Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
listlengths 1
91
| keywords
listlengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Many bacterial pathogens hijack macrophages to egress from the port of entry to the lymphatic drainage and/or bloodstream , causing dissemination of life-threatening infections . However , the underlying mechanisms are not well understood . Here , we report that Salmonella infection generates directional electric fields ( EFs ) in the follicle-associated epithelium of mouse cecum . In vitro application of an EF , mimicking the infection-generated electric field ( IGEF ) , induces directional migration of primary mouse macrophages to the anode , which is reversed to the cathode upon Salmonella infection . This infection-dependent directional switch is independent of the Salmonella pathogenicity island 1 ( SPI-1 ) type III secretion system . The switch is accompanied by a reduction of sialic acids on glycosylated surface components during phagocytosis of bacteria , which is absent in macrophages challenged by microspheres . Moreover , enzymatic cleavage of terminally exposed sialic acids reduces macrophage surface negativity and severely impairs directional migration of macrophages in response to an EF . Based on these findings , we propose that macrophages are attracted to the site of infection by a combination of chemotaxis and galvanotaxis; after phagocytosis of bacteria , surface electrical properties of the macrophage change , and galvanotaxis directs the cells away from the site of infection .
Common bacterial pathogens such as Salmonella , Shigella , and Yersinia spp . invade the gut epithelial barrier , preferentially by targeting the relatively small number of M cells located in the follicle-associated epithelium ( FAE ) [1–3] . Disruption of epithelial integrity releases chemokines that attract immune cells such as neutrophils and macrophages—a process known as chemotaxis [4–7] . Subsequent phagocytosis and clearance of these pathogens by immune cells usually stops the infection . However , some of these bacterial pathogens have developed strategies , such as the type III secretion systems in Salmonella spp . [8–12] , to evade macrophage killing and survive inside the macrophage [13–16] , an environment in which the pathogen is hidden from the immune system . Survival within the macrophage allows the pathogen to spread from its entry site to the spleen , liver , bone marrow , and other organs via the bloodstream , resulting in life-threatening consequences [17–19] . Although chemotaxis can explain how macrophages reach an infected site , it cannot explain how macrophages harboring pathogens escape from the bacterial entry site to reach the lymphatic drainage and/or bloodstream , a critical initial step in the dissemination process that is understudied and poorly understood . Bioelectric signals have been implicated in development [20–23] , wound healing [24–26] , and regeneration [27 , 28] . For example , a wound collapses the transepithelial potential ( TEP ) difference of an intact epithelial barrier , generating laterally oriented endogenous electric fields ( EFs ) of up to 1 . 4 V cm−1 , as well as local electric current densities ( JI ) of several μA cm−2 . These bioelectric phenomena actively control wound healing in the skin and cornea [25 , 29]; however , they are extremely challenging to study in the gut epithelium due to limited accessibility and have never been characterized in vivo during an active infection . Nonetheless , it is generally appreciated that an EF on this scale can drive directional cell migration—a process known as electrotaxis or galvanotaxis [21] . Many cell types , regardless of their origin , respond to an exogenous EF by directional migration toward the cathode [25 , 30] , whereas others , e . g . , human keratinocytes ( HaCat cells ) and bone marrow mesenchymal stem cells , migrate toward the anode [31] . Macrophages and lymphocytes also undergo galvanotaxis in vitro and in vivo [32–34] . In the present study , we have developed an ex vivo mouse cecum model of Salmonella infection that enables bioelectricity measurement . We report that Salmonella infection generates a directional EF at the bacterial entry sites that can recruit macrophages by galvanotaxis . We demonstrate that primary macrophages reverse galvanotaxis direction upon Salmonella infection by modifying their surface glycan composition , which reduces the negative electric charge on the surface . This directional switch is independent of the Salmonella pathogenicity island 1 ( SPI-1 ) type III secretion system , a major virulence determinant responsible for Salmonella invasion . Instead , it may require certain glycosidases that are widely conserved in Salmonella spp . , because cleavage of surface-exposed sialic acids with a potent neuraminidase caused severe defects of macrophage galvanotaxis .
The mouse is an ideal organism for understanding human infectious diseases and is widely used to study bacterial pathogenesis and mucosal immunity of enteropathogenic bacteria [35–40] . Previously , we have successfully measured bioelectric currents in various tissue and organ cultures using the noninvasive vibrating probe [41–44]; however , measuring bioelectric currents in the mouse small intestinal epithelium is challenging due to limited accessibility [45] . Therefore , we developed a new ex vivo cecum model for measuring bioelectric activity in the gut epithelium ( S1 Fig ) . This model is based on the well-established mouse typhoid model , in which mice orally infected with Salmonella enterica serotype Typhimurium ( S . Typhimurium ) develop disseminated infection that mimics human typhoid disease [36] . Although the ileum is the most commonly targeted organ to study pathogen–host interactions in vivo , it is too small for the bioelectrical measurements in our ex vivo experimental setting . Instead , the size of the mouse cecum is anatomically optimal . Salmonella invades the cecum in mice and causes acute appendicitis in humans [46] . Furthermore , we can easily identify the FAE under a dissecting microscope , because we found that 90% of C57BL/6 mice have only 1 Peyer’s patch around the blind-end apex , containing 6 to 11 lymphatic follicles . Using microelectrodes [28] in the ex vivo mouse cecum model , we measured a TEP of up to 15 mV , lumen-positive in uninfected mice . Notably , the TEP in the FAE was significantly smaller than that of the surrounding villi ( Fig 1A ) . Next , using a vibrating probe to measure the JI close to the gut surface , we detected outward currents at the FAE and inward currents at the villus epithelium , with a magnitude of around 1 μA cm−2 ( Fig 1B ) . Such currents were not detected in the serosal epithelium , despite the presence of a TEP ( Fig 1A ) , nor in formalin-fixed epithelia ( Fig 1B ) , suggesting the existence of active bioelectricity restricted to the mucosal epithelium . It has been well appreciated that mammalian intestinal mucosa maintains a large transmucosal potential difference [47 , 48] . In humans , that potential is up to 12 mV , lumen-negative , in the fasting jejunum and ileum [49 , 50] . Rats , mice , chickens , and fish all maintain a TEP of up to 5 mV in their intestinal epithelium , as measured with Ussing chambers [51–55]; however , these measurements provide no spatiotemporal information [45] . The TEP of control murine ceca ( up to 15 mV ) that we have measured directly with glass electrodes under microscopic resolution were well within the physiological range described above; however , it differed spatially , i . e . , the TEP of the FAE was significantly lower than the TEP of the surrounding villus epithelium ( Fig 1A ) . Using an ultrasensitive vibrating probe , we further detected ionic currents that run in opposite directions between the FAE and surrounding villi ( Fig 1B ) . Together , these findings unveil lateral voltage gradients and/or constant current loops running between these two structurally and functionally distinct epithelia . In the normal gut scenario , such a bioelectrical landscape may prevent commensals from accidentally entering the FAE and enable the pathogens to specifically target M cells via bacterial galvanotaxis [56] . Salmonella invades the intestinal epithelium , preferentially by targeting M cells located at an FAE [1 , 57] . If a wound disrupting an epithelium can generate a steady EF , one would envisage that similar EFs must be produced at the Salmonella entry site because of the breakage of epithelial integrity and subsequent short-circuit of the transmucosal potential difference . This is indeed the case in our model . In mice intragastrically challenged with S . Typhimurium , the peak JI remarkably reversed to become inward in FAE compared to control mice , whereas that of the villus epithelium increased nonsignificantly ( Fig 1B ) . Consistent with this finding , we also detected high variation in TEP from individual S . Typhimurium–infected mice ( up to 25 mV ) , still lumen-positive ( Fig 1A ) , which has also been observed in chemically induced colitis in rats [58] . More importantly , penetrating electrodes through disrupted FAE up to 200 μm in depth revealed a stepwise increase in TEP ( S2 Fig ) , suggesting the existence of an electric potential gradient generated by the S . Typhimurium infection . Bacterial invasion and subsequent dissemination to mesenteric lymph nodes ( MLNs ) and spleen were verified by determination of colony-forming units ( CFUs ) ( S4A Fig ) , and disruption of FAE was assessed by histological staining ( S4B Fig ) . S . Typhimurium was undetectable in some of the MLNs and spleens , indicating that some of the FAE and surrounding villi were either uninfected or mildly infected , which explains the wide distribution of TEP and JI measured in the S . Typhimurium–challenged mice . Based on these results , we coined the term infection-generated electric field ( IGEF ) ( Fig 1C ) to distinguish it from a wound-generated electric field ( WGEF ) [26] . We demonstrate , for the first time , that Salmonella infection generated a steady EF ( up to 5 V cm−1 , provided an epithelial thickness of 50 μm ) that drives minute directional electric currents , running from the breached FAE into the deep intestinal wall in a stepwise manner . Although how the IGEFs are formed is currently unknown , we speculate on a couple of possible mechanisms . First , in order to establish a steady EF , a potential gradient or a circuit must be formed . Even though there are accumulated charges segregated by the epithelium ( positive at the apical side ) , in a healthy gut a circuit is unlikely to be formed by the epithelium itself due to high resistance of the epithelium , which is sealed by tight junctions . However , given that the intestinal lumen is alkaline ( more negative in terms of electrogenesis ) , a micropotential gradient could be developed in close proximity to the apical side of the gut epithelia , which can drive outward ionic currents ( Fig 1C ) . Second , differential expression and asymmetric distribution of ion pumps and channels essential for selective absorption and/or secretion of electrogenic ions by enterocytes and M cells [47 , 48] are likely to be critical for the generation of the aforementioned bioelectricity . Future studies using specific channel blockers in combination with our bioelectrical experimental model and techniques will help to pinpoint the molecular mechanism of an IGEF by identifying the channel ( s ) or pump ( s ) involved . Third , as in the formation of a WGEF [21 , 59] , Salmonella preferentially invades and destroys M cells and collapses the epithelial barrier at the FAE ( S4B Fig ) , which short-circuits the TEP . Subsequently , the short-circuited and augmented TEP could then drive inward ionic currents as supported by the measurements of stepwise increase of the TEP ( S2B Fig ) . Fourth , in contrast to the healthy alkaline gut , the microenvironment at the Salmonella entry site ( i . e . , the FAE ) is likely to be more acidic ( more positive in terms of electrogenesis ) because of the local inflammatory responses [60] and metabolic changes [61] induced by Salmonella infection . Such a microenvironmental pH switch could be related or attributed to the reversal of ionic current flow as we detected with vibrating probes ( Fig 1B ) . Like the injury currents reported by Sawyer and colleagues in early publications [62–64] , IGEF-driving ionic currents could affect small blood vessels in the intestinal wall and mesentery to cause a transvascular potential drop or reversal , resulting in 2 possible consequences: an intravascular occlusion that may benefit transendothelial penetration of immune cells ( e . g . , monocytes , leukocytes ) and/or the creation of a galvanotactic route between the infected epithelium and the electrically impacted vessels [65] . Although IGEFs may provide a guidance cue for the enterocytes or even local stem cells , contributing to the repair process of damaged epithelium , the major focus of this work is , rather , to investigate its biopathological role in the systemic Salmonella infection , specifically during the initiation of macrophage-driven dissemination . Previous studies have shown that applying an EF in vitro can direct macrophage galvanotaxis to the anode [32–34] . We confirmed this phenotype by demonstrating that the primary mouse peritoneal macrophages ( PMs ) , in response to an exogenous EF tuned to mimic the in vivo IGEF ( mathematics in Materials and methods ) , underwent robust unidirectional migration to the anode ( S1 Movie ) . This unidirectional migration was verified by immunostaining showing that nearly 100% of macrophages were polarized to the anode with a distinct morphology characterized by a leading pseudopodium of dense actin meshwork and a rearward uropod . However , upon challenge with S . Typhimurium IR715 , the average directionality of EF-induced galvanotaxis decreased significantly , with approximately 41% of the macrophages reversing their migratory direction from the anode to the cathode ( Fig 2 and S2 Movie ) . Although PMs have been widely used in bacterial infection studies [66] , likely due to the ease of harvesting , the most commonly used primary murine macrophages are the bone marrow-derived macrophages ( BMDMs ) because of their high yield and less heterogeneous nature , which makes them phenotypically and functionally different from PMs [67] . To exclude the possibility that the observed bidirectional migration is restricted to PMs , we generated and tested galvanotaxis of BMDMs as we did with PMs . Although both types of macrophages exhibited similar unidirectional migration ( mean directionality: −0 . 98 versus −0 . 88 ) to the anode , reversal in BMDMs infected by Salmonella IR715 was more robust than in PMs infected by the same Salmonella strain ( mean directionality: 0 . 36 versus −0 . 18 ) ( Fig 2C and S5B Fig ) . Notably , this phenotype can be reproduced in BMDMs challenged by another 2 virulent S . Typhimurium strains , LT2 and SL1344 ( S5 Fig ) , suggesting that directional migration in response to electrical stimuli is an intrinsic hallmark of the macrophages regardless of their origins and that the ability to manipulate and subvert galvanotaxis in these macrophages is conserved in virulent Salmonella spp . If the cells had simply stopped sensing the EF , then we would have observed random migration , as was the case for control macrophages not subjected to an EF ( Fig 2E ) . The observed change in migration pattern can be attributed to S . Typhimurium infection for a variety of reasons . First , a gentamycin protection assay confirmed the presence of intracellular bacteria ( Fig 2D ) ; second , flow cytometry demonstrated a high S . Typhimurium infection rate ( approximately 53% ) ( S6 Fig ) ; and third , high-resolution confocal microscopy revealed that macrophages containing intracellular S . Typhimurium switched polarity to the cathode ( Fig 2G ) . Based on these findings , we conclude that Salmonella-containing macrophages respond to galvanotaxis stimuli by reversing their primary directional migration . Macrophages are professional phagocytes that uptake a broad range of substances; meanwhile , pathogenic Salmonella has developed several virulence means to evade macrophage killing , with SPI-1 being the major virulence factor responsible for colonization and invasion [10 , 68] . To better understand whether the directional reversal is phagocytosis dependent or SPI-1 specific , we monitored galvanotaxis under identical conditions in BMDMs challenged with ( i ) fluorescently labeled microspheres similar in size to bacteria , ( ii ) S . Typhimurium constitutively expressing mCherry , and ( iii ) a green fluorescent protein ( GFP ) -expressing ΔinvA mutant that lacks a functional SPI-1 ( unable to inject its effectors into cells ) [69] ( Fig 3A ) . Macrophages challenged with microspheres exhibited migratory behavior similar to that of controls , i . e . , unidirectional migration to the anode ( Fig 3B and S3 Movie ) . Macrophages challenged with ΔinvA migrated with a significantly decreased overall directionality close to that of macrophages infected with wild-type ( WT ) Salmonella ( Fig 3B ) , even though the number of intracellular mutants was indeed lower than that of the WT ( Fig 3C ) . Time-lapse recording captured a marked opposing directional migration pattern: macrophages containing microspheres migrated to the anode , and macrophages with either WT or ΔinvA bacteria inside migrated to the cathode ( Fig 3D and 3E , and S3 Movie ) , further confirming that the observed directional switch was Salmonella infection dependent . There were similar phagocytosis and/or infection rates between the macrophages challenged with microspheres or Salmonella in the given multiplicity of infection ( MOI; S7A and S7B Fig ) , as verified by flow cytometry ( S6 Fig ) . Cells containing no or variable microspheres migrated to the anode exclusively ( S7C and S7D Fig ) , ruling out mechanisms solely based on phagocytosis . The fact that cells containing either WT or ΔinvA migrated to the cathode suggests that the SPI-1 type III secretion system is not required or , at least , is insufficient for the directional switch . These data are in accordance with previous studies that identified an SPI-1-independent pathway contributing to early dissemination of S . Typhimurium in the mouse typhoid model [17] . We therefore argue for the existence of a general infection-dependent mechanism that involves phagocytosis and subsequent interplay between the macrophages and internalized bacterial pathogens . Charged cell-surface components are critical for EF-induced motility in 3H3 cells [70] and have been implicated in electro-osmosis of concanavalin A ( Con A ) binding receptors on the surface of myotomal spheres [71] . We hypothesized that the directional switch of macrophage galvanotaxis could result from bacteria-induced modifications to the charged components of the cell surface , which would not occur following microsphere challenge ( Fig 4A ) . To this end , we screened Salmonella-infected and control macrophages against a panel of fluorescently labeled lectins ( glycan-binding proteins ) capable of detecting charged and uncharged cell surface glycans ( S1 Table ) . The normalized mean fluorescent intensity of Maackia Amurensis Lectin II ( MAL-2 ) , a lectin that recognizes pathogen-binding sialic acids , was significantly decreased in macrophages infected by Salmonella but not in those carrying microspheres ( Fig 4B–4D ) . Marked Galanthus Nivalis Lectin ( GNL ) - and Con A–binding aggregates were visible within macrophages infected by Salmonella ( S8 and S9 Figs ) , raising the possibility that the decrease in MAL-2-binding sialic acids could be the result of bacterial internalization and subsequent degradation . Sialylated cell surface molecules are negatively charged , creating an electronegative zeta potential [72] . Using an electrophoretic light-scattering technique , we determined the zeta potential of BMDMs under various conditions . The negative zeta potential of macrophages infected by S . Typhimurium was significantly reduced , i . e . , less negative than that of control macrophages ( P < 0 . 05 ) ( Fig 4E ) . By contrast , macrophages challenged with microspheres showed a nonsignificant zeta potential change compared to that of control macrophages ( Fig 4E ) , suggesting that the decrease in surface-exposed sialic acids and subsequent reduction of surface negativity is mediated by active bacterial product ( s ) [73] . If a decrease of the negatively charged sialic acids of the surface glycoproteins is critical for the directional switch in macrophage galvanotaxis , then cells with their sialic acids enzymatically removed should exhibit a switch or at least a defect in directional migration when exposed to an EF . To test this , we treated freshly isolated mouse BMDMs with a potent neuraminidase ( an enzyme that cleaves terminal sialic acid residues from surface-exposed glycoproteins ) [74] . Cleavage of sialic acids following enzymatic treatment was confirmed by flow cytometry and confocal microscopy ( Fig 5A–5C ) . As predicted , the zeta potentials of neuraminidase-treated macrophages were significantly reduced ( Fig 5D ) . These cells also lost anodal migration compared to control macrophages monitored in parallel ( Fig 5E and 5F , and S4 Movie ) . Further inspection and morphological quantification of macrophages stained with fluorescently labeled actin and/or lectin revealed that 71% of the macrophages treated with neuraminidase failed to establish a polarity and notably , 12% of the cells were polarized to the cathode whereas nearly all the control macrophages ( 97% ) were polarized to the anode in response to the EF ( Fig 5G ) . It is well known that a change in the environmental pH can dramatically influence the growth and virulence of Salmonella [75 , 76] . Moreover , by exposure to acidic pH , a condition that is required for activation of several Salmonella virulence factors [77–79] , it is possible to modify macrophage surface electrical properties [30] . Therefore , to further determine the importance of surface negativity , we incubated BMDMs in medium at pH 5 . 8 that markedly reduced the zeta potential ( S11A Fig ) , presumably through protonation of the sialylated surface molecules . Similar to the neuraminidase treatment , galvanotaxis of BMDMs under acidic conditions was significantly impaired , resulting in nearly half of the macrophages ( 47% ) losing their directional migration and 14% of the macrophages reversing their polarity to the cathode ( S11B-S11D Fig and S5 Movie ) . Although these data are consistent with our previous studies , showing that low pH abrogates directional migration of epithelial cells in response to an EF [80 , 81] , it also suggests that the acidic environment is not only required for the activation of Salmonella but also contributes to triggering dissemination . How does Salmonella infection instruct macrophages to reverse EF-directed migration ? Firstly , it is important to note that infecting macrophages with a SPI-1 mutant resulted in a directional switch similar to that of WT , with nearly all the macrophages containing live fluorescent protein-expressing bacteria migrating to the cathode ( Fig 3 and S3 Movie ) . This suggests a general mechanism , independent of this major virulence factor , even though other specific factor ( s ) may still be involved [82 , 83] . Previously reported effects have suggested that negatively charged surface glycan moieties are critical for EF-induced cell motility and polarization , which are consistent with our data , and provide a long sought-after mechanism of action . Because macrophages challenged with microspheres did not show a significantly reduced zeta potential ( Fig 4E ) , it is likely that the decrease in surface-exposed sialic acids and reduction of surface negativity is mediated by active bacterial product ( s ) rather than by metabolic changes in the host itself . It is also important to note that although both Salmonella infection and neuraminidase treatment decreased the surface-exposed sialic acids , the latter caused a serious defect in macrophage galvanotaxis without reversing the overall directionality ( Fig 5E ) , in contrast to Salmonella infection ( Fig 3B ) . There may be multiple glycosidases involved in surface glycan modification to reverse directionality ( Fig 5H ) because Salmonella possesses at least 51 putative glycosidases that likely function in glycan degradation [84] . In fact , a recent study identified several glycosidases , including a putative neuraminidase , as new virulence factors essential for Salmonella infection of epithelial cells , which is again independent of the SPI-1 [85] . We are in the process of performing genetic knockouts to identify the factors involved . It is also possible that modification of surface glycan and reduction of zeta potential were mediated by internalization during phagocytosis of the bacterium itself rather than by bacterial enzymatic activities . For example , macrophages express Toll-like receptors ( TLRs ) that recognize structurally conserved molecules derived from Salmonella and other pathogens . All TLRs contain N-linked glycosylation consensus sites , and both TLR2 and TLR4 require glycosylation for surface translocation and function [86 , 87] . Binding of Salmonella to these glycosylated receptors and subsequent internalization may reduce surface negativity of macrophages , leading to directional switch under an EF . This idea is supported by our observations in which the accumulation of certain lectin binding aggregates within macrophages infected by Salmonella but not in cells challenged by microspheres ( S8 and S9 Figs ) . It is also possible that a macrophage can uptake microspheres without significantly changing its zeta potential ( e . g . , through a neutralized receptor ) and thus still migrate to the anode . Disseminated Salmonella infection is a major health problem of developing countries , responsible for approximately 433 , 000 deaths annually [88] . Understanding the mechanisms that trigger dissemination is critical for efforts to target this key process for preventive and therapeutic purposes . We propose that macrophages are attracted to the site of infection by a combination of chemotaxis and galvanotaxis , driving the cells in the same direction . After phagocytosis of bacteria , surface electrical properties of the macrophage change , and galvanotaxis directs the cells away from the site of infection ( Fig 6 ) . Our study represents a new perspective for the initiating mechanisms , suggesting that Salmonella disseminates through infection-generated bioelectrical control of macrophage trafficking . It is important to emphasize that the demonstrated bidirectional migration of macrophages to the IGEF-like EFs is not a physical electrophoresis ( movement of charged particles under direct-current EF ) but rather a complex yet poorly understood biological process that requires phosphoinositide 3-kinases and other critical signaling activities , as well as the cellular motility machinery [25 , 81] . It is also worth emphasizing that the model proposed in this work is not mutually exclusive with respect to chemotaxis but offers an alternative and/or complementary mechanism of directional migration . Both directional cues can coexist and play equally important roles in orchestrating the initial stage of the innate immune response against bacterial infection ( Fig 6 ) , although a chemical gradient could be overridden by a strong electrical stimulus [59 , 89] . Both chemotaxis and galvanotaxis likely share critical signaling pathways as suggested in studies of macrophage-like Dictyostelium , which are highly sensitive to cyclic adenosine monophosphate gradients , as well as to electrical potential gradients [90–92] . Future work utilizing transgenic animals and pharmacological perturbations to target specific pathways ( known or unknown ) will help to pinpoint key molecules mediating the infection-dependent directional reversal ( i . e . , the molecular mechanisms to initiate disseminated infection ) and response to bioelectric signaling . Triggering bacterial dissemination mediated by macrophage galvanotaxis might be a common strategy , not only for pathogenic Salmonella but also for other bacterial pathogens that are able to invade macrophages and survive intracellularly . Although the present work deals primarily with gut epithelium and enteric bacteria , the general mechanism that emerged from this work could also apply to other mucous epithelia such as the respiratory tract and its associated pathogens , e . g . , Mycobacterium tuberculosis ( the causative agent of tuberculosis ) [93] , which is another major public health concern .
The mouse strains used were in a C57BL/6 background ( both male and female mice were used in experiments ) . Mice were purchased from Jackson lab and maintained under a strict 12-h light cycle and given a regular chow diet in a specific pathogen-free facility at University of California ( UC ) , Davis . All animal experiments were performed in accordance with regulatory guidelines and standards set by the Institutional Animal Care and Use Committee of UC Davis . In brief , we dissected mouse cecum following euthanasia and opened longitudinally along the mesenteric attachment remnant to avoid incision damage to the single Peyer’s patch located under the antimesenteric mucosa near the apex ( S1A Fig ) . After thorough washing in mouse Ringer’s solution ( 154 mM NaCl , 5 . 6 mM KCl , 1 mM MgCl2 , 2 . 2 mM CaCl2 , 10 mM glucose , and 20 mM HEPES [pH 7 . 4] ) to remove the luminal contents , we placed the cecum with mucous side facing up , on a 30° slope of silicone gel prepared from polydimethylsiloxane ( PDMS ) in custom-made measuring chambers . The cecum was aligned and immobilized with fine metal ( tungsten ) pins prior to taking measurements ( S1B Fig ) . This process was usually completed within 5 min . We used glass microelectrodes to directly measure the TEP of intestinal epithelium as previously described [28 , 94] . TEPs were recorded by microelectrode impalement through the epithelial layers . Microelectrodes ( 1–2 μm tip diameter; NaCl 3 M electrolyte ) had resistances of approximately 1 to 2 MΩ , and the potentials were offset to 0 mV prior to impalement . Cecal FAE and adjacent villus epithelium were discriminated under a dissecting microscope within a Faraday cage on an antivibration table . In some cases , the TEP were measured as follows: first at the epithelial surface ( 0 μm ) , then stepwise at 50 , 100 , and 200 μm in depth , controlled by a micromanipulator ( S2A Fig ) . The potential typically returns to the baseline of 0 mV after microelectrode withdrawal . If the reference baseline was > ±1 and ≤ ±5 mV , the value was subtracted from the TEP recorded as shown in the equation ( S2B Fig ) ; if > ± 5 mV , the trace was rejected . As a control , we measured the TEP of serosa epithelium and the TEP of formalin-fixed mucous epithelium . Measurements were performed at room temperature in mouse Ringer's solution . Data were acquired ( saturated sampling at 100 Hz ) and extracted using pClamp 10 ( Molecular Devices ) and analyzed using Excel ( Microsoft , Redmond , WA ) . We used noninvasive vibrating probes to measure the electric current density ( JI ) in μA cm−2 of mouse cecum epithelium as previously described [28 , 41 , 42] . The probes , platinum-electroplated at the tip ( approximately 30 μm ball diameter ) , vibrated at a frequency between 100 and 200 Hz . Prior to measurements , the probe was calibrated to the experimental conditions by an applied JI of 1 . 5 μA cm−2 ( S3C Fig ) . Under a dissecting microscope , mounted mouse ceca were positioned in the nonconductive measuring chamber ( S3A Fig ) . The plane of probe vibration was perpendicular to the epithelial surface at a distance as close as possible ( S3B Fig ) . JI was recorded until the plateau peak was reached ( <1 min ) ( S3C Fig ) . Reference values were recorded with the probe away from the epithelium surface ( ≫1 mm ) ( S3A Fig ) . Measurements were taken at room temperature in mouse Ringer's solution . During calibrations and measurements , a Faraday “wall” ( grounded aluminum-wrapped cardboard ) covered the microscope . As a control , we measured JI near the surface of serosal epithelium and formalin-fixed mucous epithelium . Data were acquired and extracted using WinWCP V4 ( Strathclyde Electrophysiology Software , University of Strathclyde , Glasgow , United Kindom ) and analyzed using Excel ( Microsoft , Redmond , WA ) . Special reagents used in this work are listed in S2 Table . Plasmids and Salmonella strains used in this work are listed in S3 Table . Cultures of Escherichia coli ( for plasmid extraction ) and S . Typhimurium were incubated aerobically at 37°C in Luria-Bertani ( LB ) broth ( per liter: 10 g tryptone , 5 g yeast extract , 10 g NaCl ) or on LB agar plates ( 1 . 5% Difco agar ) overnight . Antibiotics were used at the following concentrations unless stated otherwise: 30 μg ml−1 chloramphenicol , 50 μg ml−1 nalidixic acid , 100 μg ml−1 ampicillin , 50 μg ml−1 kanamycin , and 10 μg ml−1 tetracycline . In an effort to monitor live intracellular Salmonella during macrophage galvanotaxis we constructed an S . Typhimurium strain derived from IR715 that constitutively expresses mCherry coded in its genome [95] . In brief , the glmS::mCherry allele was transferred to IR715 via P22 transduction from the donor strain S . Typhimurium SL1344St ( a gift from Leigh Knodler ) [96] and selected by chloramphenicol . To remove the FRT-flanked chloramphenicol cassette , the transductant was transformed with pCP20 encoding FLP recombinase . The resulting strain constitutively expressing mCherry encoded by glmS::mCherry::FRT was designated KLL18 . The GFP-expressing SPI-1 mutant ΔinvA was generated by electroporating pGFT/RalFc into AJB75 [69] . The plasmid pGFT/RalFc was constructed in 2 steps . First , a fragment of a gfp-mut3 gene under the control of the constitutively active kanamycin-resistance gene aphA3 promoter was amplified from pJC43 [97] with the primers GFPmut3-F ( 5ʹ- AGAGCTCCAGCGAACCATTTAAGGTGATAG -3ʹ ) and GFPmut3-R ( 5ʹ- ACTGCAGTTATTTGTATAGTTCATCCATGCC -3ʹ ) . This fragment was then digested with SacI/PstI and cloned into pFT/RalFc , a low–copy-number plasmid based on pBBR1-MCS4 [98] . The mouse is a well-established animal model for studying Salmonella pathogenesis [35] . C57BL/6 and other mice carrying a mutation in nramp1 develop disseminated infections when challenged by S . Typhimurium , which mimics human typhoid fever [99] . For the mouse infection experiment , we used the hyper-disseminative D23580 strain [100 , 101] . In brief , D23580 was used to inoculate LB broth and incubated overnight at 37°C . C57BL/6 mice ( 6 to 10 wk old , mixed sexes ) were intragastrically infected with 109 bacteria ( actual inoculum was determined by plating ) in 0 . 1 ml LB broth . Uninfected mice used as a control were given 0 . 1 ml sterile LB broth in place of Salmonella . Mice were euthanized at 16 h and 40 h post infection ( PI ) by CO2 asphyxiation followed by cervical dislocation as the secondary method of euthanasia . The 16 h PI was chosen based on previous studies showing that the best-characterized route for the phagocytes harboring Salmonella to reach the bloodstream normally requires between 12 and 20 h [17 , 102] . MLNs and spleens were collected aseptically and homogenized in phosphate-buffered saline ( PBS ) for CFU enumeration . Ceca were dissected , cleaned , and then either mounted for bioelectrical measurement , prepared for histopathological fixation , or homogenized in PBS for CFU enumeration . Ceca were fixed in 10% neutral buffered formalin . After fixation , tissues were routinely processed , embedded in paraffin , sectioned , and stained with hematoxylin–eosin . Both PMs and BMDMs were isolated from C57BL/6 mice ( 6 to 10 wk old , mixed sex ) following standard procedures as previously described [103] . PMs were seeded onto 6-well plates and allowed to adhere to the plastic for 1 to 2 d in DMEM ( Invitrogen ) with 10% fetal bovine serum ( Invitrogen ) and 1× antibiotic-antimycotic solution ( Invitrogen ) . BMDMs were cultured in the same medium as described above but supplemented with 20% L-929 conditioned medium for 6 d ( plus an extra feed at day 3 ) , followed by 1-d culture without the conditioned medium . Adherent macrophages were then harvested by gently scraping with a “policeman” cell scraper and used for subsequent experiments accordingly . Cell viability was determined by trypan blue staining and counting . Our initial galvanotaxis experiments were carried out in PMs . Because we observed better directional switch in BMDMs infected by Salmonella ( S5 Fig ) , subsequent experiments were done in BMDMs , unless stated otherwise . The gentamycin protection assays were carried out as previously described [104] . In 24-well tissue culture-treated plates , 2 × 105 cells were seeded per well for 5 to 6 h in culture medium ( DMEM with 10% fetal bovine serum and no antibiotics ) . Salmonella were grown overnight and used to infect macrophages at an MOI of 20 . After 60 min of incubation , cells were gently washed 3× with PBS and further incubated in gentamycin-containing culture media at a final concentration of 50 μg ml−1 for an additional 60 min . Afterwards , media were replaced with culture media containing 10 μg ml−1 gentamycin for the duration of the experiment . Intracellular CFU was measured at 16 h PI . To measure intracellular CFU , macrophages were lysed using 0 . 5% Tween 20 for 5 min at room temperature and released by scraping with 1 ml pipette tips . CFUs were enumerated by plating . Typically , 4 × 104 primary mouse macrophages were seeded per well of engineered silicon stencils sealed in custom-made EF chambers ( see “Engineering silicone stencil and EF chamber design” for details ) or 96-well glass bottom plates ( Nunc ) or 2 × 105 cells per well in 24-well tissue culture-treated plates depending on different experiment needs for 5 to 6 h in culture medium . Overnight cultures of Salmonella or fluorescently labeled microspheres were used to infect and/or challenge macrophages at an MOI of 20 . The rest of the procedures were similar to the gentamycin protection assay . Cells were cultured in medium containing 10 μg ml−1 gentamycin for 16 h , and subsequent galvanotaxis experiments were carried out in the same medium containing gentamycin . For the neuraminidase treatment , cells were incubated in culture medium containing 100 mU ml−1 neuraminidase from Vibrio cholerae ( Sigma-Aldrich ) for 30 min at 37°C [70] . Cells were then washed with culture medium , and subsequent galvanotaxis experiments were carried out in the culture medium containing no neuraminidase . For the low pH experiments , cells were incubated in culture medium of pH 5 . 8 buffered with 15 mM MES ( Sigma-Aldrich ) for 60 min , and subsequent galvanotaxis experiments were carried out in the same media of pH 5 . 8 [81] . Control experiments were carried out always in parallel in culture medium of pH 7 . 4 buffered with 14 . 4 mM HEPES ( Invitrogen ) . Macrophages were seeded in either 96-well glass bottom plates or custom-made EF chambers and infected , challenged , and/or treated by following procedures as described above . The cells were fixed with 4% paraformaldehyde immediately or after EF exposure for 3 h with field orientation marked . Salmonella were detected with a polyclonal antibody specific to Salmonella spp . ( Mybiosource , San Diego , CA ) stained by an Alexa Fluor 488-conjugated secondary antibody . F-actin was labeled by Alexa Fluor 555 Phalloidin . Nuclei were labeled by Hoechst 33342 . In the cases of lectin staining , fixed cells were incubated with FITC-labeled lectin ( S1 Table ) overnight at 4°C , washed extensively , and then stained with DAPI for 10 min on ice . Cells were photographed using either an inverted ( for cells on cover glass with no EF ) or an upright ( for cells on plastic EF chambers ) Leica TCS SP8 confocal microscope ( Leica microsystem ) . Images were processed using ImageJ . Quantification and comparison of fluorescent intensity were done in images taken in the same batch with the same optical setup and parameters . Lectin binding aggregates stained after permeabilization were quantified by thresholding . Cells were counted using particle analysis function . Infected , challenged , and/or treated macrophages , handled according to the procedures described above , were then incubated with Fc-block ( BD , Franklin Lakes , NJ ) on ice for 15 min , stained with FITC-labeled lectin ( S1 Table ) for 1 h on ice , and then stained with Aqua-LIVE/DEAD ( Invitrogen , Carlsbad , CA ) for 30 min at room temperature . Cells were washed after each step and before being analyzed on a BD Fortessa flow cytometer . Data were analyzed using FlowJo software ( Tree Star Inc . Ashland , OR ) . After gating single cells and live cells , the geometric mean fluorescence intensity and standard error ( SE ) were collected for each FITC lectin in each condition in addition to Fluorescence Minus One ( FMO ) for FITC-lectin ( no FITC-lectin staining ) . The geometric mean fluorescence intensity was then standardized across experiments using the following equation: xFMO ( MFI ) =x ( MFI ) −FMO ( MFI ) FMO ( SD ) , ( 2 ) where xFMO ( MFI ) is the standardized geometric mean fluorescence intensity of a specific lectin for a specific experiment , x ( MFI ) is the geometric mean fluorescence intensity of a specific lectin for a specific experiment , FMO ( MFI ) is the geometric mean fluorescence intensity of the FMO for a specific experiment , and FMO ( SD ) is the standard deviation of the FMO for a specific experiment . Standardized geometric mean fluorescent intensities were then plotted and tested for statistical significance ( S10 Fig ) . Macrophages were seeded onto 24-well tissue culture plates and infected , challenged , and/or treated following procedures as described above . Cells were fixed in 2% paraformaldehyde and washed with motility buffer ( 10−4 M potassium phosphate buffer at pH 7 . 0 , with 10−4 M EDTA ) [56] . Cells were then gently collected by scraping with a “policeman” cell scraper , and subsequent measurements were done in motility buffer , except for the macrophages tested for acidic treatment , which were measured in either pH 5 . 8 medium or pH 7 . 4 medium as a control . Zeta potential was determined by electrophoretic light scattering at 25°C with a Zetasizer ( Malvern Panalytical Ltd , Malvern , United Kingdom ) . Zeta potential was calculated in mV , and differences between groups were analyzed by Student t test . Galvanotaxis data from representatives of at least 4 independent experiments were routinely presented as mean ± SE , unless stated otherwise . Distributions of macrophage polarity between control and neuraminidase treated or between neutral and acidic conditions were analyzed using χ2 test . Student t test and one-way ANOVA analysis followed by post hoc Tukey HSD test were used for paired or unpaired comparisons among 2 groups or multiple groups ( more than 2 ) , respectively .
|
Bacterial pathogens can invade and survive within macrophages and use them as a vehicle to reach important organs of a human body , resulting in life-threatening infections , but the underlying mechanisms are not well understood . Our current understanding is that macrophages are recruited to the infected site by sensing gradients of chemokines and/or cytokines released by damaged cells—a process known as chemotaxis . However , this mechanism does not explain how macrophages containing the pathogens escape the site to reach the bloodstream . Here , we use a disseminated Salmonella infection model in mice and detect electric fields ( EFs ) generated by Salmonella infection at the gut epithelium , which can drive unidirectional migration of macrophages towards the anode—a biological process also known as galvanotaxis . We further demonstrate that macrophage galvanotaxis can be reversed to the cathode by phagocytosis of the bacteria . Based on these findings , we propose that macrophages are attracted to the site of infection by a combination of chemotaxis and galvanotaxis; after phagocytosis of bacteria , the electrical properties of the macrophage change , and galvanotaxis directs the cells away from the site of infection allowing the escape of the macrophages that contain pathogens .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"engineering",
"and",
"technology",
"electronics",
"cathodes",
"pathogens",
"immunology",
"electricity",
"microbiology",
"salmonellosis",
"bacterial",
"diseases",
"enterobacteriaceae",
"electric",
"field",
"bacteria",
"bacterial",
"pathogens",
"salmonella",
"typhimurium",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"biological",
"tissue",
"salmonella",
"physics",
"anodes",
"cell",
"biology",
"anatomy",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"macrophages",
"electrodes",
"organisms"
] |
2019
|
Infection-generated electric field in gut epithelium drives bidirectional migration of macrophages
|
To regulate shape changes , motility and chemotaxis in eukaryotic cells , signal transduction pathways channel extracellular stimuli to the reorganization of the actin cytoskeleton . The complexity of such networks makes it difficult to understand the roles of individual components , let alone their interactions and multiple feedbacks within a given layer and between layers of signalling . Even more challenging is the question of if and how the shape of the cell affects and is affected by this internal spatiotemporal reorganization . Here we build on our previous 2D cell motility model where signalling from the Rho family GTPases ( Cdc42 , Rac , and Rho ) was shown to organize the cell polarization , actin reorganization , shape change , and motility in simple gradients . We extend this work in two ways: First , we investigate the effects of the feedback between the phosphoinositides ( PIs ) , and Rho family GTPases . We show how that feedback increases heights and breadths of zones of Cdc42 activity , facilitating global communication between competing cell “fronts” . This hastens the commitment to a single lamellipodium initiated in response to multiple , complex , or rapidly changing stimuli . Second , we show how cell shape feeds back on internal distribution of GTPases . Constraints on chemical isocline curvature imposed by boundary conditions results in the fact that dynamic cell shape leads to faster biochemical redistribution when the cell is repolarized . Cells with frozen cytoskeleton , and static shapes , consequently respond more slowly to reorienting stimuli than cells with dynamic shape changes , the degree of the shape-induced effects being proportional to the extent of cell deformation . We explain these concepts in the context of several in silico experiments using our 2D computational cell model .
In order to address these issues , we have developed a computational model that integrates the signalling biochemistry with actin-based motility in a spatial setting . Our main philosophy in constructing the model has been to assemble modules of the signalling repertoire for which there is biological consensus or strong experimental evidence , to identify model parameters based on quantitative biological information , and to study the dynamics of these modules individually [34] , [35] , with dynamic actin cytoskeleton [25] , [36] and in concert with other signalling modules [25] in 2D spatiotemporal computations . Details of the assumptions , steps , parameter choices and strategy have been extensively reviewed elsewhere [25] , [34] , [36] and are abbreviated in the Materials and Methods . In view of our main aim to understand the role of feedback from PIs to GTPases , we here revised the model in Dawes and Edelstein-Keshet [25] so as to “tune” the magnitude of feedback from the PIs to the Rho GTPases over a full range ( from absent to essential ) and compare the resulting behaviours . ( See dashed lines in Fig . 1 ) . This in silico tuning represents our depiction of in vitro or in vivo knockout , silencing , and overexpression experiments . To meet our second aim of elucidating how cell shape influences biochemical repolarization , we implement a fully 2D computation with evolving cell geometry . The biochemistry is summarized schematically in Fig . 1 . The model consists of a set of coupled partial differential equations ( PDEs ) that describe the kinetics , crosstalk , diffusion , and exchange of the following intermediates: Cdc42 , Rac , and Rho ( active and inactive forms – 6 PDEs ) , PIP , , ( lipids diffusing in the membrane – 3 PDEs ) and Arp2/3 ( active cytosolic form , 1 PDE ) . ( See equations in the Materials and Methods and parameter values based on biological data in Table 1 ) . Initially , all concentrations are uniform in the interior of a circular domain , representing an unstimulated resting cell . Stimulation is depicted by imposing a transient , spatially dependent activation of Cdc42 on a cell that is initially at rest , with no other spatial bias in any internal component . To keep the size of the model modest , we do not explicitly model the Rho GEFs and GAPs nor the kinases or phosphatases ( PI5K , PI3K , PTEN ) as dynamic variables . Feedback from GTPases to kinases and phosphatases are included in concentration-dependent rate-constants . In order to appraise the effect of PI feedback onto the small GTPases , we introduced a parameter , for the efficacy of such feedback ( where means no feedback from PIs and means that PIs are essential for activation of the small GTPases ( see Eqn . ( 4 ) ) . As described in [36] , the discretized densities and orientations of actin filaments and barbed ends are represented as evolving spatial distributions , where Arp2/3-dependent branching enhanced by Cdc42 nucleates new barbed ends . We keep track of a special category of barbed ends engaged with and applying force to the membrane , the “pushing barbed ends” . These promote local outwards protrusion as in the thermal polymerization ratchet mechanism [37]–[39] . Areas of high Rho are interpreted as sites where actomyosin contraction would be enhanced . This is depicted by a force directed inwards and perpendicular to the cell membrane . ( Such zones tend to spontaneously become the “back” of the evolving cell . ) To combine reaction-diffusion ( RD ) kinetics with fully dynamic cell shape so as to show the important feedbacks between the geometry and the biochemistry , we use the Cellular Potts Model ( CPM ) framework [36] , [40] , [41] . In this multi-scale approach , the CPM specifies the domain and boundary conditions for the RD equations ( PDEs ) at each time point . The PDEs are solved efficiently “on the fly” in the irregular domain , generating the intracellular patterns that lead to differential forces on the cell membrane . The shape is then updated by an energy-minimization ( Hamiltonian based ) stochastic edge update algorithm [40] , [41] . ( See Materials and Methods . )
To assess whether the model can capture basic experimental observations we ran the full model ( Eqs . ( 1 ) – ( 16 ) ) with biologically-based parameter values ( Table 1 ) . In the absence of stimuli , the resting cell is stable , and does not deform significantly nor move . We imposed a transient ( 10 s ) gradient in the Cdc42 activation rate on the resting cell , and followed the dynamics of the GTPases ( Eqs . 1 , 2 ) , the PIs ( Eqs . 5 ) , Arp2/3 ( Eqn . 11 ) , and actin ( Eqs . 7–8 ) for 90 simulated minutes . Other than the graded stimulus , we do not a priori define a “front” or a “back” in the cell; all other dynamics develop spontaneously . The final cell shape and spatial distributions of all variables are shown in Fig . 2 . Profiles of the GTPases and PIs ( rows 1–3 ) are shown both as 2D heat maps ( left ) and as line plots ( right ) . As observed experimentally , active Cdc42 and Rac , as well as and , are enriched at one end , whereas active Rho and PIP are most prevalent at the opposite end . forms the steepest gradient , followed by . Due to their very rapid rates of diffusion in the cytosol , the inactive GTPases distribute more or less uniformly over the cell ( Fig . 2 , Row 2 ) , even when their active forms are spatially segregated . The transition between resting and motile cell is indicated in several panels in Fig . 2 . In the rest state , the cell is disk shaped , with radially symmetric filaments and barbed end densities . Stochastic noise leads to a fluctuating edge and small displacements of the centre of mass , but the cell as a whole does not move ( red curve , bottom left panel , Fig . 2 ) . Once stimulated , the cell rapidly takes on a roughly oval shape and attains a velocity of . This speed remains constant and is maintained after the transient stimulus is removed , unless other stimuli are introduced ( See Video S1 , and black curve , bottom left panel , Fig . 2 ) . Elevated Rac and Cdc42 enhance Arp2/3 activation and branching of actin filaments . ( Eqs . 7–8 and Fig . 2 , Row 4 . ) In the case of Rac , this takes place through the activation of PI5K ( ) , which elevates and in turn induces Arp2/3 activation . Cdc42 further accelerates the -induced Arp2/3 activation , which promotes a local increase in barbed ends ( Eqn . 8 ) . The orientations and degree of anisotropy of the filaments and their barbed ends are indicated for the motile cell ( Row 4 ) and resting cell ( Row 5 ) . Some barbed ends ( Eqn . 13 ) contribute to a protrusive force that pushes out that part of the cell . This results in the spontaneous formation of a leading edge that defines the front of the cell . As seen in Fig . 2 , Rho is highest at the rear of the polarized cell . This leads to a distributed isotropic inwards contractility that causes retraction , and formation of a trailing edge that becomes the “rear” of the cell . Following a stimulus , there is a transient reorganization in the biochemistry and then a sharp transition is formed separating the zone of high Cdc42 activity ( “front” ) from the zone of low Cdc42 activity ( “back” ) . We refer to the border between these zones as the “front-back interface” . For visual convenience , we use green to denote the mean concentration in all 2D chemical distributions , so the green isocline ( see Cdc42 , Row 1 Fig . 2 ) can be identified with that “front-back interface” . The appearance of robust polarization with a sharp transition zone recapitulates results of our previous models on GTPases [34]-[36] , where we showed that the proximity of bistable kinetics , mass conservation , and disparity in the rates of diffusion of active and inactive GTPases leads to formation of a zero speed interface separating “front” and “back” in the cell ( wave-pinning ) . ( Even though with PI feedback the transition from front to back gets smeared out , in both cases we can select one of the intermediate isoclines to represent the demarcation between the front and the rear of the cell , and we informally refer to that boundary as an “interface” in both cases . ) As a cell edge extends outwards , chemical isoclines also relocate . Once the cell starts to move , it increases the region of “frontness” . But then , the buildup of active GTPase at the front is at the expense of the inactive GTPase pool . This means that the front-back interface moves forward , compensating for that depletion . A balance occurs when the speed of the front-back interface matches the forward motion of the leading edge of the cell , i . e . moves in perfect pace with the cell edge ( see Video S1 ) . In this sense , the system exhibits a self-correcting internal structure . We later discuss how perturbing this internal chemical distribution causes it to return to the basic robust polarization here described . In view of the above , the basic model reproduces essential aspects of cell motility and reasonable distributions of the signalling chemicals and the cytoskeleton . We can now use this basic simulation run as a control against which to appraise in silico experiments . In Fig . 3 we contrast the GTPase profiles that occur with and without PI feedback in simulations ( a ) , and schematically ( b , c ) . When there is no PI feedback ( ) in this model , and in previous models where PIs were not explicitly included [34] , [36] , profiles of active GTPases are plateaus ( high and low ) separated by a narrow transition zone ( a generic property , discussed mathematically in [35] ) . Without PI feedback , since active zones are flat plateaus , their interactions drop off steeply so that two such plateaus ( Fig . 3c , left ) hardly influence each other at a distance . Moreover , in this case , inactive GTPases ( dot-dash line , Fig . 3b , c left ) are essentially uniform in space [34] , [36] . Increase or decrease in the total amount of the GTPase or its basic activation rate then determines the width of the region of activity . Including PI feedback results in an auto-amplification positive feedback of Rac and Cdc42 ( via ) on themselves . At an intermediate level of PI feedback , this leads to lower , rounded peaks with higher shoulders . For example , corresponds to smoother , more realistic profiles for the active forms of small GTPases ( Fig . 3c , right ) . It also leads to growth in the heights of the active Cdc42/Rac peaks at the expense of the inactive forms . This creates a local depletion of the inactive GTPases , and substantial gradients in their availability i . e . growing activity peaks can ‘rob’ one another by depleting this pool ( Fig . 3b , right ) . In presence of excessively high PI feedback , this can lead to highly localized peaks of GTPases with flattened tails . The stronger the PI feedback , the stronger the kurtosis of GTPase peaks . This results in concurrent depletion of the inactive GTPases as more and more activity is turned on locally at the expense of the global pool of available inactive forms . Peaks of the active small GTPases Cdc42 and Rac correspond to zones of activity that spawn nascent lamellipodia . Hence , the communication of such peaks has an important influence on competition of protrusions . Complex stimuli can lead to multiple zones of protrusion . Without PI feedback , since active zones are flat plateaus that hardly interact , multiple peaks merge on an exponentially slow timescale [35] , too slow for biologically relevant resolution of competing lamellipodia . This will be discussed further in the section on V-shaped gradient stimuli below . As shown in the right panel of Fig . 3c ) , with appropriate feedback from PIs , zones of Rac activity communicate spatially through their augmented depletion of inactive Rac ( similarly for Cdc42 ) . Large-scale gradients of inactive cytosolic GTPases are formed as a result of the intense local exhaustion . This means that the spatial scale of communication is governed by the relatively fast effective cytosolic diffusion of inactive GTPases , rather than by the significantly slower diffusion of the membrane-associated active forms . This implies that communication between competing peaks of active GTPases is accelerated by 100–500 fold due to feedback from PIs . We find that the modulating influence of PIs depends on the right balance between enough feedback for auto-amplification to enhance peaks of GTPase activity , versus excessive feedback that causes overly dramatic kurtosis of those peaks . If the magnitude of the feedback is tuned to values closer to , the resulting Rac and Cdc42 peaks become sharper and more highly localized , with resultant aberrations in cell behaviour ( described below ) . Similar tuning of other parameters has the same consequences . Increasing the kinetic parameters of PIs ( , , , ) or decreasing the rate of diffusion has similar outcomes ( results not shown ) . The availability of inactive small GTPases and factors that influence this similarly play a role . Such factors include availability of GDIs , and their efficacy at extracting inactive small GTPases from the plasma membrane , which affects an effective rate of diffusion of these proteins [34] , [36] . The longer the inactive GTPase spends on the membrane , the smaller this rate of diffusion , and the more significant are the effects of depletion described above . Note that , in contrast , without PI feedback the effective rate of diffusion of the inactive forms only plays a very marginal role , as long as that rate is at least a few times higher than the diffusion rate of the active form . This is due to the fact that without PI feedback the GTPase levels at the flat plateaus are not limited by the diffusion of the inactive form . There is a fine balance , however , between sufficient and excessive autoamplification due to PI feedback . When is too low , as already discussed , the competition of zones of high GTPase activity takes too long to resolve . Having values closer to 1 leads to rapid resolution of competing peaks of GTPase activity , but at the same time , this also tends to “freeze” single peaks , reducing the ability of cells to respond by moving or turning . Consequently , we have found that the most effective strategy for cell motility is to adopt some intermediate level of feedback . There are two distinct geometric factors that affect dynamics: the shape of the cell and the geometry of isoclines . Since actin remodelling is a direct consequence of the signalling system , it is clear that the shape of the cell is downstream of the signalling modules , so this direction of influence is obvious . The possibility that there is feedback in the opposite direction , from cell shape to signalling biochemistry is more subtle . Here we show that cell shape also influences the biochemical kinetics through a geometric ( rather than hard-wired ) effect . This implies a feedback loop between cell shape and intracellular dynamics with important consequences for cell behaviour , such as rate of turning towards an external signal . Mathematical investigations of reaction-diffusion systems have shown that geometry and dynamics are linked . In many systems , it is well known that curvature of a moving interface can locally accelerate or retard its motion [42]–[44] . The shape of the boundary of a domain , and the conditions imposed at those boundaries ( e . g . impermeability ) , put constraints on the possible behaviour . For example , no-flux ( also called Neumann ) boundary conditions ( BCs ) imply the orthogonality of chemical isoclines at points of intersection with the boundary . Given that the cell boundary is nonpermeable for lipids and proteins forming the signalling system , and given that the keratocytes that have been used as a paradigm system for this study have a very flat , almost two-dimensional shape , no flux BCs hold in our model for the GTPases , PIs and signalling components and their isoclines are therefore always perpendicular to the cell edge . As this constraint holds at any time , it implies that isoclines bend or rotate whenever the boundary deforms locally , preserving that orthogonality . In the case of a dynamic cell shape , e . g . when the cell turns or reorients , the regions of locally high curvature on points of the boundary result in deformed and highly curved segments of the isoclines , including the front-back interface discussed above ( that interface is itself an isocline . ) As a result , the effects of heightened curvature drive accelerated dynamics and result in a faster biochemical response . First we analyze the feedback between cell shape and intracellular polarity by uncoupling the dynamics of the cell shape changes from the dynamics of the internal biochemistry . We did this by studying the effect of the shape of the cell perimeter in immobilized cells with PI feedback ( ) and without ( ) , as shown in Fig . 4a and Video S2 . ( Immobilized disk-shaped cells can be obtained in vitro using latrunculin , e . g . , see [45] and others . ) Specifically , we asked how confining the cell to a specific , immobilized elongated shape would impact the chemical polarity in each case . We use an immobile ellipsoidal cell , initially polarized along its shortest axis by means of an external signal . Note that if the cell is polarized along its shortest axis , the front-back interface is parallel to the longest axis . Once the applied stimulus gradient is turned off , the chemical distribution ( but not the cell perimeter ) is allowed to evolve . Interestingly , the direction of polarity spontaneously reorientates to align itself along the longest axis of the cell , thereby minimizing the length of the front-back interface , which becomes positioned along the shortest axis . With PI feedback ( Fig . 4 ( a ) , left ) , the broad region of “frontness” in the cell rapidly relocates to the pole of the ellipsoid . Without PI feedback ( Fig . 4 ( a ) , right ) , the interface also decreases , but spatial coupling is much weaker . Hence , a globally optimal configuration of the active zone is only attained after an excessively long ( biologically unreasonable ) time scale . Note the difference in timescale for repolarization with PI feedback ( within 10 min ) and without it ( more than 90 min ) ( see also Video S2 ) . Again there is an optimal feedback strength , because when , the amplification caused by the PIs becomes too high , causing a freezing of the initial polarization and a complete failure to reorient ( results not shown ) . To dissect this process of interface minimization further , we purposely initiated a circular cell with an irregular “wavy” front-back interface . Fig . 4b illustrates how the curved interface straightens , with highest curvature regions changing most rapidly , so that the overall length and curvature of the interface decreases ( see also Video S3 ) . Again , this process is significantly faster when PI feedback is included ( , top row ) , than when it is absent ( , bottom row ) . Similarly , in silico experiments of “micro-injecting” active Cdc42 in the middle of a polarized cell have a similar effect ( Fig . 4c; Video S3 ) ( see Materials and Methods for details ) . This results in a perturbed interface , which then rapidly reestablishes its flattened geometry . These results together illustrate the PI-enhanced tendency to shorten and flatten the border between the front and the back of the cell . The observation that biochemical kinetics coupled to diffusion can drive the length minimization of the interface between two stable states within a system is known to mathematicians . This fact can be explained by the following argument , as pointed out by a reviewer of this paper . Reaction-diffusion systems with multiple steady states can be represented as gradient flow problems with an assigned effective energy . This “energy” depends on the reaction terms as well as on a gradient squared term that captures the diffusion terms . In such a representation , the stable states of the dynamical system are given by the minima of the reaction “energy” . Given that the gradient flow acts to minimize the energy , the configuration of a spatial system with a fraction in one stable state and the rest in another stable state will therefore evolve so as to continuously reduce the length of the transition region ( i . e . to minimize the integral of the gradient squared of the concentration ) . ( See , e . g . [46] . ) Note that in this study an extra complexity arises because the two “steady states” are a consequence of the fast diffusion of the inactive forms , i . e . the reaction part does not itself entail multiple steady states . Nevertheless , the argumentation underlying the interface length minimization still holds . As noted above , cell shape influences the dynamics of signalling even when the shape of the cell is static . But signalling cascades also cause the shape of a cell to evolve ( unless specifically blocked as in the previous test ) . Thus , cell shape and signalling concurrently influence one another . Here we aim to illustrate the effect of this feedback . Fig . 5 demonstrates the effect of cell shape changes on the internal dynamics . We here contrast the speed with which repolarization occurs in an immobilized cell with static shape ( left sequences in Fig . 5a , b ) versus a cell in which shape is dynamic ( right sequences in Fig . 5a , b ) . Cells were first polarized using standard protocol with a gradient of 10 s duration . At , a new gradient of smaller magnitude was introduced at ( Fig . 5a ) and at ( Fig . 5b ) to the original gradient . For both angles , motile and static cells detected and chemically repolarized to the new gradient , i . e . were capable of changing their directionality to track the new cue . ( Even for the extreme angle of , motile cells performed a “U turn” to align to the new gradient , as described in the literature , e . g . by [47] . ) However , the speed of repolarization is significantly faster in a cell with dynamic shape . The most noticeable acceleration of repolarization is obtained in cases where cell shape changes induce the most dramatic curvatures in the front-back interface ( see also Video S4 ) . For example , during the U-turn , with shape-dependent feedback the cell is able to turn 90 degrees within 10 minutes ( Video S4 , from 13:00 till 23:00 ) , while without feedback during the same period the rotation is only 30 degrees . The feedback from membrane curvature to local biochemistry , and hence to overall polarity , is quantitatively dependent on the magnitude of the curvature changes in the cell shape during the turning of the cell . The strength of the feedback will therefore be more prevalent under conditions that favor more dramatic cell shape changes , basically when the cell interfacial tension is low . Thus , when the membrane coupling energy ( ) and/or membrane stiffness ( ) are lowered , the cell deformations become more extreme , and hence the feedback becomes more pronounced ( see Video S5 ) . The changing cell shape results from actin dynamics described previously . Importantly , no additional feedback from actin to PIs or small GTPases has here been assumed explicitly in the model . Rather , the dynamic shape itself leads to a faster chemical repolarization . With the above observations , we are now in position to understand the results of further in silico experiments . As shown above , when there is a single front and back , the interface separating these is linked to the cell shape and to the dynamics of the leading edge . Here we asked what happens when there are multiple interfaces , due , for example , to many “fronts” that form spontaneously . To address this question , we challenged the model cell with a variety of stimuli to investigate its ability to resolve multiple conflicting cues , comparable to in vitro experiments . Details of the stimulus protocol are given in the Materials and Methods .
The famous in vitro clip of a neutrophil navigating between red blood cells while chasing a bacterium reinforces our intuition that real cells encounter complex environments where multiple decisions and rapid changes in orientation are essential . This is even more dramatic in vivo , as shown , e . g . in [8] . To understand how cells can respond to such cues , we here simulated actin-based cell motility and its regulation by small GTPases , modulated by feedback from the phosphoinositides . The main motivations were ( 1 ) to explore the role of PIs in fine-tuning direction sensing in response to complex stimuli and ( 2 ) to demonstrate the bi-directional feedback between the signalling modules and the dynamically evolving shape of the cell . We explored a biophysical feedback , wherein geometric effects , and cell shape affected the biochemical dynamics . When local protrusion occurs , the cell perimeter becomes extended and curved , causing chemical isoclines to be curved . The curvature-reducing effect of the reaction-diffusion dynamics then speeds up the response significantly relative to the basal rate of polarization for a cell with static shape . The internal chemical pattern , in turn , specifies the sites where actin nucleation and growth will lead to protrusion of the cell edge , and affect the isoclines yet again , closing the feedback loop between biochemistry and cell shape . While the importance of morphology and cell shape has been discussed in other papers [48]–[50] , here we have incorporated full dynamics changes in both shape and chemical distribution , allowing for feedback in both directions . The implication of this finding is that cell shape is not just a downstream consequence of regulatory pathways that impinge on the cytoskeleton but rather , an integral part of the feedback mechanism . Papers in the literature have suggested that actin filaments feed back onto PI localization . Here we have shown that part of that feedback could stem directly from the changing cell shape , and not only from a direct interaction between actin and PIs . Mechanical effects ( e . g . via integrin signalling not here considered ) would substantially magnify such purely geometric effects . As shown in Figs . 4–6 , a frozen cell shape with absence of feedback from PIs to GTPases takes up to 10 times longer to respond to a repolarizing signal , or to decide between conflicting cues . Cells with dynamic shapes respond more quickly , and those with PI feedback in the appropriate range ( not too high , and not too low ) are even faster . A second theme in our paper concerns chemical feedbacks . In our simulations , decreasing the parameter ( ) corresponds to a gradual PI3K silencing ( e . g . as in [8] with the PI3K inhibitor , LY294002 ) and we have here examined the effects of tuning this parameter between overexpressed PI3K to full knockout . We have shown that both extremes are pathological , so that wild-type behaviour resides at some optimal level of feedback ( ) . In particular , peaks of Cdc42/Rac activity tend to be platykurtic when PIs have no feedback ( ) , leptokurtic at high levels of feedback ( ) . In the former case , communication between plateaus of activity is restricted . In the latter case , shoulders are broader , and so the zones of activity interact directly . Autoamplification due to PI feedback also raises the magnitudes of the activity zones , and depletes the inactive GTPases , leading to longer-range global communication and depression of competing peaks . We found that PI feedback works optimally at some intermediate level . At that level , it can help to speed up the response to new stimuli and to resolve confusing or contradictory external cues . As PI influence is tuned to higher intensity ( ) , the ability to displace a peak of active Cdc42 ( Rac ) decreases , but the ability to resolve conflicts by a “winner take all” mechanism increases . A too-high PI feedback is inappropriate for motile cells exposed to challenges such as conflicting stimuli or obstacles . We showed that if PI feedback is too strong , cells get pinned to an obstacle or face difficulty in reorienting to new cues . ( This would present serious challenges in the complex environment of a living tissue . ) The optimal level of feedback from PIs depends on the type of cell and its function , and whether multiple peaks of Cdc42 or a single peak is needed for some cell function . Plant cells have ROPs , which are analogues to the small GTPases of the Rho family [51]–[56] . For example , ROP 2 and 4 play a role similar to Cdc42 , defining a “leading edge” zone whereas ROP 6 in plants is analogous to Rho in animal cells . In pavement cells of plants , for instance , multiple lobes are a functionally important feature to allow cells to interlock like jigsaw puzzle pieces . Hence , multiple static peaks of small GTPase activity are observed , suggesting that a different PI feedback would be optimal there: either higher or much lower . For , we observe firm immovable peaks of GTPase activity , and for we found that distant peaks hardly interact . Such extremes would possibly fit the plant repertoire more closely . Also , the specific details of the small G-protein crosstalk can significantly differ between different experimental systems , and this can influence both biophysical and chemical feedbacks . For example , using biosensors for the three Rho GTPases and mouse embyonic fibroblasts ( MEFs ) , Machacek . et al . [57] showed that Cdc42 activates Rac1 , as is assumed in this study . They , however , find mutual inhibition between Rac1 and RhoA , and , contrary to older work by Bourne's lab , they find that RhoA gets activated right at the advancing cell edge , and that Cdc42 and Rac1 are activated farther back with a delay ( ) . It remains to be seen to what extent such findings are cell-specific or wide-spread . By exploring this model , we gain several insights that help to understand the biochemistry . For example , the importance of GDIs emerges from our analysis of factors that influence communication of activity peaks . We argued that one such factor in peak communication is the “effective cytosolic diffusion” of inactive small GTPases . Upregulating GDIs extracts inactive small GTPases from the cell membrane , effectively increasing their diffusivity . Downregulating GDIs means that inactive small GTPases spend more time on the membrane , and have smaller diffusivity [58] . Similarly , increasing the kinetics of the PIs ( equivalent to up/down regulating PTEN , PI3K , etc ) produces an analogous tuning of the interpeak communication . Such parameters are tuneable outcomes of evolution , with species-specific and cell-type-specific variability . A range of dynamical effects would thus be expected in control and mutant cells , or cells treated with inhibitors or drugs . Recent reviews of the models for eukaryotic chemotaxis and their relation to experiments include [33] , [59]–[61] . Existing models based on Rho GTPase and/or PI signalling [48] , [62]–[67] are mainly concerned with explaining polarization . Other theoretical models [68]–[71] describe circuits with capability for adaptation , direction sensing , or polarization . Previous models for 2D cell motility include steady state cell shapes [72]–[74] and evolving shapes using force-based methods [75] , [76] . Recent computational models for cell motility have also been based on level-set approaches [77]–[79] phase-field methods [80] , and other approaches [81] . Our model , based on energy minimization [40] , [41] allows for rapid and convenient reaction diffusion of chemicals on an irregular domain , and for effective forces of protrusion and contraction that can be put into correspondence with real forces due to actin filament barbed ends and actomyosin contraction [36] . This method has the advantage of providing a good description of thermal-noise induced stochastic shape change of the cell edges , while affording speed and efficiency of computation . Such energy-minimization techniques have become more widely adopted for describing cells and tissues [82]–[84] because they can dramatically speed computation . The efficiency of the implementation allowed us to focus on exploring the response of the model cell to specific stimuli , with a variety of geometries . Future work should address a comparison of similar ideas in other 2D simulation platforms . We anticipate that results discussed here would carry over universally to a variety of approaches for capturing the evolving shape of the cell . Using a mathematical model , Meyers . et al . [85] considered the effect of cell spreading ( and flattening ) on rates of ( de ) phosphorylation due to proximity of the plasma membrane to cytosolic intermediates . They noted that this effective change in activation/deactivation rates links cell size and shape to regulation of signaling pathways . While they were concerned with the “thickness” dimension of motile cells ( that we take to be constant ) , we are here describing the effects of curvature and 2D cell boundary shape on the dynamics of interfaces of the internal RD system . Model limitations include absence of direct mechanical forces and integrin signalling . Thus , this model would not be appropriate for describing keratocytes “bouncing” off walls they encounter , or cells following mechanical cues . The pathways and rate constants used for the signalling module could be variations on specific versions at play in specific cell types and conditions , but behaviour was robust to modest changes in most parameter values . In [35] , we showed that far simpler GTPase circuits ( consisting of a single GTPase in its active and inactive forms ) can already account for polarization reinforcing our belief that such overall dynamic motifs could operate in a more universal setting . Other cases where mutual inhibition between Rac and Rho are dominant would retain many of the overall features described here , while differing in subtle details , as do cells of distinct species . Also , the model does not capture possible effects of fluid convection in the cytosol ( see Material and Methods for details on our implementation of moving boundary conditions ) . It would be interesting for future studies to address possible effects of intracellular convection by implementing reactant transport and developing a more complete description of the actin network , membrane and cytoplasmic flow of the moving cell . While it has been shown experimentally that PI3K is not essential for chemotaxis in Dictyostelium discoideum , Yoo . et al . [8] found that PI3K is required for the interstitial migration of neutrophils in live zebrafish embryos . The mechanism of this effect was difficult to untangle . Our work in this paper highlights the fact the PI3K product ( and other PIs ) facilitate the resolution of contradictory or multiple stimuli to the Rho GTPases . Such complex stimuli arise repeatedly as cells navigate through the complex environment of tissues ( where blood vessels , other cells , or structures create obstacles that have to be circumnavigated ) . The results suggest a number of important experimental investigations . First , although more data is becoming available , simultaneous measurement of the distributions of multiple GTPases and PIs in single cells is rare . Obtaining such correlated data would be valuable in characterizing the typical resting and stimulated states . Second , to check the effect of PI feedback , pharmacological inhibitors of PI3K such as LY294002 applied at successive level ( very weak , to full inhibition ) , or mutants lacking PI3K could be compared with wild-type behaviour . To detect the differences , it would be important to challenge both treated/mutant cells and wild-type cells with multiple stimuli ( as we have done in silico ) or environments with obstacles to be resolved . Third , to check the predictions about cell shape , one can compare responses of cells treated with latrunculin ( where the actin cytoskeleton is disrupted so that cell shape does not change ) with untreated cells . When both are subjected to the appropriate time-varying stimuli , it would be possible to test our prediction that shape provides an additional feedback to speed of repolarization . As this effect can be subtle , one would need stimuli that lead to dramatic shape changes in the untreated cells to detect a substantial difference .
The parameter regime we use here allows the cell to have both a stable rest state as well as a polarizable state . Standard Polarization: Polarity is initiated by applying a transient ( ) spatial gradient in ( the Cdc42 activation rate ) with slope , which corresponds to a roughly 15% variation across the cell . The gradient is then turned off . Repolarization: ( Fig . 5 ) The cell is polarized as before . After 5 min , we applied a shallower second gradient , corresponding to roughly 3% variation in the Cdc42 basal activation rate ( values ) across the cell . The second stimulus was either orthogonal ( Fig . 5a ) or opposite ( Fig . 5b ) to the direction of the first . Shown are results for . ( Here turning is slightly faster than for , though qualitatively similar . ) We used here ( rather than ) to slightly enlarge the front of the cell , and make the curved front-back interface visually more pronounced . Initial Predetermined Patterns: The intracellular distributions for the simulations of Fig . 4b were initialized by horizontally shifting a stabilized intracellular pattern in a sinusoidal fashion , such that all chemical species ( PIs and ( in ) active forms of the small GTPases ) have corresponding lower and higher levels , distributed in a sinusoidal pattern with three peaks . Injection: In the simulation of Fig . 4c , a spot of activation of Cdc42 was introduced by adding active Cdc42 concentrations at the highest level as found within the cell and diminishing this amount from the inactive pool . V-shaped Gradient: In Fig . 6a , b a V-shaped gradient across the horizontal axis ( from right to left extremities of the cell ) of 7 . 5% difference in values was employed . ( By a V-shaped gradient , we mean two superimposed , simultaneous , diametrically opposed gradients . ) Noise: In Fig . 6c , the initial condition was a cell at steady state with uniform concentrations of all substances . On this we superimposed , only for the initial condition , normally distributed noise in the Cdc42 and Rac concentrations , with a spatial autocorrelation distance of and a standard deviation of , i . e . for Cdc42 , and for Rac ( see the first frame of Fig . 6c for a visual display of the initial noise level ) . A relatively high amount and/or spatially correlated noise is needed to push the cell out of the stable resting state . ( We tested that continuously adding such a level of noise does not significantly change the results , as noise has only a small effect on the cell once polarity has been established . ) Reactions and cell shape are computed on a 2D hexagonal grid . The cell , in a top-down view ( approximated as having constant thickness ) is represented as a set of pixel points on that lattice . Both cell interior ( cytosol ) and membrane are so represented , and chemical concentrations ( in number of molecules per hexagonal cylinder of constant thickness ) are tracked by implicitly solving the reaction-diffusion PDEs on the evolving domain . As the intracellular small G-protein and PI dynamics evolve , leading to down-stream effects upon the cytoskeleton , forces generated by the actin barbed-ends and myosin contraction change the cell's shape . To study the resulting cellular dynamics , and how these influence the internal chemical dynamics , we utilized a modelling framework in which membrane displacement is described according to an energy function . This is an approach that has recently become more widely recognized in modelling cell and tissue movement [82]–[84] . The core of this energy function includes biophysical properties such as cell adhesion , cell volume conservation , membrane and cortical tension , which together lead to an effective cell surface tension [83] . We utilize such an energy description within the Cellular Potts Model [40] , [86] to describe the dynamics of the change of the cell's edge . The Hamiltonian is defined by summing the energy contribution of each pixel over the entire field: ( 14 ) ( summed over neighbours up to order ) . In 2D , depends on cell area and boundary length ( in 3D , on volume and surface area ) . is the coupling energy per boundary site , is the actual cell area , the target area , and a parameter that describes resistance to deviation from the target area , describes resistance to changing the perimeter away from a target perimeter . The perimeter constraint represents a high effective interfacial tension and energetic costs of stretching the cell membrane . The dynamics of cellular movement result from the above Hamiltonian through the Monte Carlo simulation utilizing the Metropolis Algorithm , an energy-minimization method that allows the cell edge to change stochastically . Briefly , during each Monte Carlo step ( MCS ) , each lattice site in the field will be evaluated in a random sequence . Sites at the cell's perimeter are queried for possible change to the state of a randomly chosen neighbour ( “copying” ) . In the simulations here , in which we consider single cell dynamics , this local change implies protrusion or retraction of the edge of the cell . The net change of energy due to a “copying” event , is computed , and the event accepted with probability ( 15 ) where represents a yield , which is the ability of the membrane to resist a force , and determines the fluctuations . Changes in state that decrease by at least have probability 1 , and other changes are made with a Boltzmann probability . Tuning the ‘temperature’ allows us to tune the magnitude of stochastic fluctuations ( of various possible origins ) in the model . For example , Mombach et al . [87] interpreted the parameter as the membrane fluctuation amplitude of cells , and they compared this with effects of the drug cytochalasin-B ( a suppressor of membrane ruffling ) . Here , given that we describe the state of the cytoskeleton at the membrane , we are able to directly relate this parameter to the density and biophysical properties of the actin barbed ends . Note however , that the cell is not expected to relax to a surface-driven equilibrium shape , as there are internal forces generated by the force-bearing barbed ends at the membrane . Thus , we describe these internal forces by altering probabilities of expansion/retraction dependent on the internal densities of barbed ends at the membrane as well as on the amount of myosin contraction as a downstream effect of Rho . Presence of barbed ends biases the probability towards protrusion , whereas presence of Rho GTPase biases towards retraction ( see Fig . 9 for a schematic representation ) , and leads to the following forces at the membrane: ( 16 ) describes the forces exerted by all barbed ends pushing against the membrane towards the empty site . The term describes the effective Rho-dependent contraction forces when Rho exceeds the threshold level , ( ) . The term scales a unit of Rho elevation to the force of one pushing barbed end per nm membrane length . Note that according to Eqn . 16 , ( and thereby and ) carry the same units as , i . e . the number of extending filaments pushing against the membrane per unit edge length ( here ) . We can further relate the above expression to known physics of cell protrusion . An effective force-velocity relationship for protrusion speed as a function of the number of barbed ends pushing at the cell edge has previously been derived [37] , [88] . In a thermal ratchet driven by actin polymerization , the relationship between the number of barbed ends at the membrane and the speed , , of the lamellipodial protrusion is approximately ( 17 ) where is the free polymerisation speed , the density of barbed ends per unit length at the membrane , and the renormalised membrane resistance force per unit length ( , where is the membrane resistance , the size of one monomer , and is the thermal energy ) . Neglecting capping and side-branching , and assuming that all barbed ends are directed normal to a straight cell edge , it can be shown [36] , [41] that within the Cellular Potts Model Eqn . 16 implies a mean speed of protrusion ( 18 ) Here and are the grid size and time step corresponding to one MCS , respectively . This is in line with Mogilner and Edelstein-Keshet [88] . While not identical to Eqn . ( 17 ) , this equation also describes a relationship between protrusion velocity and the number of barbed ends . Here the relationship is expressed in terms of the CPM parameters and . By fitting this relationship to Eqn . ( 17 ) ( for which the parameter values are well-established ) , we obtained the optimal values ; . For these values , the thermal ratchet force-velocity relationship of Eqn . 17 and the effective force-velocity relationship of Eqn . 18 are highly comparable over the whole range of biologically relevant barbed end densities , which are typically observed to be in the range of at the lamellipod edge [88] , [89] . Accordingly , the CPM gracefully leads to a reasonable depiction of actin-based protrusion forces and the model quantitatively describes the response of the cell membrane to any possible load of pushing barbed ends . Having matched this relationship , we can now apply it in a simulation of a complex shaped 2D motile cell , with large variations in pushing barbed ends along the edge , implicitly locally solving for the large variation in the applied forces . This also allows us to determine the feedback between the cell shape and deformation on the underlying cytoskeleton dynamics . Further details of how the model has been parametrized to biophysical measurements are given in [36] , [41] . We use a 400×400 hexagonal grid with periodic ( toroidal ) boundary conditions and grid mesh size equivalent to . A time step corresponds to , and the same timestep is used to numerically integrate the PDEs . Diffusion processes were integrated using the Alternating Direction Implicit ( ADI ) method [90] , but modified to be performed in units of one-third timestep along each of the three principal directions given by the hexagonal symmetry of the field . At a retracting site , all filaments and barbed ends that were in that site are pushed back with the edge , and pile up at adjacent sites with their original orientations . Their barbed ends push against the new edge , and some become load-bearing . When the edge protrudes outwards , barbed ends formerly pushing lose contact with the membrane . In this way , filaments and barbed ends are not lost or generated de novo when the membrane retracts or extends , and the build-up and release of internal forces are directly coupled to the cytoskeleton . Fig . 9 illustrates this process . Moreover , as the cell moves ( due to the dynamics given by Eqn . 16 ) the edge of the cell deforms , and hence the local unit normal vector changes , which results in changes of the boundary conditions for which the intracellular dynamics are run ( see Eqn . ( 6 ) ) . As explained above , we choose to update concentrations locally in such a way that we preserve mass conservation ( nothing is added or lost in a pixel extension or retraction ) .
|
Single cells , such as amoeba and white blood cells , change shape and move in response to environmental stimuli . Their behaviour is a consequence of the intracellular properties balanced by external forces . The internal regulation is modulated by several proteins that interact with one another and with membrane lipids . We examine , through in silico experiments using a computational model of a moving cell , the interactions of an important class of such proteins ( Rho GTPases ) and lipids ( phosphoinositides , PIs ) , their spatial redistribution , and how they affect and are affected by cell shape . Certain GTPases promote the assembly of the actin cytoskeleton . This then leads to the formation of a cell protrusion , the leading edge . The feedback between PIs and GTPases facilitates global communication across the cell , ensuring that multiple , complex , or rapidly changing stimuli can be resolved into a single decision for positioning the leading edge . Interestingly , the cell shape itself affects the intracellular biochemistry , resulting from interactions between the curvature of the chemical fronts and the cell edge . Cells with static shapes consequently respond more slowly to reorienting stimuli than cells with dynamic shape changes . This potential to respond more rapidly to external stimuli depends on the degree of cellular shape deformation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology",
"computational",
"biology",
"biophysics"
] |
2012
|
How Cells Integrate Complex Stimuli: The Effect of Feedback from Phosphoinositides and Cell Shape on Cell Polarization and Motility
|
In plants , innate immune responses are initiated by plasma membrane-located pattern recognition receptors ( PRRs ) upon recognition of elicitors , including exogenous pathogen-associated molecular patterns ( PAMPs ) and endogenous damage-associated molecular patterns ( DAMPs ) . Arabidopsis thaliana produces more than 1000 secreted peptide candidates , but it has yet to be established whether any of these act as elicitors . Here we identified an A . thaliana gene family encoding precursors of PAMP-induced secreted peptides ( prePIPs ) through an in-silico approach . The expression of some members of the family , including prePIP1 and prePIP2 , is induced by a variety of pathogens and elicitors . Subcellular localization and proteolytic processing analyses demonstrated that the prePIP1 product is secreted into extracellular spaces where it is cleaved at the C-terminus . Overexpression of prePIP1 and prePIP2 , or exogenous application of PIP1 and PIP2 synthetic peptides corresponding to the C-terminal conserved regions in prePIP1 and prePIP2 , enhanced immune responses and pathogen resistance in A . thaliana . Genetic and biochemical analyses suggested that the receptor-like kinase 7 ( RLK7 ) functions as a receptor of PIP1 . Once perceived by RLK7 , PIP1 initiates overlapping and distinct immune signaling responses together with the DAMP PEP1 . PIP1 and PEP1 cooperate in amplifying the immune responses triggered by the PAMP flagellin . Collectively , these studies provide significant insights into immune modulation by Arabidopsis endogenous secreted peptides .
Immune signaling in plants is typically initiated when immune-related receptors perceive the presence of pathogen molecules , including so-called “pathogen-associated molecular patterns” ( PAMPs ) and race-specific effectors [1] . PAMPs , such as bacterial flagellin and fungal chitin , are recognized by plasma membrane-located pattern recognition receptors ( PRRs ) , which activate PAMP-triggered immunity ( PTI ) . In addition , pathogen infection causes the release of endogenous damage-associated molecular patterns ( DAMPs ) , such as peptides , oligogalacturonides ( OGs ) , or cutin monomers . DAMPs are released from the cytoplasm or the cell wall into the extracellular space , where they induce immune responses resembling PTI following perception by PRRs [2]–[4] . Over a dozen PRRs have been identified . Most belong to the superfamily of receptor-like kinases ( RLKs ) , characterized by an extracellular domain , a transmembrane region and a cytoplasmic kinase domain . Arabidopsis thaliana has more than 600 RLKs . Among these , the leucine-rich repeat RLKs ( LRR-RLKs ) constitute the largest group which has been divided into 13 categories ( I through XIII ) [5] . Flagellin-sensitive 2 ( FLS2 ) , a LRR-RLK from category XII , binds to a 22 residue epitope ( flg22 ) present at the N terminus of flagellin from Gram-negative bacteria [6] . Perception of flg22 induces the dimerization and rapid phosphorylation of FLS2 and BRASSINOSTEROID INSENSITIVE 1-associated receptor kinase 1 ( BAK1 ) , as well as phosphorylation of the receptor-like cytoplasmic kinase ( RLCKs ) BIK1 [7]–[10] . The activated receptor complex triggers elevation of cytosolic calcium , generation of reactive oxygen species ( ROS ) , phosphorylation of mitogen-activated protein kinases ( MAPKs ) , callose deposition , and transcriptional reprogramming of the cell , leading to enhanced resistance against pathogens [11]–[14] . PEP1 was identified as an extracellular 23-aa peptide derived from the C-terminus of the A . thaliana precursor protein proPEP1 . Since proPEP1 lacks an N-terminal signal peptide , the release of PEP1 into the apoplast was suggested to result from cellular damage caused by pathogen attack or wounding , suggesting that PEP1 functions as DAMP . Two XI category LRR-RLKs , PEPR1 and PEPR2 , were shown to act as receptors of PEP1 and homologous peptides in A . thaliana [15] , [16] . Perception of PEP1 by PEPR1/2 activates PTI and enhances host resistance against the pathogens Pseudomonas syringae and Pythium irregulare [3] , [16] . PEPR1 also modulates ethylene ( ET ) -dependent resistance to Botrytis cinerea via the phosphorylation of BIK1 [17] , [18] . Since expression of PEP1-PEPR1/2 is induced by flg22 and PEP1 itself , and since PEP1-PEPR1/2 employs shared components with PAMPs signaling , PEP1-PEPR1/2 has been proposed to function as an amplifier of PTI signaling [16] , [19] . Secreted peptides coordinate a variety of plant developmental processes , including stem cell maintenance , stomatal development , lateral root initiation , vascular formation , floral abscission and cell expansion [20]–[22] . Recently , several secreted peptides have been reported to modulate plant immune signaling . For instance , the CLAVATA3 peptide ( CLV3p ) , known to regulate stem cell homeostasis in the shoot apical meristem , was suggested to be recognized by FLS2 and activate FLS2-dependent immune responses in the shoot meristem [23] . The sulfated peptides phytosulfokine ( PSK ) and PSY1 , were initially identified as promoters of cell proliferation and tissue growth , and were recently shown to attenuate PTI responses and to enhance susceptibility to biotrophic pathogen and resistance to necrotrophic pathogen [24] , [25] . A . thaliana has been suggested to produce over 1000 secreted peptides [26] , the overwhelming majority of which remain functionally uncharacterized . To look for secreted peptides potentially involved in regulation of immunity , we searched the available A . thaliana microarray data for flg22- and elf18-induced genes . This led to the identification of a novel gene family of secreted peptide precursors , termed “prePIPs” ( precursors of PAMP-Induced Peptides ) . We provide evidence showing that PIP1 and PIP2 , two peptides obtained from processing of the representative prePIP family members prePIP1 and prePIP2 , are able to activate immune responses in A . thaliana and to enhance resistance against P . syringae and Fusarium oxysporum . Using a reverse genetics approach , we demonstrate that RLK7 , a class XI LRR-RLK , is required for PIP1 and PIP2-elicited immune activation , and that PIP1-RLK7 has a crucial role in PTI amplification .
Analysis of flg22- and elf18-induced transcription data ( microarray accession number E-MEXP-547 ) resulted in the identification of 12 genes encoding putative secreted peptide precursors [27] . The predicted gene products were 70–110 amino acid residues in length and included an N terminal signal peptide , as predicted by the SignalP 3 . 0 server [28] . Of these , four have known or predicted functions . They include PSK4 precursor [29] , PSY1 precursor [30] , IDA [31] , and an IDA-like protein ( At1g05300 ) . The other eight are functionally uncharacterized . Three of these eight genes ( At4g28460 , At4g37290 , and At2g23270 ) share a highly conserved C terminus , and their products were named prePIP1 , prePIP2 and prePIP3 , respectively ( Table S1 ) . A blastp search based on the prePIP1 C terminus sequence revealed that A . thaliana has at least 11 prePIP homologs , including seven annotated and four non-annotated proteins . Orthologs of prePIP proteins are present in numerous species of dicots and monocots , such as soybean , grape , maize , and rice ( Figure S1 ) . All the prePIP family members exhibit the hallmarks of post-translationally modified secreted peptide precursors: a signal peptide at the N terminus , a highly conserved cysteine-poor region at the C-terminus ( hereafter referred to as the SGPS motif ) , and a variable region between the signal peptide and the SGPS motif ( Figure 1A ) [20] . Eight A . thaliana family members contain a single SGPS motif while three ( prePIP2 , prePIP3 and prePIPL1 ) harbor two SGPS motifs . The prePIP SGPS motif shares structural features with CLV3/CLE peptides [32] , [33] , the IDA peptide ( IDAp ) , CEP1 [34] , and PEP1 [3] . Since all these peptides carry conserved Ser , Gly , Pro , and His residues ( Figure S2 ) , we propose that they form a superfamily called “SGP-rich” peptides . During the process of translation , the prepropeptide , the original form of secreted peptide precursor , is targeted to the endoplasmic reticulum/Golgi-dependent secretory pathway where the N-terminal signal peptide is removed resulting in the propeptide . The propeptide is subsequently secreted into the apoplast and subjected to proteolytic processing , releasing the mature C-terminal peptide [20] . To experimentally determine whether the prePIP1 propeptide is secreted , the green fluorescent protein gene ( GFP ) was fused to the C-terminus of prePIP1 ( prePIP1-GFP ) under the control of the cauliflower mosaic virus 35S ( CaMV 35S ) promoter and transiently expressed in tobacco leaves using agro-infiltration . Confocal microscopy imaging showed that prePIP1-GFP fluorescence was distributed in the pericellular apoplastic space . In contrast , GFP protein alone was present in the cytoplasm and the nucleus . The secreted peptide precursor CLV3 , which was previously shown to localize in the extracellular matrix , exhibited a similar localization as prePIP1-GFP when a C-terminal GFP fusion allele was expressed in tobacco leaves ( Figure 1B ) . These results suggest that the prePIP1 product is secreted into the plant extracellular space . An in-vitro assay was conducted to determine whether prePIP1 and prePIP2 are proteolytically processed . Glutathione S-transferase-tagged signal peptide-deleted prePIP1 and prePIP2 ( GST-ΔP1 and GST-ΔP2 ) were expressed in E . coli strain BL21 ( DE3 ) and purified through Glutathione Sepharose ( Figure S3 ) . Incubation of GST-ΔP1 and GST-ΔP2 in a reaction solution supplementing extracts of A . thaliana seedlings but not BSA ( negative control ) resulted in a reduction of 1–2 kDa in size ( Figure 1C and D ) . When GST-ΔP1 was injected into A . thaliana leaves , a similar reduction in molecular size was detectable after a 2 h incubation , consistent with a cleavage of GST-ΔP1 by a plant protease ( s ) present in the extracellular space ( Figure S4 ) . Transgenic plants carrying the GFP gene under control of the prePIP1 promoter exhibited strong fluorescence in guard cells , hydathodes and vascular tissue ( Figure 2A ) . Interestingly , all these tissues represent either potential entry points or proliferation routes for invading pathogens . In contrast , no fluorescence was detected in these tissues in untransformed plants ( data not shown ) . When A . thaliana seedlings were exposed to flg22 or chitin , prePIP1 transcription was markedly up-regulated ( Figure 2B ) . Subsequent experiments , based either on transcript abundance or on the expression of a transgene carrying the β-glucuronidase ( GUS ) gene driven by the prePIP1 promoter , confirmed that prePIP1 was up-regulated during infection with the bacterial pathogen P . syringae DC3000 ( Pst DC3000 ) or with the fungal pathogen F . oxysporum f . sp . conglutinans strain 699 ( Foc 699 ) ( Figure 2C ) . Transcript abundance increased about eight folds following inoculation with Pst DC3000 and about 15 folds with Foc 699 , extending throughout the leaf and root system within 24 h after inoculation ( Figure 2C–E ) . PrePIP1 expression was also increased in A . thaliana seedlings after treatment with immune-related phytohormones . Quantitative RT-PCR ( RT-qPCR ) analysis showed that the prePIP1 transcript was induced by methyl salicylate ( MeSA ) , but not by methyl jasmonate ( MeJA ) or the ethylene precursor 1-aminocyclopropane-1-carboxylate ( ACC ) . Importantly , expression of the SA pathway marker pathogenesis-related protein 1 ( PR1 ) and of the JA pathway marker PDF1 . 2 was induced by MeSA and MeJA treatments , respectively ( Figure 2F ) . A . thaliana transgenic lines overexpressing prePIP1 or prePIP2 ( 35S::prePIP1 and 35S::prePIP2 ) ( Figure 3A ) consistently exhibited a shorter main root than the wild type ( WT ) plants ( Figure 3B and C ) . In contrast , transgenic plants overexpressing IDA and IDA-like were abnormal with respect to their floral abscission zone ( AZ ) [35] . In spite of the high sequence similarity between the C termini of IDA and prePIPs , overexpression of prePIP1 or prePIP2 did not affect AZ structure ( Figure S5 ) , indicating different functions of the two protein families . Because post-translationally modified secreted peptides generally coincide with the C-terminal conserved region of their precursors [20] , exogenous application of synthetic peptides such as CLV3p , IDAp , and CEP1 reproduces the phenotypes of overexpression lines of the respective precursor gene [32] , [34] , [35] . We tested whether addition of synthetic peptide PIP10 comprising the conserved SGPS-motif of prePIP1 , could reproduce the effect on root growth of prePIP1 overexpression . PIP10 significantly inhibited the elongation of the main root when applied at a concentration of 100 nM ( Figure 3D ) . Since SGP-rich peptides usually undergo proline hydroxylation , the inhibitory effect on root growth of three PIP1 derivatives , PIP1Hpy6 , PIP1Hpy8 , and PIP1Hpy6 , 8 ( Table S2 ) was investigated . Of these , PIP1Hyp6 ( hereafter denoted “PIP1” ) and PIP1Hpy6 , 8 were more active than PIP10 ( Figure 3D ) , suggesting that proline hydroxylation at position 6 contributes to biological activity of the peptide . PIP1 activity was also pH dependent , since root growth inhibition was most active in the pH range 5 . 8–6 . 8 ( Figure 3E ) . Results for PIP2 , the synthetic hydroxylated peptide corresponding to prePIP2 , were similar to those obtained with PIP1 ( Figure 3F ) . The role of the peptide derived from prePIP1 in plant immunity was explored initially using a transient expression assay in mesophyll protoplasts . The firefly luciferase gene ( LUC ) driven by the promoter of Flg22-induced Receptor-like Kinase 1 ( FRK1 ) , a marker gene of PTI signaling , was co-transfected as a reporter with either prePIP1 , prePIP1ΔSP ( prePIP1 lacking the signal peptide ) , or prePIP1ΔSGPS ( prePIP1 lacking the SGPS-motif ) all driven by the CaMV 35S promoter . Activation of the FRK1 promoter was only detected with a full length copy of prePIP1 , implying that both secretion of prePIP1 and its SGPS-motif are required for FRK1 induction ( Figure 4A and B ) . Similar to flg22 and PEP1 , exogenous application of PIP1 and PIP2 induced the expression of pFRK1::LUC in protoplasts , but neither IDL2 nor CEP1 did ( Figure 4C ) . Moreover , RT-PCR and RT-qPCR analyses revealed that PIP1 and PIP2 induced transcription of the immune response genes FRK1 , WRKY30 , WRKY33 , WRKY53 , and PR1 ( Figure 5A–C and S6 ) . Other characteristic PTI responses such as stomatal closure ( Figure 5D ) , ROS production ( Figure 5E ) , callose deposition ( Figure 5F ) , and MAPK phosphorylation ( Figure 5G ) were also induced by these two peptides . In comparison with flg22 , PIP1 , PIP2 and PEP1 induced significantly lower ROS production and leaf callose deposition ( Figure 5E and F ) . Similarly , the effect of PIP-induced immunity on host resistance against Pst DC3000 was weaker than that induced by flg22 . Treatment with 1 µM PIP1 or PIP2 delayed Pst DC3000 proliferation in leaves by ∼70% , while 1 µM flg22 decreased bacterial growth by >90% ( Figure 5H ) . The prePIP1 gene is abundantly expressed in A . thaliana roots . We therefore measured PIP-induced immunity in roots using a MYB51p::GUS reporter which was previously employed to monitor flg22-triggerred immune responses [36] . PIP1 , PIP2 , PEP1 and flg22 strongly activated MYB51 promoter activity in the root elongation zone ( EZ ) ( Figure 6A ) . MYB51-dependent indole-glucosinolate synthesis is required for callose deposition [37] . All peptides induced callose deposition in the root EZ ( Figure 6A ) , while no such induction was detectable in the presence of CEP1 or IDL2 ( Figure S7 ) . Given that prePIP1 expression was induced upon Foc 699 infection , resistance against this pathogen was compared between WT and 35S::prePIP1 or 35S::prePIP2 plants . When A . thaliana seedlings were challenged with microconidia of GFP-labeled Foc 699 ( Foc 699-GFP ) , fungal hyphae penetrated the EZ cortex 3–6 h post infection and reached the vascular tissue ∼12 hours later ( Figure S8 ) . However , the extent of Foc 699 penetration in the roots of 35S::prePIP1 and 35S::prePIP2 plants was significantly lower than in the roots of WT , as estimated from the GFP fluorescence signal ( Figure 6B and C ) . When Foc 699 infected seedlings were potted into soil and left to grow for three weeks , the overexpression lines displayed a significantly reduced mortality compared to the WT plants ( Figure 6D ) . These results indicate that overexpression of prePIP1 or prePIP2 enhances Arabidopsis resistance against Foc 699 . Secreted peptides are typically recognized by plasma-localized LRR-RLKs [21] . The sequence similarity between PIPs and other SGP-rich peptides suggested that the hypothetical PIP1 receptor ( s ) could be structurally related to the CLV3p receptor CLV1 [38] , the IDAp receptors HAE and HSL2 [35] , or the PEP1 receptors PEPR1/2 [15] , [16] , all of which are class XI LRR-RLKs [5] . Like PEPR1/2 and other immune-related receptors , the hypothetical PIP1 receptor ( s ) is likely to be up-regulated by pathogen attack or PAMP induction . The A . thaliana genome harbors 28 category XI LRR-RLKs genes , six of which are induced by PAMP treatment or pathogen infection [27] , [ 39]: PEPR1/2 , HAE , RLK7 ( At1g09970 ) , At5g25930 ( here named HSL3 ) , and SOBIR1 . The SOBIR1 product was shown to act as a co-regulator of multiple receptor-like proteins ( RLPs ) that are involved in immune recognition [40]–[42] , and was suggested not to function directly in ligand recognition due to its short LRR domain . To identify the putative receptors of PIP1 and PIP2 , we analyzed the response of T-DNA insertion mutants of RLK7 , HAE , HSL2 , HSL3 , and FLS2 to PIP1 and PIP2 treatments . No inhibition of root growth was observed in two rlk7 mutants , rlk7-2 and rlk7-3 , while the other mutants responded similar as the WT ( Figures 7A , S9 ) . The roots of 35S::prePIP1 or 35S::prePIP2 plants were significantly shorter than those of WT plants , while roots of the double homozygous F2 progeny of a cross between 35S::prePIP1 or 35S::prePIP2 and rlk7-3 grew normally as did those of rlk7 mutants . Thus , inhibition of root growth by prePIP1 and prePIP2 is RLK7 dependent ( Figure 7B ) . In contrast to the WT , the rlk7-3 plants failed to up-regulate expression of FRK1 , WRKY33 , and WRKY53 upon treatment with PIP1 or PIP2 ( Figure 7C and S10A and B ) . In contrast , flg22 strongly induced expression of FRK1 both in WT and rlk7-3 plants , but not in the fls2 mutant ( Figure S10C ) , suggesting that RLK7 responds specifically to PIPs . Moreover , PIP1-induced MPK3 and MPK6 phosphorylation was also abolished in rlk7-3 ( Figure 7D ) , as was the increase of host resistance against Pst DC3000 infection by pre-treatment of Arabidopsis leaves with PIP1 ( Figure 7E ) . The prePIP1 overexpression line displayed a significantly reduced mortality compared to the WT plants as indicated above , while the double homozygous F2 progeny of a cross between 35S::prePIP1 and rlk7-3 displayed a higher mortality as did those of rlk7-3 mutants ( Figure 7F ) . We next asked whether RLK7 directly binds the PIP1 peptide . This was first addressed through a pull-down assay with biotinylated PIP1 in A . thaliana plants expressing hemagglutinin ( HA ) tagged-RLK7 ( RLK7-HA ) . Two derivatives of biotin labeled PIP1 ( Biotin-PIP1 and PIP1-biotin ) were confirmed to maintain their biological function by determining their activities on root growth inhibition and marker gene induction ( Figure S11 ) . Since PIP1-biotin exhibited a higher activity , it was used for all subsequent experiments . We found that RLK7-HA was pulled down with PIP1-biotin-associated streptavidin beads from membrane protein extracts of rlk7-3 plants harboring RLK7-HA , but not from rlk7-3 plants ( Figure 7G ) . Binding of RLK7-HA to the beads was inhibited by a 100× excess of unlabelled PIP1 but not by unlabelled IDA . Next , a chemical cross-linking assay was employed to prove a direct binding of PIP1-biotin to RLK7-HA . PIP1-biotin peptide was incubated with protein extracts of RLK7-HA transgenic plants or rlk7-3 mutants , and cross-linked with its potential receptor using a chemical cross-linker . After separation by SDS-PAGE , protein samples were hybridized with an anti-biotin antibody . A protein of 130 kD , consistent with the molecular mass of RLK7-HA , was detected in RLK7-HA plants but not in rlk7-3 mutants ( Figure 7H ) , suggesting that the protein corresponds to the RLK7-HA protein . Binding of PIP1 to RLK7 was further corroborated using a photoaffinity labeling assay . RLK7-HA or GFP ( negative control ) were transiently expressed in tobacco leaves , and homogenized leaf tissues were incubated with 1 nM 125I-labeled PIP1 in the presence or absence of 10 µM unlabeled PIP1 . Specific binding of 125I-labeled PIP1 was detected in the homogenate from leaves expressing RLK7-HA protein , but not in those from leaves expressing GFP ( Figure 7I ) . The receptor kinase BAK1 plays an important role in PTI immune activation by forming heteromeric co-receptor complexes with multiple LRR-RLK receptors , including FLS2 and PEPR1 [7] , [8] , [39] . Sensitivity to flg22 and PEP1 was partially reduced in bak1 T-DNA insertion mutants bak1-3 and bak1-4 [43] . While dimerization of FLS2 with BAK1 occurs after flg22 perception by FLS2 , PEPR1 interacts constitutively with the kinase domain of BAK1 . Since PIP1 triggers similar early immune responses as flg22 and PEP1 , we asked whether BAK1 also contributes to PIP1 responses . Indeed , PIP1-induced ROS production and root growth inhibition were both reduced in bak1-4 than in WT plants ( Figure 8A and B ) . In contrast , while PEP1-induced ROS production was also reduced in the bak1-4 mutant , inhibition of root growth was unaffected ( Figure 8A and B ) . Thus , while PIP1-RLK7 signaling is partially dependent on BAK1 , PIP1 and PEP1-induced responses have different requirements for BAK1 . FLS2 and PEPR1 initiate downstream signaling by directly interacting with the receptor-like cytoplasmic kinase BIK1 [9] , [17] . Therefore , we investigated the possible interaction between BIK1 and RLK7 . Yeast two-hybrid results did not indicate an interaction between BIK1 and the kinase domain of RLK7 , while confirming the interaction between BIK1 and the kinase domain of PEPR1 reported previously ( Figure 9A ) . In plants lacking BIK1 , flg22- and PEP1-induced root growth inhibition was attenuated while the effect of PIP1 was unchanged ( Figure 9B ) . Given the known role of PEPR1-BIK1 in mediating ET responses [17] , we compared hypocotyl elongation in WT and rlk7 seedlings treated with ACC , but found no significant difference ( Figure 9C and D ) . However , sensitivity to ACC treatment was attenuated in both ein2 ( ethylene insensitive 2 ) and bik1 mutants . Taken together , these results suggest that PIP1-RLK7 signaling is independent of BIK1 . Because the expression of prePIP1 and RLK7 is induced by flg22 and PIP1 triggers a similar immune response to flg22 , we hypothesized that PIP1-RLK7 , like PEP1-PEPR1 , may serve to amplify PAMP signaling . In support of this idea , flg22- or chitin-induced callose deposition was more pronounced in leaves and roots of 35S::prePIP1 or 35S::prePIP2 plants than in WT plants ( Figure 10A–C ) . Moreover , we observed an additive effect in elevation of host resistance against Pst DC3000 in plants pre-treated simultaneously with flg22 and PIP1 , compared to each single peptide elicitor ( Figure 10D ) . Furthermore , activation of WRKY33 and PR1 , two genes representing , respectively , early- and late-response immune reporters , by flg22 was reduced in rlk7 plants compared to WT plants ( Figure 10E and F ) , and the level of flg22-induced host resistance against Pst DC3118 ( a coronatine deficient Pst DC3000 mutant ) was less marked in the rlk7 mutant ( Figure 10G ) . Finally , PIP1 and PEP1 both appeared to enhance flg22 responses via up-regulation of FLS2 expression ( Figure 10H ) . A crosstalk between PIP1 and PEP1 signaling was further supported by the finding that PEP1-induced root growth inhibition and WRKY33 expression were impaired in mutants lacking RLK7 ( Figure S12 ) . Either PEP1 or PIP1 induced the transcription of all the genes encoding precursors and receptors of the two peptides ( Figure 10I–L ) . Thus , PIP1-RLK7 and PEP1-PEPR1 act cooperatively to amplify FLS2-initiated immunity .
The identification of elicitors to date has relied on various bioassays conducted on extracts of pathogen and/or host tissue [3] , [44] , [45] . Because the active components are typically present in low abundance , this mode of analysis is technically challenging . With the widespread development of genomic and transcriptomic data in A . thaliana , bioinformatics is increasingly offering potential for predicting the identity of elicitors . Here , by analyzing PAMP-induced gene transcription data , a gene family encoding precursors of the secreted peptide elicitors PIPs was identified . The release from precursor proteins by proteolysis in the extracellular space is a critical process for secreted peptides [20] . In vitro , prePIPs are typically cleaved close to the C terminus . Specific cleavage was confirmed in vivo , since recombinant GST-ΔP1 protein suffered a similar processing pattern when injected into leaves of A . thaliana . Although it is generally assumed that mature peptides are released from precursors through endopeptidase-mediated cleavage [46] , the only cleavage recognition site identified so far in A . thaliana is a specific sequence in the peptide PSK4 which was confirmed to be proteolytically cleaved by the subtilase SBT1 . 1 [46] . In most post-translationally modified secreted peptide precursors , cleavage occurs before or after Arg , Asp , His or Asn residues located at both sites of the C-terminal conserved motifs [20] . Members of the prePIP1 family harbor a conserved Arg or His residue at each side of the SGPS-motif . We found that exogenous application of synthetic PIP1 peptide corresponding to the conserved SGPS-motif successfully mimicked the phenotypes of A . thaliana plants transiently or constitutively expressing prePIP1 . This indicates that PIP1 is a biologically active form derived from prePIP1 , and shares part or all of the sequence with the genuine mature peptide cleaved from the precursor . However , considering that PIP1 peptide was saturated at micromolar concentration in root growth inhibition assays , we cannot exclude the presence of a more active peptide . Further , a mass spectrometry analysis is needed to confirm the cleavage site and to identify the mature peptides cleaved from prePIPs precursors . Proline hydroxylation is common in SGP-rich peptides such as CLV3p and CEP1 [33] , [34] . PIP family members harbor two conserved proline residues . A comparison of the root growth inhibitory effect of proline hydroxylated and non-hydroxylated forms of PIP1 revealed that hydroxylation enhances the biological activity of the peptide . In contrast , unmodified CLV3 and hydroxylated CLV3 peptides had similar activities in root growth inhibition [32] , [33] . This suggests that proline hydroxylation differentially affects the biological activities of PIP1 and CLV3 . It is currently not clear whether proline hydroxylation of PIP1 affects its affinity for the receptor or its stability . We found that the PIP1 and PIP2 peptides activate similar immune responses as flg22 and PEP1 , including expression of marker genes , ROS production , callose deposition and MAPK activation . The possibility that this result was caused by contamination with flg22 and/or PEP1 can be excluded for several reasons . First , independently synthesized PIP peptides exhibited the same activity; second , IDL2p and CEP1 , two peptides with a similar sequence structure to PIPs that were synthesized together with PIPs , failed to activate immune responses; third , a fls2 loss-of-function mutant that is insensitive to flg22 still responded to PIPs; and fourth , PIP1 and PEP1 differed functionally from each another . A reverse genetics screen identified the class XI LRR-RLK RLK7 as the responsible for PIP1- and PIP2-triggered responses . RLK7-PIP1 binding data implicate that RLK7 acts as the PIP1 receptor . However , the flg22 receptor FLS2 which was previously proposed to perceive CLV3p and Ax21 [23] , [47] , [48] , failed to recognize PIP1 since fls2 mutants were still responsive to PIP1-induced up-regulation of FRK1 . Although RLK7 was required for the PIP1-induced enhancement of host resistance against Pst DC3000 , loss-of-function rlk7 mutants showed no reduction in the level of resistance in the absence of PIP1 treatment . This is reminiscent of the finding that the pepr1/pepr2 double mutant is not affected in the level of resistance against Pst DC3000 [16] . The virulence of Pst DC3000 relies heavily on secreted effector proteins which can suppress host immunity by blocking various signaling pathways [49] . The resistance conferred by the PIP-RLK7 signaling pathway may thus be severely disrupted by pathogen effectors . Moreover , the expression pattern of prePIP1 suggests that PIP1-RLK7 resistance is perhaps more specific to pathogens infecting through the hydathodes or proliferating in the vascular tissue . This idea is consistent with the high host resistance conferred by prePIP1 or prePIP2 over-expression against the fungus Foc 699 , a soil-borne pathogen that colonizes the root vascular tissue . PIP1 activates an almost identical set of signaling events as flg22 and PEP1 , suggesting that the three pathways likely share a number of components . BAK1 regulates several of the immune signaling pathways triggered by LRR-RLK type immune receptors , including FLS2 and PEPR1 [7] , [8] , [39] . We found that PIP1-RLK7 mediated responses are less pronounced in bak1-4 mutants , suggesting that BAK1 contributes to PIP1-RLK7 signaling . Previous studies suggested that BAK1 and BAK1-LIKE1 ( BKK1 ) function in parallel in FLS2- and PEPR1-activated immune signaling , since the bak1 mutant is only partially insensitive to flg22 and PEP1 while the bak1/bkk1 double mutant is completely insensitive [7] , [39] , [43] . We noted that the bak1-4 mutant retained some sensitivity to PIP1 , implying some degree of redundancy between BAK1 and BKK1 . However , both flg22- and PIP1-induced ROS production and root growth inhibition were attenuated in the bak1-4 mutant , whereas only ROS production was affected upon induction with PEP1 . This suggests possible differences in the requirement for BAK1 between flg22 , PEP1 and PIP1 responses . BIK1 , another important regulator of the FLS2 and PEPR1 signaling pathways , is rapidly phosphorylated when flagellin binds to FLS2 [9] . BIK1 phosphorylation can also be induced with PEPR1 or PEPR2 in the presence of ET or PEP1 [17] . This is consistent with the results from our root growth inhibition assay and previous ET-induced triple response analysis . No direct protein-protein interaction between RLK7 and BIK1 could be detected by yeast two-hybrid analysis , and no parallels were found between PIP1-RLK7 and PEP-PEPR1 in the context of the ET response . Neither did the rlk7 mutants show reduced sensitivity to ACC , nor was PIP1-induced root growth inhibition attenuated in the bik1 mutant . BIK1 is a member of class VII RLCKs , which have been suggested to integrate immune signaling in A . thaliana from cell-surface-localized receptors [9] , [10] . Thus it is possible that other members of class VII RLCKs mediate RLK7 signaling and are responsible for the observed differences in signaling outputs between RLK7 and PEPR1 . ProPEP1 family members lack a classical signal peptide , and therefore the mechanism underlying PEP release is unclear . Since expression of proPEP1 is up-regulated by wounding and treatment with the wound signal MeJA , it was suggested that release of PEP1 from plant cells may be the result of cell injury caused by pathogen attack or wounding [3] . Consistent with this , PEPR signaling was recently shown to operate predominantly at local pathogen challenged sites , though systemic immunity can be activated by treatment with PEP1 [50] . In contrast , PIP1 is secreted into the extracellular spaces through a cell-autonomous secretory pathway and massive expression of prePIP1 is detected in vascular tissues , suggesting that PIP1 is likely to act as a mobile signal involved in systemic immune activation . Activation of immunity by endogenous signals is a common strategy exploited by animals and plants to amplify immune responses after perceiving a limited number of invading pathogens [51] . In animals , many endogenous peptides such as interleukins which are generated upon PAMP recognition , were confirmed to function in inflammation [52] . In plants , PEP1 was suggested to act as a PTI amplifier because ( 1 ) PAMP treatment increases transcription of proPEP1 , ( 2 ) PEP1 and PAMPs activate similar immune responses , and ( 3 ) PEP1 receptors are required for full activation of PTI signaling and resistance against bacterial infection [16] , [18] , [53] . In this study , prePIP1 and RLK7 were induced by flg22 , and flg22-triggered immunity was impaired in rlk7 mutants . These findings imply that PIP1-RLK7 and PEP1-PEPR1 have similar functions in FLS2 signal amplification . PIP1 and PEP1 , respectively , induce their corresponding precursor and receptor genes showing that self-amplification mechanisms act in both signaling pathways . Importantly , PIP1 and PEP1 also induce the expression of each other's precursor and receptor genes . Further , the level of PEP1 responses was decreased in rlk7 mutants . These demonstrate that the two endogenous peptide signaling pathways are interdependent and cooperate to amplify the immune response . We propose a working model ( Figure 11 ) in which FLS2 signaling is initially primed by the perception of flg22 , followed by upregulation of the host peptide elicitors PIP1 and PEP1 and their respective receptors PEPR1 and RLK7 . Once PIP1 and PEP1 are released and processed in the apoplast , they initiate the immune response and also increase expression of prePIP1 , RLK7 , proPEP1 , PEPR1 and FLS2 , leading to an amplification of the immune responses via the combined effect of FLS2 , PEPR1 and RLK7 .
A . thaliana were grown in potting mix or on 1/2 MS medium ( containing 1/2 MS salts , 1% w/v sucrose and 0 . 8% w/v agar , pH 5 . 7 ) in a controlled growth chamber providing a 10 h photoperiod ( 140 µmol•m−2•s−1 light ) at 22°C/20°C day/night and 60% relative humidity . fls2 [54] , rlk7 [55] , hae/hsl2 [35] , ein2-1 [17] , bak1-4 [7] , and bik1 [10] mutants used were described earlier . Verification of homozygous T-DNA insertion mutants was carried out by a PCR assay based on locus-specific primers ( Table S3 ) . Arabidopsis seedlings were germinated on 1/2 MS media , and then transferred to 1/2 MS liquid medium ( 1/2 MS salts , 1% sucrose , pH 5 . 7 ) adding various concentrations of PIP1 or other peptides in a 6-well plate . The length of the seedling roots was measured after 5–7 days . PrePIP1 , prePIP2 , and RLK7 coding sequences were PCR-amplified from A . thaliana genomic DNA using locus-specific primers , and the products were separately inserted into pCAMBIA1300-HA vector downstream of the CaMV 35S promoter to generate pCAMBIA1300-35S::prePIP1-HA , pCAMBIA1300-35S::prePIP2-HA and pCAMBIA1300-35S::RLK7-HA . An ∼2 . 8 kb fragment upstream of the prePIP1 start codon was amplified from A . thaliana genomic DNA and inserted into the pGFPGUSPlus vector [56] to construct prePIP1p::GUS and prePIP1p::GFP . Truncated prePIP1 , prePIP2 and prePIPL5 coding sequences were amplified from A . thaliana genomic DNA using locus-specific primers and inserted into pGEX-6p-1 to generate GST-ΔprePIP1 , GST-ΔprePIP2 , GST-ΔprePIPL5 . The BIK1 coding sequence was amplified from A . thaliana cDNA and inserted into pGBKT7 to generate pGADT7-BIK1 . The sequences encoding the kinase domains of PEPR1 ( residues 827–1123 ) and RLK7 ( residues 671–977 ) were amplified from A . thaliana cDNA and inserted into pGADT7 to generate pGADT7-PEPR1KD and pGADT7-RLK7KD . All the sequences primers are listed in Table S3 . Peptides of purity level 98% were synthesized by Yaguang Biochemical Company ( Shanghai , China ) . Their sequences are given in Table S2 . Transient expression in tobacco leaves was performed as described previously [57] . Agrobacterium tumefaciens strain GV3101 harboring pCAMBIA1300-RLK7-HA , pCAMBIA1300-GFP , pCAMBIA1300-prePIP1-GFP or pCAMBIA1300-CLV3-GFP were grown overnight in YEB medium and transferred to 1/2 MS liquid medium containing 50 µM acetosyringone for 4 h until an OD600 of 0 . 4–0 . 6 had been reached . The culture was then diluted 1∶1 with 10 mM MES ( pH 5 . 6 ) , 10 mM MgCl2 , 150 µM acetosyringone , and pressure-infiltrated into the leaves of 4–5 week old tobacco plants . Transfected leaves were collected after 48–72 h . In-vitro cleavage assays were performed as described previously [58] . In brief , GST-tagged truncated prePIPs ( GST-ΔPIPs ) were expressed in E . coli BL21 ( DE3 ) and purified using glutathione Sepharose ( GE Healthcare ) . The purified proteins were incubated with Arabidopsis protein extracts or BSA ( control ) for 0–2 h at room temperature . The samples were then subjected to SDS-PAGE to determine the protein composition . For the in-vivo cleavage assay , GST-PIP1 ( 1 µg/µL ) or GST ( 1 µg/µL ) was syringe-injected into A . thaliana leaves and incubated for 2 h , then extracellular fluids were extracted and analyzed by SDS-PAGE . GUS staining was performed as described previously [36] . In brief , plant tissues were immersed in staining buffer ( 100 mM sodium phosphate buffer , pH 7 . 0 , 10 mM EDTA , 1 mM potassium ferrocyanide , 1 mM potassium ferricyanide , 1 mM X-Gluc , and 0 . 1% Triton X-100 ) and incubated at 37°C for 2–6 h . Stained samples were cleared in 70% ethanol and observed by the Olympus BX53 microscope . Protoplast transfection and subsequent luciferase reporter assay were performed as described previously [14] . FRK1p-LUC reporter was co-transfected with prePIP1 constructs and UBQ10p-GUS ( internal control ) . After 6 hours' incubation , luciferase activities were tested with a Luciferase Assay kit and a GloMax-20/20 luminometer ( Promega ) . For analysis of FRK1p-LUC induction by exogenous application of peptide elicitors , protoplasts were incubated overnight after transfection with FRK1p-LUC reporter , and then were induced with 1 µM peptide for 4 hours before detection of luciferase activity . Total RNA was extracted from plant tissues by the TRIzol reagent ( Invitrogen ) following the manufacturer's protocol . A 2 µL aliquot of the total RNA preparation was subjected to reverse transcription using a RevertAi First Strand cDNA Synthesis kit ( Fermentas ) . The resulting cDNA was amplified using the SYBR Green Mix ( Roche ) and gene-specific primers ( Table S3 ) . AtActin2 was used as the reference sequence . A luminol-based assay was used to quantify ROS in treated leaves [59] . The same amount of 1–2 mm leaf fragments cut from Arabidopsis leaves were incubated in 100 µL water for 12 h , and then 100 µM luminol ( Sigma ) , 10 µg/mL horseradish peroxidase ( Sigma ) and 1 µM peptide were added rapidly in turn . The resulting luminescence was measured using a GloMax-20/20 luminometer ( Promega ) at one minute intervals over 15 min . Staining of callose deposits was achieved following methods described previously [36] , [59] . Adult leaves were infiltrated with either water or 1 µM peptide for 8 h , and the roots of 10-day old seedlings were immersed in 1/2MS liquid medium with or without peptides ( 1 µM ) or chitin ( 500 µg/L ) for 18 h . The materials were then fixed in 3∶1 ethanol∶acetic acid for 6 h , changing the fixative solution every 2 h . The samples were rehydrated in 50% ethanol for 2 h , and then thoroughly rinsed in water . Finally the samples were incubated in staining solution ( 150 mM K2HPO4 ( pH 9 . 5 ) , 0 . 01% ( w/v ) aniline blue , Sigma-Aldrich ) for 30 min . Callose was visualized using UV-epifluorescence microscopy . Signal intensities were estimated using Image J software . Ten seedlings were immersed in sterile water overnight . Peptides were then added to a final concentration of 1 µM for 5–15 minutes induction . After induction , the seedlings were snap-frozen in liquid nitrogen and ground to a fine powder , from which total protein was extracted by suspension in 50 mM HEPES ( pH 6 . 8 ) , 150 mM NaCl , 1% ( w/v ) SDS , 2 mM DTT , 10 mM NaF , 10 mM NaVO3 , 5 mM EDTA , 1× protease inhibitor cocktail ( Roche ) . An anti-phospho p44/p42 MAPK antibody ( Cell Signaling Technology ) was used to detect active MPK6 and MPK3 via immunoblotting . Y-PIP1 peptide was labeled with 125I as described previously [60] . In brief , 2 nmol Y-PIP1 peptide and 600 µCi Na125I ( PerkinElmer ) dissolved in 100 µL sodium phosphate buffer ( 10 mM , pH 7 . 4 ) were added into a glass vial pre-coated with 1 , 3 , 4 , 6-tetrachloro-3α , 6α-diphenylglycouril , and were incubated for 15 min at root temperature . After passing through a Sephadex G25 column ( PD-10 column , GE Healthcare ) , ∼800 µL 125I-Y-PIP1 containing 1 . 7×107 counts per minute ( cpm ) was collected . Plasma membrane fragments were extracted from 200 mg tobacco leaves and re-suspended in binding buffer ( 25 mM MES , pH 6 . 0 , 3 mM MgCl2 , 10 mM NaCl , 2 mM dithiothreitol and protease inhibitor cocktail ( Roche ) ) with a final concentration of 2 µg/µL total protein . The plasma membrane ( 100 µL ) was incubated with 2 µL 125I-Y-PIP1 ( ∼100 fmol ) in the presence or absence of 10 µM unlabelled PIP1 for 15 min at 4°C , then were collected by a vacuum filtration system through glass fibre filters ( Millipore , 2 . 5-cm diameter ) . After washed with cold washing buffer ( binding buffer supplemented with 1% BSA , 1% bactotrypton , 1% bactopepton ) , the binding was determined by γ-counting . Plasma membrane proteins were extracted from the Arabidopsis leaves of rlk7 mutant and rlk7/35S::RLK7-HA with an extraction buffer ( 25 mM MES/KOH ( pH 6 . 0 ) , 3 mM MgCl2 , 10 mM NaCl , 0 . 5% SDS and 1× protein inhibitor cocktail ( Roche ) ) , then were diluted ten folds with a binding buffer ( 25 mM MES/KOH ( pH 6 . 0 ) , 3 mM MgCl2 , 10 mM NaCl and 1× protein inhibitor cocktail ( Roche ) ) . Biotinylated PIP1 ( 1 µg ) was coupled to 20 µL streptavidin beads ( Pierce ) for 1 h at 4°C . After three rinses in 500 µL binding buffer , the beads were incubated with 200 µL of the prepared plasma membrane proteins in the presence or absence of 100× excess of unlabelled PIP1 or IDA for 2 h at 4°C . After rinsed three times in 500 µL binding buffer , the beads were boiled for 5 minutes in 50 µL 1× Laemilli buffer . The RLK7-HA was detected with an anti-HA monoclonal antibody ( Qiagen ) . Chemical cross-linking of PIP1-biotin to RLK7 was displayed as described previously . PIP1-biotin ( 1 µM ) was incubated with the total protein ( 50 µg ) extracted from rlk7 or 355S::RLK7-HA plants in the presence or absence of excess ( 50 µM ) unlabeled PIP1 for 30 min at 4°C . After adding 1/10 volume of 25 mM EGS ( Pierce ) , the mixture was incubated for another 30 minutes at room temperature before the reaction was terminated by the addition of 1 µL Tris-HCl buffer ( 1 M , pH 7 . 5 ) . Proteins in samples were separated by SDS-PAGE and detected with anti-biotin antibody ( Cell Signaling Technology ) . Interactions between BIK1 and the kinase domain of PEPR1 ( residues 827–1123 ) or RLK7 ( 671–977 ) were tested using the GAL4 yeast two-hybrid system ( Clontech ) . In brief , the pGADT7-PEPR1KD or pGADT7-RLK7KD plasmid was co-transfected with pGBKT7-BIK1 into Saccharomyces cerevisiae strain AH109 . The transformed yeast cells were spotted on a synthetic dropout ( SD ) medium ( Difco Yeast Nitrogen Base ) lacking tryptophan , leucine , and histidine ( SD-Y−-L−-H− ) but supplementing with 3 mM 3-amino-1 , 2 , 4-triazole ( 3-AT , Sigma ) to detect the His reporter activity . Transformants were also detected on the basis of lacZ reporter activity with 50 µg/mL X-gal dissolved in 25 mM phosphate buffer . Pst DC3000 inoculation assay was performed as described previously [61] . The bacterial suspension ( 2×105 colony-forming units ( cfu ) /mL ) with or without 1 µM peptide was syringe infiltrated into leaves of 5-week old A . thaliana plants . Foc 699-GFP strain was obtained by cotransformation of the F . oxysporum f . sp . conglutinans strain 699 with the sGFP coding region driven the Aspergillus nidulans gpdA promoter and the trpC terminator , and the hygromycin resistance cassette , as described previously [62] , [63] . Foc 699-GFP was grown in half strength potato dextrose broth at 28°C for 2 to 3 days . Ten day old seedlings were exposed to a 2 mL volume of a microconidia suspension ( 1×106 spores/mL sterile water ) and incubated for 3–24 h at 22°C . To quantify Foc 699-GFP biomass , genomic DNA was extracted from 30 infected seedlings after rinsing them three times in sterile water , and used as a template for qPCR with GFP-specific primers ( Table S3 ) . The AtActin2 gene was used as the reference sequence . To monitor infection , Arabidopsis seedlings were rinsed three times with sterile water after 6-hour incubation with spore solution , planted into soil , and survival of the plants was assessed after 21 days . Sequence information of genes involved in this article can be found in the Arabidopsis information resource or the Arabidopsis unannotated secreted peptide database under the following accession numbers: At4g28460 ( prePIP1 ) , At4g37290 ( prePIP2 ) , At2g23270 ( prePIP3 ) , At1g49800 ( prePIPL1 ) , At3g06090 ( prePIPL2 ) , At4g37295 ( prePIPL3 ) , At5g43066 ( prePIPL4 ) , ath_mu_ch1_43150top ( prePIPL5 ) , ath_mu_ch5_43674top ( prePIPL6 ) , ath_mu_ch4_17161top ( prePIPL7 ) , ath_mu_ch5_43661top ( prePIPL8 ) , At1g09970 ( RLK7 ) , At5g46330 ( FLS2 ) , At1g73080 ( PEPR1 ) , At1g17750 ( PEPR2 ) , At2g31880 ( SOBIR1 ) , At4g28490 ( HAESA ) , At5g65710 ( HSL2 ) , At5g25930 ( HSL3 ) , At5g64900 ( proPEP1 ) , At4g33430 ( BAK1 ) , At2g39660 ( BIK1 ) , At5g24110 ( WRKY30 ) , At2g38470 ( WRKY33 ) , At4g23810 ( WRKY53 ) , At2g19190 ( FRK1 ) , At2g14610 ( PR1 ) , At5g44420 ( PDF1 . 2 ) , At1g18570 ( MYB51 ) , At5g03280 ( EIN2 ) , At1g68765 ( IDA ) , At5g64667 ( IDL2 ) , At1g47485 ( CEP1 ) .
|
Both animals and plants have evolved mechanisms to trigger innate immunity through perception of exogenous and endogenous molecules . In the model plant Arabidopsis thaliana , endogenous molecules such as the peptide elicitor PEP1 activate the immune response by means of cell surface-located receptors . Here we describe a new gene family in A . thaliana named prePIPs , whose members encode secreted peptide precursors , and show that one of its members , prePIP1 , is secreted into extracellular space and cleaved at the C-terminus . Exogenous application of PIP1 , the synthetic 13-amino acid peptide corresponding to the conserved C-terminal region of prePIP1 , triggered immune responses and led to enhanced pathogen resistance in A . thaliana . We further provide evidence showing that PIP1 signals via the receptor-like kinase 7 ( RLK7 ) and employs both shared and distinct components with the PEP1 signaling pathway . Both PIP1 and PEP1 cooperatively amplify the immune response triggered by flg22 , the active epitope of bacterial flagellin .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology",
"and",
"life",
"sciences",
"plant",
"science"
] |
2014
|
The Secreted Peptide PIP1 Amplifies Immunity through Receptor-Like Kinase 7
|
Cells interacting through an extracellular matrix ( ECM ) exhibit emergent behaviors resulting from collective intercellular interaction . In wound healing and tissue development , characteristic compaction of ECM gel is induced by multiple cells that generate tensions in the ECM fibers and coordinate their actions with other cells . Computational prediction of collective cell-ECM interaction based on first principles is highly complex especially as the number of cells increase . Here , we introduce a computationally-efficient method for predicting nonlinear behaviors of multiple cells interacting mechanically through a 3-D ECM fiber network . The key enabling technique is superposition of single cell computational models to predict multicellular behaviors . While cell-ECM interactions are highly nonlinear , they can be linearized accurately with a unique method , termed Dual-Faceted Linearization . This method recasts the original nonlinear dynamics in an augmented space where the system behaves more linearly . The independent state variables are augmented by combining auxiliary variables that inform nonlinear elements involved in the system . This computational method involves a ) expressing the original nonlinear state equations with two sets of linear dynamic equations b ) reducing the order of the augmented linear system via principal component analysis and c ) superposing individual single cell-ECM dynamics to predict collective behaviors of multiple cells . The method is computationally efficient compared to original nonlinear dynamic simulation and accurate compared to traditional Taylor expansion linearization . Furthermore , we reproduce reported experimental results of multi-cell induced ECM compaction .
Cell-induced compaction of fibrous extracellular matrix ( ECM ) is an important mechanism for numerous processes such as wound healing and tissue development [1–3] . During wound healing , for example , traction forces exerted by fibroblasts and myofibroblasts result in ECM compaction at the site of injury [2 , 3] . In vitro experiments using cell-populated collagen gel reveal global compaction of the matrix as a result of cooperative effect of multiple cells at the boundaries as well as propagation through the bulk [4–6] . Furthermore , matrix densification is observed in the regions around [7] and in-between cells . Here we examine the mechanical aspect of intercellular communication through the ECM and how contractile cells can induce emergent mechanical changes leading to matrix compaction . From a simplified mechanics point of view , compaction results when the traction forces exerted by the contractile cells embedded within the ECM overcome the resistive forces of the ECM structure , including viscoelastic forces and elastic energy forces . As a result the matrix is deformed from its original stress-free state and the elastic modulus increases [4–7] . In reality , the compaction process is far more complex . The ECM forms a network of cross-linked fibers that is highly nonlinear and intricate , but is critical for predicting large compaction and long-range transmission of forces [4] . As a large deformation is induced by contractile cells , the standard linear mechanics model yields substantial errors . The ECM fiber network is anisotropic and causes irreversible deformations as a large compaction takes place . This prominent nonlinearity prohibits use of simple methods for predicting the ECM compaction by a multitude of cells . In addition , cells can internally modulate their state in response to local mechanical stresses within the ECM , which influences cell polarity , contractility , stiffness and strength of focal adhesion’s [8–11] . These cell properties are highly nonlinear and complex . Consideration of these nonlinear physical and physiological properties involved in the cell-ECM mechanics often result in differential equations that are intractably complex due to high-dimensional , nonlinear coupled dynamics . Many in silico modeling approaches in the areas of wound healing and fibrotic disease have helped elucidate and explore the underlying phenomena involved in cell-induced ECM compaction , and have been used to supplement in vitro experiments for fast and inexpensive methods of evaluation . Approaches in previous works include: i ) a hybrid continuum-discrete framework consisting of the macroscopic finite element domain and local microscopic fiber network [12] , ii ) rule-based models with deformable cells and ECM fibers to explain matrix remodeling and durotaxis [13 , 14] , iii ) a discrete fiber model of cell populated fibrous matrix [15] , and iv ) continuum models of ECM gel compaction [7 , 16 , 17] . Even though these works provide many insights , they also simplify the ECM gel compaction mechanism by: a ) 2-D representation of a 3-D system , b ) exclusion of intracellular mechanics such as mechanobiology of actin stress fibers , focal adhesions , and remodeling of cellular and nuclear membranes , and c ) consideration of linear elastic spring model of ECM fibers without including the viscoelastic nature of the fibers . Consequently , these prior models abstract detailed cell-ECM interactions , resulting in limitations to understanding how these interactions enable characteristic gel compaction . In addition , models examining the complex dynamics surrounding cell morphology , contractility , and polarity based on finite element methods do exist for 2 dimensional cases [10] . And , focal adhesion-stress fiber dynamics have been modeled for 2-D PDMS substrates based on non-equilibrium thermodynamics [11] . In the current work , the ECM is modeled as a 3-D cross-linked network of discrete , viscoelastic fibers , and detailed mechanistic cell dynamics , including focal adhesion dynamics , cytoskeleton remodeling , actin motor activity and lamellipodia protrusion , are derived from basic principles . The resultant model is computationally complex , especially for a larger number of cells . The governing differential equations are highly nonlinear , coupled , and of high dimension . Here , we solved this difficulty by introducing a methodology having its disciplinary basis spanned in system dynamics , machine learning , and statistics . It is known that a nonlinear system can behave more linearly when recast in a larger space [18] . In our approach , the original nonlinear dynamics derived from physical and physiological principles are recast in an enlarged state space by augmenting independent state variables with auxiliary variables that inform input-output characteristics of the nonlinear elements involved in the system . Once recast in the augmented space , the nonlinear system can be represented as an augmented set of linear dynamic equations . The linear representation facilitates model reduction using latent variable analysis , which can be shown is difficult to apply to highly nonlinear systems [19–22] . Furthermore , linearization in the augmented space allows for superposition of multiple subsystems . In the current work , collective behaviors of multiple cells are predicted via superposition of single cell subsystems through the linearization in the augmented state space . The proposed methodology is general , and is applicable to a broader class of problems where large-scale , collective behaviors must be predicted while retaining sufficient mechanistic details .
We construct a computational model for predicting cell-mediated gel compaction by multiple ( ncell ) cells having a uniform phenotype and interacting through a surrounding 3-D ECM fiber network . The ECM is modeled as a network of many fibers connected at a large number of nodes ( Ne ≈ 2000 ) , whereas each cell is represented with a mesh structure consisting of multiple nodes ( Nc ≈ 200 ) which forms the cell outer membrane ( see Fig 1A ) . The cell outer membrane deforms and gains traction as the nodes on the membrane bond to the nodes of the surrounding ECM fiber network and form focal adhesions , which occur when bonding molecules ( or integrins ) on the cell membrane bind to ligands on ECM . Consider the i-th outer membrane node of the k-th cell with three dimensional spatial coordinates x i c , k ∈ ℜ 3 × 1 ( See Fig 1B ) . The forces acting on it include the cell’s cortical tension force and elastic energy force ( collectively denoted as F C o r t - E l a s , i c , k ∈ ℜ 3 × 1 ) , focal adhesion force ( denoted as F F A , i c , k ∈ ℜ 3 × 1 ) , lamellipodium force ( F L , i c , k ∈ ℜ 3 × 1 ) , and frictional damping force ( F D a m p , i c , k ∈ ℜ 3 × 1 ) [23 , 24] . Assuming that the mass of the node is negligibly small and the damping force is given by F D a m p , i c , k = - D c d x i c , k / d t , where Dc is damping constant , the equation of motion is given by: F C o r t - E l a s , i c , k + F F A . i c , k + F L , i c , k - D c d x i c , k d t = 0 i = 1 , ⋯ , N c , k = 1 , ⋯ , n c e l l ( 1 ) The generation of lamellipodium force pertains to the polarity of the cell . Namely , lamellipodia extend in a particular direction of the cell determined by the cell’s polarity [23–26] . The cell polarity and the lamellipodium forces can be treated as a cell’s decision or , in the system dynamics terminology , control inputs . Let x c , k = ( x 1 c , k T ⋯ x N c c , k T ) T ∈ ℜ 3 N c × 1 be a vector containing the 3-D coordinates of all the cell membrane nodes . Here the superscript in XT represents the transpose of matrix or vector X . The above equation of motion can be written collectively as: d x c , k d t = W C E c F C o r t - E l a s c , k + W F A c F F A c , k + L c u k k = 1 , ⋯ , n c e l l ( 2 ) where F C o r t - E l a s c , k ∈ ℜ 3 N C × 1 is a vector comprising cortical tension and elastic energy forces for all the cell nodes ( i = 1 , ⋯ , NC ) , F F A c , k ∈ ℜ 3 N C × 1 is a vector of focal adhesion forces at all the cell nodes , u k ∈ ℜ 3 N C × 1 is an input vector containing all the lamellipodium forces ( F L , i c , k ) , and W C E c , W F A c and Lc are constant matrices of consistent dimensions . The equation of motion of the surrounding ECM fiber network can be represented in a similar manner . The forces acting on the j-th node of the fiber network are the elastic energy forces , including both lateral restoring forces and the one associated with bending moments , ( F E l a s , j e ∈ ℜ 3 × 1 ) , focal adhesion forces from the shared attachment with the cell ( F F A , j e ∈ ℜ 3 × 1 ) and damping forces ( F D a m p , j e ∈ ℜ 3 × 1 ) [23–26] . The equation of motion can be written as: F E l a s , j e + F F A , j e - D e d x j e d t = 0 , j = 1 , ⋯ , N e ( 3 ) Let x e = ( x 1 e T ⋯ x N e e T ) T ∈ ℜ 3 N e × 1 be a vector containing the 3-D coordinates of all the ECM nodes . Then Eq 3 can be written as: d x e d t = W E l a s e F E l a s e + W F A e F F A e ( 4 ) The ECM elastic energy force is a nonlinear function of ECM coordinates xe . The cortical tension and elastic energy force of the k-th cell is a nonlinear function of its membrane coordinates xc , k . Here xe and xc , k are independent state variables of the multi-cell ECM system . F E l a s e = F E l a s e ( x e ) , F C o r t - E l a s c , k = F C o r t - E l a s c , k ( x c , k ) k = 1 , ⋯ , n c e l l ( 5 ) The focal adhesion force is modeled as a stochastic binding process between nodes on the cell membrane and those on the ECM . Using Monte Carlo simulations it has been found that focal adhesion forces can be approximated to a nonlinear algebraic function of cell membrane and ECM nodes as well as the biochemical parameters involved in integrin-ligand binding [23 , 24] . F F A c , k = F F A c , k ( x c , k , x e ) , k = 1 , ⋯ , n c e l l ( 6 ) Assuming that no two cells bind to the same ECM node , we can find that the focal adhesion force of the i-th membrane node of the k-th cell attached to the j-th ECM node must satisfy: F F A , i c , k + F F A , j e = 0 ( 7 ) Namely , F F A , i c , k and F F A , j e have the same magnitude with the opposite signs . Therefore , all the focal adhesion forces of the k-th cell can be mapped to the corresponding ECM nodes . Collectively , the focal adhesion forces of all the nodes within the ECM may be written as: F F A e = ∑ k = 1 n c e l l P m a p k F F A c , k ( 8 ) where P m a p k ∈ ℜ 3 N e × 3 N c is a parameter matrix ( consisting of either 0 or -1 elements ) that maps the membrane focal adhesion forces of the individual cells ( F F A c , k , k = 1 , … , n c e l l ) to the corresponding ECM focal adhesion forces ( F F A e ) . The focal adhesion connections between the membrane nodes and ECM nodes change over time as the cell membrane deforms , gains traction and generates lamellipodial protrusions . Therefore , the mapping matrix P m a p k is updated at each time step . Details on the formation and structure of P m a p k are given in the Methods Section . The ncell cells interact with each other through the surrounding ECM by generating focal adhesion forces , which propagate through the ECM fiber network and influence the other cells . The resultant collective behavior of the multiple cells is complex due to coupled , nonlinear dynamics . Although the governing equations derived above are rigorous and based on basic principles , they are complex and can become computationally expensive as the number of cells increases . Computational complexity is a key challenge in predicting collective behaviors of multiple cells . The number of state variables for the given system is 3Ne + 3Nc ncell , which is on the order of 7 , 000 for ncell = 2 and 9000 for ncell = 5 . We aim to transform the governing equations into a linear latent variable representation in order to considerably reduce the number of state variables but also facilitate prediction of collective behaviors of the multiple cells through superposition of individual cell dynamics . Model reduction is a challenging problem particularly for highly nonlinear , dynamical systems [21 , 22 , 27–29] , as in the presented problem of collective behaviors of multiple cells within an ECM . If the system is linear or near linear , model reduction is more amenable and simple methods , such as Principal Component Analysis and Partial Least Squares , can reduce dimensionality . Here , we propose a unique linearization method , termed Dual-Faceted ( DF ) Linearization , and then apply a model reduction method to the linearized model . In DF Linearization , we represent the nonlinear dynamical system in an augmented space consisting of independent state variables ( xe and xc , k ) and nonlinear forces ( F E l a s e , F C o r t - E l a s c , k , F F A c , k ) as the additional variables , termed auxiliary variables . Standard linearization , such as Taylor series expansion , is limited in accuracy , which may be valid only in the vicinity of a reference point . In DF Linearization , instead of taking “algebraic” linearization of these nonlinear terms , we consider “dynamic” linearization by representing their dynamic transitions using linear regressions . Before formally introducing DF Linearization , let us consider a simple example that delineates the basic principle of the method . Suppose that the system consists of one spring and a damping element with negligibly small mass , F - D x ˙ = 0 . If the spring is a linear spring , F = kx , there is absolutely no difference between the equation in terms of state variable x , x ˙ = ( k / D ) x , and the one in force F , F ˙ = ( k / D ) F . However , it is not the case if the spring is nonlinear , for example a hard spring: F = ax + bx3 where a > 0 , b > 0 . Representing the differential equation in two variables , one with the state variable x and the other with the auxiliary variable F , provide different equations . d x d t = a D x + b D x 3 ( 9 ) d F d t = 1 D ( a + 3 b ( g ( F ) ) 2 ) F ( 10 ) where x = g ( F ) s the inverse function of F = ax + bx3 . Both equations represent the same nonlinear system , yet the expressions are different , hence Dual-Faceted representations . Linearizing these differential equations lead to two linear differential equations viewed from the augmented space . Note that Eq 9 can be represented as a linear equation by using both state and auxiliary variables: d x d t = W F F ( 11 ) where WF = 1/D . The augmented state Eq 10 can be approximated to a linear regression: d F d t ≃ S x x + S F F ( 12 ) where Sx , SF are regression parameters . The expression given by Eq 12 differs from the one based on the first order Taylor expansion ( or “algebraic” linearization ) which yields: F ( x ) ≃ F ( x ¯ ) + d F d x Δ x ( 13 ) Furthermore , if we evaluate the derivative J ( x ) ≡dF/dx at a particular point , J ¯ ( x ¯ ) , and then use Eq 13 to express the augmented state equation , it reduces to F ˙ = J ¯ x ˙ . This implies that F ˙ and x ˙ are proportional . Using an “algebraic” linearization yields a differential equation representing the transition of F that is collinear to the one representing the transition of x , and thereby an auxiliary state equation would not provide any new information . Conversely , the regression model in Eq 12 provides us with a diverse view of the original nonlinear system , thus providing a richer representation of the nonlinearity than the standard first order Taylor expansion . Applying the above principle of Dual Faceted Linearization to our problem , we note that the original state equations governing the transition of the independent state variables , 2 and 4 , are linear in the augmented state space . All we need is to obtain the transition of the auxiliary variables . Let the regression of the dynamic transition of auxiliary variable F E l a s e , be expressed as: d F E l a s e d t ≈ R x e x e + R F E l a s e F E l a s e ( 14 ) where R x e , R F E l a s e ∈ ℜ 3 N e × 3 N e are parameter matrices . If an “algebraic” linearization using the Jacobian J ¯ = ∂ F E l a s e / ∂ x e | x ¯ e was utilized , the above equation would be: d F E l a s e / d t = J ¯ · d x e / d t . This state transition equation through “algebraic” linearization is equivalent to one of the original independent state Eq 4 because d F E l a s e / d t and d x e / d t are collinear within this formulation which renders it redundant . In contrast , the state transition equation presented in Eq 14 is not collinear , providing a diverse facet of the nonlinear system . Similarly , for the auxiliary variables F C o r t - E l a s c , k , F F A c , k , let the regression equations be written as: d F C o r t - E l a s c , k d t ≃ Q x c x c , k + Q F C E c F C o r t - E l a s c , k + Q u u k ( 15 ) d F F A c , k d t ≃ H x c x c , k + H x e x e + H F F A c F F A c , k + H u u k ( 16 ) where Q ⋆ ⋆ , H ⋆ ⋆ ( ⋆ -corresponding subscript and superscript ) are parameter matrices with consistent dimensions . The high-dimensional parameter matrices ( R ⋆ ⋆ , Q ⋆ ⋆ , H ⋆ ⋆ ) do not need to be determined explicitly as discussed in the subsequent sections . DF Linearization represents a nonlinear dynamical system with two sets of differential equations . One set is the original state equations governing the transition of the independent state variables and the other set is the regression of the dynamics of auxiliary variables . The original state Eqs 2 and 4 , are apparently linear in terms of the auxiliary variables and inputs . In these equations , all the forces acting on each node sum to zero . These are linear expressions when the nonlinear forces are treated as auxiliary variables . In addition , the auxiliary state transitions ( Eqs 14–16 ) are given by linear regressions in the augmented space . Therefore , both differential equations are linear . The two linear differential equations represent different ( or dual ) facets of the original nonlinear system viewed from the augmented space , thus providing a richer representation of the nonlinearity . Now that the original nonlinear system has been represented as a linear dynamical system in the augmented space , we can apply a latent variable modeling method to reduce model order . Represented in the augmented space , the differential equations may contain similar modes , or some variables are close to collinear . These similar modes and collinear variables can be eliminated by model order reduction methods . Let ζ c , k be the augmented variable vector containing membrane node coordinates and forces of the k-th cell . ζ c , k = ( x c , k F C o r t - E l a s c , k F F A c , k ) ∈ ℜ 9 N c × 1 ( 17 ) Here uk ( the cell’s lamellipodial force ) is treated as input variables that are excluded from the augmented state space . Similarly , let ζe be the augmented variable vector containing ECM node coordinates and forces: ζ e = ( x e F E l a s e ) ∈ ℜ 6 N e × 1 ( 18 ) Focal adhesion forces F F A e are determined by the individual cells by Eq 8 and , thereby , excluded from the augmented space of the ECM . We apply latent space analysis to vectors ζc , k and ζe , respectively . First we generate data by using Eqs 2 and 4–7 . Computation of the nonlinear state equations is amenable for a single cell interacting with ECM . A data set can be created by simulating those nonlinear equations by placing a single cell at diverse locations , i . e . repeating the simulation with different initial conditions . Let C ζ ζ c , C ζ ζ e be the covariance matrices of simulation data sets of augmented states ζc , k and ζe , respectively . Each covariance matrix contains both independent state and auxiliary variables , where the latter is nonlinear functions of the former . If auxuliary variables were linear functions of the state variables , then the rank of the covariance matrix would be equal to the number of independent state variables . However due to the nonlinearity , the rank is higher . Details on the formation of C ζ ζ c , C ζ ζ e are given in the Methods Section . The covariance analysis also reveals that the system represented in the augmented space contains many components that may be negligibly small . Using Principal Component Analysis , the original data of ζc , k and ζe can be represented with latent variables of truncated dimension mc ≪ 9Nc and me ≪ 6Ne , respectively [29]: ζ c , k = ( V x c V F C E c V F F A c ) ︸ V c z c , k , ζ e = ( V x e V F E l a s e ) ︸ V e z e ( 19 ) where matrices V c ∈ ℜ 9 N c × m and V e ∈ ℜ 6 N e × m are orthogonal matrices comprising the eigenvectors of the covariance matrices , and z c , k = ( V x c V F C E c V F F A c ) T ( x c , k F C o r t - E l a s c , k F F A c , k ) ∈ ℜ m c × 1 z e = ( V x e V F E l a s e ) T ( x e F E l a s e ) ∈ ℜ m e × 1 ( 20 ) Differentiating the latent variable state vector zc , k and substituting Eqs 2 , 15 , 16 and 19 yields: d z c , k d t = V x c T d x c , k d t + V F C E c T d F C o r t - E l a s c , k d t + V F F A c T d F F A c , k d t = A z c , k + B u k + C z e , k = 1 , ⋯ , n c e l l ( 21 ) where: A = V x c T ( W C E c V F C E c + W F A c V F F A c ) + V F C E c T ( Q x c V x c + Q F C E c V F C E c ) + V F F A c T ( H x c V x c + H F F A c V F F A c ) B = V x c T L c + V F C E c T Q u + V F F A c T H u C = V F F A c T H x e V x e ( 22 ) Eq 21 provides the latent variable state equation of the k-th cell interacting with the ECM . Given the latent variable state of ECM ze and the input uk reflecting the cell’s decision , the transition of the cell’s latent variable state is determined locally without directly including the states of the other cells . Cells interact indirectly through the strain field created by other cells over the ECM fiber network . The ECM dynamics can be represented in the latent variable space spanned by Ve . Differentiating the latent variable state vector and substituting Eqs 4 , 14 and 19 yield: d z e d t = V x e T d x e d t + V F E l a s e T d F E l a s e d t = G z e + ∑ k = 1 n c e l l D k z c , k ( 23 ) where: G = V x e T W E l a s e V F E l a s e + V F E l a s e T ( R x e V x e + R F E l a s e V F E l a s e ) D k = V x e T W F A e P m a p k V F F A c ( 24 ) Fig 2 shows numerical examples of the DF linearization and subsequent latent variable transformation in reproducing accurate cell morphologies of the original nonlinear computational model over time . Remarkably , the DF linearized latent variable model can correctly reconstruct the complex cell membrane topology with m = mc + me = 50 + 50 = 100 total latent variables . Fig 2C quantifies the root mean square error and computation time as a function of latent variable model dimension . As can be seen , the computation time for the latent variable cell-ECM system increases with increased latent variables while the root mean square error decreases . Conversely , the standard Taylor expansion linearization is not capable of representing cell morphologies without marked error which is quantified in Fig 2B . We also compare our DF linearization approach to a more sophisticated method for approximation of nonlinear systems termed trajectory piece-wise linear ( TPWL ) [30 , 31] . This method uses collection of ( algebraic ) linearizations of the original nonlinear system about suitably selected states to approximate the nonlinear system . As can be seen from Fig 2B , although the TPWL method yields lower error than the first order taylor expansion , our DF linearized model still leads to significantly lower prediction error . This is because in DF linearization , instead of taking “algebraic” linearization of nonlinear terms , we consider “dynamic” linearization by representing their dynamic transitions using linear regressions . These results demonstrate the effectiveness of DF linearization and model reduction in reconstructing simulations from a high dimensional complex nonlinear model . This latent space model provides not only a low-dimensional structure for efficient computation , but also contains natural insights into the interactions among the multiple cells . Fig 3A shows the dynamic interactions in block diagram form based on Eqs 21 and 23 . The ECM changes its latent variable state ze with the autoregressive feedback through matrix G as well as with the forward path that collects the latent variable states of all the individual cells , as shown by the summing junction Σ . Each single cell changes its latent variable state zc , k ( k = 1 , … , ncell ) with autoregressive feedback through A as well as with two forward paths . The first path ( fed through C ) represents global feedback from the ECM ( ze ) . The second path ( fed through B ) represents the updated lamellipodial forces uk determined from ECM state ze . The lamellipodial forces can be thought of as the individual cell’s decision based on its position and updated ECM properties as explained more in the following section . The actions taken by all the cells are integrated into the global ECM state transition , which is fed back to the individual cells . Therefore , each cell is connected to other cells through the global feedback of the ECM latent variable state ze . Fig 3A manifests the control-theoretic interpretation of multiple cells interacting through ECM . Since the system is represented in a lower dimensional space , the high dimensional regression coefficient matrices ( R ⋆ ⋆ , Q ⋆ ⋆ , H ⋆ ⋆ ) are not computed explicitly . Instead , the lower dimension coefficient matrices A , B , C , G are computed from numerical simulation data that can be transformed into latent variable space . Details are given in the Methods Section . As discussed previously , input vector uk pertains to the lamellipodial forces generated at each membrane node within the leading edge of the cell . The cell continuously updates its lamellipodial protrusions depending on the orientation of the leading edge as the cell’s polarization ( or polarity ) changes . The polarity of a cell is important to determine the orientation of the leading edge and is influenced by the direction of local maximum stiffness in the ECM [23–26] . Here we aim to extend the dynamics model of cell polarity developed in [23–25] to predict the formation of lamellipodia . Let d P o l k ∈ ℜ 3 × 1 be a 3-dimensional unit vector indicating the direction of polarity in the k-th cell and d M a x - S t i f f e , k ∈ ℜ 3 × 1 be a 3-dimensional unit vector pointing in the direction of the maximum stiffness of ECM in the vicinity of the k-th cell’s current location . According to [25] , the cell polarity rotates dynamically in response to ECM’s local stiffness in such a way that the polarity vector may align with the direction of the maximum stiffness: d d P o l k d t = κ d P o l k × ( d M a x - S t i f f e , k × d P o l k ) ( 25 ) where × indicates vector product , and κ is a scalar parameter . Fig 3B illustrates this relationship . The polarity vector d P o l k tends to align with the maximum stiffness direction , d M a x - S t i f f e , k . The leading edge of the cell is indicated by a right circular cone with apex angle 2 α L k having its centerline aligned with the polarity direction . The membrane nodes of the k-th cell within the cone have nonzero lamellipodial forces ( F L , i c , k ≠ 0 ) . Membrane nodes outside this cone have zero lamellipodial forces ( F L , i c , k = 0 ) . The direction of maximum ECM stiffness d M a x - S t i f f e , k depends on the stress field within the ECM , which pertains to the latent state vector of ECM ze . The details are given in Appendix C in S1 Text . Compaction of the ECM by the collective efforts of multiple cells is numerically analyzed based on the model reduction and superposition of the nonlinear cell-ECM dynamics via DF Linearization . We first consider the case where two cells placed 30 μm apart are embedded in a 3-D cylindrical ECM that measures 40 μm in diameter and 100 μm in length as seen in Fig 4A . The boundary conditions of the ECM fiber network are set such that the two flat planes on sides are fixed to space ( constrained ) , while the curved surface surrounding the ECM is kept free ( unconstrained ) . The volume of the cylindrical ECM shrinks over time from its initial unstressed state as the cells interact with the surrounding ECM . To quantify the spatiotemporal compaction process , the original ECM cylinder is segmented into 10 slices of 10 μm thickness along its longitudinal axis as shown in Fig 4B . The volumetric changes to the individual slices are plotted in Fig 4C . The prediction of decreased cell volume by the latent variable superposition model ( blue ) agrees well with the ground-truth , full-scale nonlinear simulation results ( green ) . This is further verified by the corresponding cross-sectional images of the 2-cell cylindrical ECM simulations in Fig 4C . The polarity directions of both cells ( shown by red arrows initially pointing in arbitrary directions ) shift to point inward , indicating that larger stresses are detected in the area between the cells . A video of the simulation comparison is shown in S1 Video . The proposed model is able to reproduce collective behaviors of multiple cells causing the characteristic compaction of ECM gel , which is not observed for single isolated cell models . This is further verified in Fig 4D which compares the ECM compaction results between single cell and two cell models . As can be seen , the single cell model predicts more localized shrinkage of the ECM volume whereas the two cell model shows more global shrinkage extended to within the region between the cells . A video of the simulation comparison is shown in S2 Video . Fig 4D suggests the presence of more than one cell is necessary for the pronounced ECM compaction leading to emergent changes within the ECM . However , the emergence of pronounced compaction entails not only plurality of cells but proper cell spacing . Fig 5A shows that , as the spacing between cells increases , compaction is less pronounced between them , indicating decreased interaction and integration of cell induced propagated forces . This is summarized in Fig 5B which quantifies the average ECM elastic force in-between cells against cell spacing . From Fig 5B , we see that the average ECM elastic force in-between cells spaced at 100 μm is an order of magnitude less than that of the cells spaced at 30 μm . A video of this simulation is shown in S3 Video . The above computational results shown in Figs 4 and 5 verify that the proposed method can capture collective behaviors of multiple cells . The verification was made by comparing the reduced-order superposition model using DF Linearization against the full-scale , nonlinear model . To further verify the capability of the reduced-order superposition model , a comparison is also made against in vitro experimental data of ECM compaction by a larger number of cells . As shown in Fig 5C and 5D , the computational model successfully reproduces the in vitro experiment conducted by Fernandez , et al [5] , in which a heterogeneous planar distribution of MC3T3-E1 osteoblasts were plated in 3-D rectangular prism collagen gel of 50 μm height , 100μm width and 250 μm length . The boundary conditions of ECM in the computational model were set to be consistent with experimental conditions in [5] . The multi-cell latent variable simulation is able to predict the characteristics of the ECM over time . Whereas the group of 5 cells at the left edge exhibit anisotropic contraction of the ECM at the boundary , the single isolated cell at the right edge barely contracts the gel . Fig 5C compares the isolated cell to the group of cells in terms of maximum edge displacement . The isolated cell’s ECM edge displacement is so small that five times its displacement is still substantially lower than the displacement of the group of cells . A video of this simulation is shown in the S4 Video . The presented method for predicting collective behaviors of cell-mediated ECM gel compaction is scalable . Since the individual cell-ECM interactions are local computations , as given by Eq 21 , the computational complexity does not increase exponentially , although the number of cells increases .
The nonlinear state Eqs 2 and 4–6 were computed with custom C-code based on references [23 , 24] . The computation of a single isolated cell embedded in the cylindrical ECM took approximately 24 hours for a single simulation of physical time T ≈ 3600 seconds with sampling interval of 1 second . The simulation was repeated for N ≈ 10 times at different initial cell locations each time . The simulations with a single isolated cell embedded in the large rectangular ECM for reproducing the experimental result were run for approximately 5 days ( 120 hours ) . The physical time of simulation was T ≈ 3600 . The simulation was repeated for N ≈ 10 for various initial locations of a cell . For each simulation , the total number of sample points for all the variables was over 5 , 000 , 000 . The number of sample points was 5 × 107 . The computation was performed on Intel Xeon CPU E5-2687W @ 3 . 10 GHz ( 2 processors ) with 32 logical cores . More details on the formulation of the full-scale nonlinear state equations are summarized in Appendix A in S1 Text . We create the covariance matrix using simulated training data . The training data consists of 3 , 600 time points of both state and auxiliary variables of a cell embedded in an ECM environment . The simulation is repeated N = 10 times , each time with the cell embedded in distinct locations within the ECM . Then the data covariance matrices may be formed: C ζ ζ c = 1 K · N · T ∑ k = 1 K ∑ n = 1 N ∑ t = 1 T ζ ˜ c , k , n ( t ) ζ ˜ c , k , n ( t ) T C ζ ζ e = 1 N · T ∑ n = 1 N ∑ t = 1 T ζ ˜ e , n ( t ) ζ ˜ e , n ( t ) T ( 26 ) where ζ ˜ c , n , k ( t ) represents the mean centered t-th time sample ( of augmented variable vector ζc , k ) for the k-th cell ( here K = 1 or 2 ) embedded within for the ECM at the n-th simulation , and ζ ˜ e , n ( t ) represents the mean centered t-th time sample of the augmented variable vector ζe in the n-th simulation . By performing eigen-decomposition on the covariance matrices we obtain the orthogonal matrix Vc , Ve comprised the eigenvectors of the data covariance matrix: C ζ ζ c ≈ V c Λ c V c T ∈ ℜ m c × m c C ζ ζ e ≈ V e Λ e V e T ∈ ℜ m e × m e ( 27 ) where Λc , Λe are diagonal matrices containing the largest mc and me eigenvalues of the covariance matrices , respectively . It is important to check whether the covariance matrices contain sufficiently rich data , and their first mc and me components are sufficient to capture the cell-ECM dynamics at any cell location within the ECM . Standard techniques can be applied to validate the data and truncation of components [29] . With these , the ECM dynamics of a cell embedded within the ECM at an arbitrary location will be well represented in linear latent variable space which is critical to the success of the method . Covariance matrix calculation were conducted by using Matlab . The following outlines the steps to compute coefficient matrices A , B , C , G by Eqs 21 and 23: Parameter matrices A ∈ ℜ m c × m c , B ∈ ℜ m c × 3 N c , C ∈ ℜ m c × m e , G ∈ ℜ m e × m e are substantially lower in dimension than the regression coefficient matrices R * * , Q * * , H * * given in Eqs 14 , 15 and 16 . Therefore , fewer data points allow us to determine these parameter matrices in the latent variable space . It should be noted that matrices Dk’s are of high dimension , but are not computed with regression since they consist of known matrices as shown in Eq 24 . Matlab was used for estimation of parameter matrices and subsequent computations of the latent variable model . 3-D visualization of simulation data was conducted using Tecplot 360 . When a focal adhesion is formed between the i-th node of the k-th cell and the j-th node of ECM , the two focal adhesion forces sum to zero , as described previously ( F F A , i c , k + F F A , j e = 0 where F F A , i c , k , F F A , j e ∈ ℜ 3 × 1 ) . Representing this relationship in terms of the collective focal adhesion force vectors , F F A c , k ∈ ℜ 3 N c × 1 and F F A e ∈ ℜ 3 N e × 1 , requires a matrix P m a p k ∈ ℜ 3 N e × 3 N c . Let F F A e , k be the forces acting on the ECM nodes caused by focal adhesion between the k-th cell and the ECM nodes . This can be written as i ↓ F F A e , k = ( ⋮ F F A , j e , k ⋮ ⋮ ) = j → ( ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ - I 3 × 3 ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ) ︸ P m a p k ( ⋮ ⋮ F F A , i c , k ⋮ ) ( 28 ) where I3×3 is the 3-dimensional identity matrix . Obtaining this mapping matrix P m a p k for all the cells , the complete focal adhesion forces in the ECM can be expressed in relation to the cell’s focal adhesion forces . F F A e = ∑ k = 1 K F F A e , k = ∑ k = 1 K P m a p k F F A c , k ( 29 ) As previously mentioned , the focal adhesion connections between the cell membrane nodes and ECM nodes can vary over time as the cell membrane deforms , gains traction , and generates lamellipodial protrusions . Therefore , P m a p k is updated to reflect the new focal adhesion attachments and detachments at each time step . The original nonlinear computational model has developed a functional relationship between the focal adhesion force , number of integrins , and distance between the membrane and ECM node ( see details supplementary materials Appendix A in S1 Text ) . In the presented framework , the change in focal adhesion attachments can be derived from simulated training of the nonlinear computational model .
The collective ECM compaction by multiple cells is predicted through superposition of individual cells’ contributions in latent variable space . This is made possible by DF Linearization , latent variable transformation and subsequent superposition of single-cell models to predict the collective behavior among multiple cells . As shown in Fig 3A , the DF linearization has two-order-of-magnitude higher accuracy than the first-order Taylor expansion , and can approximate the original full scale model with a reasonable root-mean-square error . This representation of nonlinear dynamics is markedly different from standard linearization methods . The DF linearization was also compared to the TPWL method . The figure shows an order-of-magnitude better result for the DF linearization compared to the sophisticated technique . Note that the TPWL does not yield a linear model since the state equation includes a product of two state functions . Therefore , superposition as applied with our DF linearization approach cannot be applied . As applied to the analysis of multi-cell ECM compaction , linear augmented equations describing single cell-ECM interactions were derived from DF linearization , and then converted to a reduced-order linear representation by transformation onto a basis of eigenvectors derived from simulated data set . Unlike model reduction of nonlinear dynamical systems , which still remains a challenging problem in the field [19–22] , the model reduction of a linear system through DF Linearization is straightforward . It allows for the evolution of independent and auxiliary states to be described within a lower dimensional linear manifold . The resulting reduced order latent variable model is capable of reproducing nonlinear dynamics , and the linearized structure of individual models facilitated their integration to describe multi-cell behaviors . The prediction of collective behaviors of a group of cells was achieved by superposing contributions of individual cells represented by latent variables zc , k , which evolves based on their own dynamics in response to the global ECM state represented by latent variable ze . The linear representation of the collective multi-cell-ECM interactions manifests the two types of feedback actions by the individual cells . As shown in the block diagram in Fig 3A , the individual cells are exposed to the ECM forces represented by latent variable vector ze in two separate paths . The path through the cell polarity block and matrix B , leading to lamellipodia formation , can be viewed as an “active input” as addressed in [5] . This feedback path includes a cell’s internal decision as to which direction it extends lamellipodia . In contrast , the other feedback path through a gain matrix C does not have a high-level cell decision , but is reactive , playing a “passive role” [5] . These feedback interactions support the prior experimental work [5] . It is interesting to note that ECM compaction begins almost instantaneously , but the magnitude of compaction is rather limited . Once the “active” feedback loop is initiated in , the ECM compacts further , resulting in a large deformation . As the polarity dynamics are rather slow , the second stage ECM compaction does not start immediately . The time scale is determined by the constant κ involved in the polarity dynamics Eq 25 . Using the proposed methodologies , we are able to reproduce intercellular mechanical interactions consistent with published experimental observations . In particular , the global compaction of gel volume via collective cell-contractile activities is characteristically different from local deformations of single isolated cells embedded within the same gel . Through study of emergent behaviors of groups of cells embedded in a 3-D ECM fiber network , we can advance our understanding of intercellular mechanical signaling during tissue formation [1–7] . There are a few limitations to our method , however . While the presented method can predict complex nonlinear behaviors , the method is still a type of approximation . Care must be taken with the validity period . In Fig 4C at the sample time of t = 50 minutes , the latent variable superposition simulation over predicts the volume shrinkage by 12% . With the current mathematical formulation , we have not yet incorporated the degradation of ECM fibers through matrix metalloproteinases . ECM degradation would be necessary to reproduce sustained movement and migration of the cells particularly in 3-D embedded matrices [32] . Since ECM degradation continuously changes the fiber connectivity through ECM remodeling , a methodology to update the node grid structure describing the ECM field would need to be developed . However , ECM degradation may not be necessary for predicting gel compaction since a cluster of cells remains stationary when contracting the surrounding gel [5] . Finally , in the current work , it was assumed that the cell’s polarity mechanism is a dominating internal response to mechanical cues . Cells change their internal state through a complex process of mechanotransduction and intracellular signaling . Incorporating these more complex mechanisms is an exciting avenue for future research . While the method has been developed and demonstrated for multi-cellular interactions with 3D ECM , the basic methodology is applicable to a broad range of systems where nonlinear dynamics of many interacting subsystems are prohibitively complex to compute .
|
Collective behaviors of multiple cells interacting through an ECM are prohibitively complex to predict with a mechanistic computational model due to its highly nonlinear dynamics and high dimensional space . We introduce a methodology where nonlinear dynamics of single cells are superposed to predict collective multi-cellular behaviors through a developed linearization method . We represent nonlinear single cell dynamics with linear state equations by augmenting the independent state variables with a set of auxiliary variables . We then transform the linear augmented state equations to a low-dimensional latent model and superpose the linear latent models of individual cells to predict collective behaviors that emerge from multi-cellular interactions . The method successfully reproduced experimental results of cell-induced ECM compaction .
|
[
"Abstract",
"Introduction",
"Results",
"Methods",
"Discussion"
] |
[
"cell",
"physiology",
"focal",
"adhesions",
"random",
"variables",
"covariance",
"cell",
"polarity",
"simulation",
"and",
"modeling",
"developmental",
"biology",
"systems",
"science",
"mathematics",
"molecular",
"development",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"nonlinear",
"systems",
"adhesion",
"molecules",
"extracellular",
"matrix",
"nonlinear",
"dynamics",
"cell",
"membranes",
"probability",
"theory",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences"
] |
2019
|
Multi-Cell ECM compaction is predictable via superposition of nonlinear cell dynamics linearized in augmented state space
|
The African sleeping sickness parasite Trypanosoma brucei evades the host immune system through antigenic variation of its variant surface glycoprotein ( VSG ) coat . Although the T . brucei genome contains ∼1500 VSGs , only one VSG is expressed at a time from one of about 15 subtelomeric VSG expression sites ( ESs ) . For antigenic variation to work , not only must the vast VSG repertoire be kept silent in a genome that is mainly constitutively transcribed , but the frequency of VSG switching must be strictly controlled . Recently it has become clear that chromatin plays a key role in silencing inactive ESs , thereby ensuring monoallelic expression of VSG . We investigated the role of the linker histone H1 in chromatin organization and ES regulation in T . brucei . T . brucei histone H1 proteins have a different domain structure to H1 proteins in higher eukaryotes . However , we show that they play a key role in the maintenance of higher order chromatin structure in bloodstream form T . brucei as visualised by electron microscopy . In addition , depletion of histone H1 results in chromatin becoming generally more accessible to endonucleases in bloodstream but not in insect form T . brucei . The effect on chromatin following H1 knock-down in bloodstream form T . brucei is particularly evident at transcriptionally silent ES promoters , leading to 6–8 fold derepression of these promoters . T . brucei histone H1 therefore appears to be important for the maintenance of repressed chromatin in bloodstream form T . brucei . In particular H1 plays a role in downregulating silent ESs , arguing that H1-mediated chromatin functions in antigenic variation in T . brucei .
The African trypanosome Trypanosoma brucei is a unicellular parasite causing African sleeping sickness , which is transmitted by tsetse flies in sub-Saharan Africa . As an extracellular parasite of the mammalian bloodstream , T . brucei has evolved a sophisticated strategy to antigenically vary its major surface coat protein , variant surface glycoprotein ( VSG ) [1] , [2] . The T . brucei genome contains a vast repertoire of silent VSG genes and pseudogenes , most of which are located in tandem arrays at subtelomeric locations [3] , [4] . The VSG repertoire varies in both size and composition between different T . brucei strains , with the exact sizes still unclear due to the technical complications of cloning , sequencing and assembling these subtelomeric sequences [5] . However , a conservative estimate proposes that the T . brucei 927 strain contains more than 1500 VSGs , of which only one VSG is expressed at a time [6] , [7] . The active VSG is located in one of about 15 telomeric VSG expression sites ( ES ) . ESs are transcribed by RNA polymerase I ( Pol I ) [8] , [9] , which normally exclusively transcribes ribosomal DNA ( rDNA ) [10] . For antigenic variation to work , it is key that only one VSG is expressed at a time , and the extensive repertoire of VSGs is kept transcriptionally silent . These restrictions need to operate within the context of a T . brucei genome which is primarily organised as very extensive polycistronic transcription units constitutively expressed by Pol II [6] , [11] . Although it is unclear how ESs are controlled , it has recently been shown that chromatin remodeling must play a key role in their regulation [12]–[14] . In eukaryotes DNA is packaged into nucleosomes , whereby ∼146 bp of DNA is wrapped around a histone octamer consisting of two histone H2A/H2B dimers and two histone H3/H4 dimers . A linker histone H1 ( H1 ) typically interacts with both the nucleosome and the linker DNA to stabilize higher order chromatin structure [15] . H1 has been shown to be dispensable in several unicellular eukaryotes including yeast and Tetrahymena [16]–[18] . The exact role of H1 has been surprisingly hard to discern despite its association with heterochromatin and proposed function as a general transcriptional repressor [19]–[23] . Knock-out of H1 in S . cerevisiae , Tetrahymena or mammalian cells affects transcription of a relatively small subset of genes in these different organisms , and does not have a major effect on global transcription [24]–[26] . In addition , yeast cells lacking histone H1 demonstrate genomic instability , most likely due to increased homologous recombination ( HR ) in its absence [27] . The chromatin of T . brucei has several unusual properties . The core histones of T . brucei are divergent compared with those of higher eukaryotes , particularly at the N-termini which can be post-translationally modified [28]–[30] . In addition , T . brucei chromatin has a more open conformation , does not form 30-nm fibres in vitro , and chromosomes fail to condense prior to nuclear mitosis [31] . These characteristic features of T . brucei chromatin are typically influenced by the linker histone H1 in other eukaryotes arguing that T . brucei H1 could play a different role [15] . Histone H1 proteins in T . brucei are distinct from those in other eukaryotes , in that they lack the central globular domain thought to be responsible for interaction with the nucleosome [32] . Instead , they consist of a single domain corresponding to the C-terminal domain of H1 proteins in higher eukaryotes [33] . This C-terminal domain has been shown to be essential for both the DNA binding and chromatin compaction functions of H1 [33]–[36] . Single-domain linker histones are also found in other kinetoplastid species , as well as in Tetrahymena and in eubacteria [21] . Importantly , this type of truncated H1 protein has been shown to affect chromatin structure through a mechanism of DNA compaction which may be mechanistically distinct from higher eukaryotes [16] , [37]–[39] . We have investigated the role of histone H1 in the regulation of antigenic variation in T . brucei , as well as in the maintenance of higher order chromatin states . Depletion of histone H1 , while having a minimal effect on cell growth , causes significant changes in chromatin structure . H1 knock-down resulted in increased sensitivity to endonucleases in bloodstream T . brucei , but not in the procyclic form of the parasite which replicates in the midgut of the insect vector and does not express VSG . In particular , reduced levels of H1 result in the formation of a more open chromatin structure in the vicinity of silent ES promoters in bloodstream T . brucei , which is correlated with an increase in transcription at silent ESs . Knockdown of a number of chromatin proteins in T . brucei results in disruption of VSG ES silencing or VSG switching [40]–[43] . We now show that histone H1 plays a role in regulating ES repression , thereby providing a link between this linker histone and the process of antigenic variation in bloodstream form T . brucei .
The T . brucei 927 genome sequence contains five predicted histone H1-like genes arranged in two clusters on chromosome 11 ( Fig . 1A ) [3] , [44] . The predicted T . brucei H1 proteins are small ( 7–8 kDa ) , very basic proteins ( pI of ∼12 ) with at least one serine or threonine residue at the N-terminus . We raised antibodies using peptides which theoretically should allow recognition of all five T . brucei histone H1 isoforms . Histone H1 proteins have characteristic biochemical properties including the ability to be extracted with perchloric acid [31] , [37] , [45] . We therefore isolated H1 proteins from procyclic form T . brucei using perchloric acid extraction . Several proteins were enriched , and Western blot analysis confirmed that a subset reacts with our H1 antibody ( Fig . 1B ) . We compared the H1 species present in either bloodstream or procyclic form T . brucei using Tricine-SDS-PAGE gels ( Fig . 1C ) . We observed a slightly different histone H1 banding pattern in the two T . brucei life-cycle stages [45] , possibly due to post-translational modifications of different H1 isoforms , as has been found in other organisms [46] . Determining the significance of these different banding patterns is complicated by the fact that our histone H1 antibody may have different affinities for different isoforms and post-translationally modified versions of H1 . However , it is interesting to note that other laboratories have observed similar H1 banding patterns using non-antibody-dependent methods [45] . We next determined the association of T . brucei histone H1 with chromatin in bloodstream form T . brucei , by isolating chromatin containing fractions in increasing concentrations of NaCl ( Fig . 1D ) [42] . As expected , the core histone H3 remained associated with the chromatin fraction in all but the highest salt concentration ( 0 . 8 M NaCl ) , while the RNA binding protein La did not associate with the chromatin fraction [47] . Histone H1 proteins were detected in the chromatin fractions at salt concentrations of up to 0 . 2 M NaCl , indicating association with DNA . However , some proportion of H1 is soluble in the absence of NaCl . This suggests that as expected , T . brucei linker histone H1 has a weaker affinity for chromatin than the core histones [37] , [45] . Similar results were obtained using procyclic form lysates ( Sup . Fig . S1 ) . We determined the subcellular localisation of histone H1 using immunofluorescence microscopy , and identified a strong nuclear signal in both bloodstream and procyclic form T . brucei ( Fig . 2 ) . Notably , H1 appeared to be depleted from the nucleolus . We confirmed that the nucleolus was indeed accessible to antibodies by co-staining with the monoclonal L1C6 antibody which specifically identifies the nucleolus [48] . This observed relative depletion of H1 from the nucleolus is presumably a consequence of the extremely high rates of transcription of the ribosomal DNA ( rDNA ) in this location . We subsequently used chromatin immunoprecipitation ( ChIP ) to investigate the genomic distribution of H1 in T . brucei using either our affinity-purified H1 antibody , or a histone H3 antibody as a positive control ( Fig . 3 ) . We first determined the distribution of H1 on two extensive nontranscribed regions of the T . brucei genome . The 50 bp simple sequence repeats form large nontranscribed arrays flanking all known VSG ESs ( Fig . 3B , 3C ) [49] . H1 is significantly enriched here , as well as on the 177 bp repeats which comprise the bulk of the transcriptionally inactive T . brucei minichromosomes [50] . We subsequently used quantitative PCR ( qPCR ) to determine H1 distribution on different transcription units in bloodstream form T . brucei ( Fig . 3D ) . Statistically significant levels of H1 were present in RNA polymerase II ( Pol II ) transcribed regions , including the actin , γ-tubulin , RNA polymerase I large subunit ( Pol I ) , URA3 , paraflagellar rod protein B ( PFR ) and spliced leader ( SL ) gene loci ( Fig . 3D ) . Significantly less H1 was found at the SL promoter , compared with at the region upstream of the SL gene ( P<0 . 0001 ) . Similarly , H1 was enriched on the non-transcribed rDNA spacer compared with the rDNA promoter or the 18S rRNA gene ( P = 0 . 003 , or P = 0 . 0159 respectively ) . A similar pattern was observed at the procyclin locus , where more H1 was bound upstream of the EP procyclin promoter compared with at the EP procyclin promoter itself ( P = 0 . 006 ) . Similar results were found in procyclic form T . brucei ( Sup . Fig . S2 ) . H1 was found associated with ESs using qPCR primers that detect all ESs . However using primers specific for either the VSG221 or VSGVO2 ES , we showed that H1 is depleted from the active VSG221 ES ( hygromycin and VSG221 genes ) compared with the silent VSGVO2 ES ( neomycin and VSGVO2 ) . Note that the primer pair for VSGVO2 will also detect an additional non-ES located copy of VSGVO2 . This H1 distribution is similar to that of the core histones H2A , H3 and H4 [13] , [14] . H1 is also enriched on the nontranscribed VSG118 gene located at a chromosome internal VSG basic copy array . The distribution of H1 in procyclic form T . brucei is similar , with the exception that H1 does not appear to be depleted from the SL promoter ( Sup . Fig . S2 ) . It is interesting to note that the Pol II transcribed polycistronic arrays are not as depleted of H1 as the active ES , implying that a completely open chromatin structure may not be required for efficient transcription of these regions . To study the function of histone H1 proteins in T . brucei , we performed inducible H1 RNAi using a construct that would be expected to target all five H1 genes . We cloned two fragments corresponding to different regions of the polymorphic H1 gene arrays in tandem into the p2T7-177 vector allowing tetracycline inducible H1 RNAi [51] . In bloodstream form T . brucei , a small but reproducible reduction in growth was observed after the induction of H1 RNAi ( Fig . 4A ) . Western blot analysis revealed that the different H1 proteins were maximally depleted by 24 hours ( Fig . 4B ) . However H1 levels were already reduced in the uninduced ( 0 h ) samples compared with in the parental T . brucei line , presumably as a consequence of leaky transcription of the RNAi construct [52] . In procyclic form T . brucei , cells containing the H1 RNAi construct grew slower than the parental line even in the absence of tetracycline , again presumably due to leaky transcription ( Fig . 4C ) . Western blot analysis of procyclic form T . brucei after the induction of H1 RNAi showed that H1 depletion was maximal at 96 hours ( Fig . 4D ) . However , the additional H1 knockdown observed after induction of H1 RNAi in procyclic form cells had no further effect on growth . These RNAi experiments provide evidence for specificity of our anti-histone H1 antibody , as the putative histone H1 proteins detectable by Western blot indeed decreased after the induction of H1 RNAi ( Fig . 4B and 4D ) . In addition , RNAi-mediated depletion of histone H1 leads to a loss of H1 nuclear staining as observed using immunofluorescence microscopy ( data not shown ) . We next investigated the role of H1 in the maintenance of T . brucei chromatin structure using micrococcal nuclease ( MNase ) . MNase preferentially digests DNA in between nucleosomes , such that an open chromatin structure is more readily cleaved than closed chromatin . Bloodstream form parental T . brucei and cells in which H1 RNAi had been induced were treated with increasing concentrations of MNase ( Fig . 5A ) . DNA was analysed , and characteristic ladders corresponding to mono- , di- , and tri-nucleosomal species were observed ( Fig . 5B ) [14] . Chromatin from cells in which H1 RNAi had been induced for 48 hours was reproducibly more sensitive to MNase digestion compared with that from the parental line ( compare lanes 1 in Fig . 5B ) . This indicates that knockdown of H1 results in DNA becoming more accessible to digestion by MNase , indicating that H1 helps maintain chromatin in a closed state . Interestingly , this increase in the accessibility of chromatin to MNase digestion after H1 knockdown was not observed in procyclic form T . brucei ( Sup . Fig . S4 ) . Possibly , the chromatin structure in procyclic form T . brucei is already more open than that in bloodstream form T . brucei [45] , thereby minimising the impact of H1 depletion . We next investigated the effect of H1 knockdown on chromatin structure at different genomic regions in bloodstream form T . brucei in more detail . We treated permeabilized cells with MNase and fractionated the different nucleosomal species on sucrose gradients ( Fig . 5C , D ) . Again , we observed a dramatic increase in DNA in the mononucleosomal fraction after knockdown of H1 ( Fig . 5D ) . We next pooled DNA fractions according to whether they primarily contained mono- , di- , di-/tri- , tri-/tetra- , or >tetranucleosomes . We determined the presence of different T . brucei genomic regions in the various fractions using qPCR , and expressed the amount of each qPCR target detected in each pool as a percentage of the total amount amplified in all pooled fractions ( Fig . 5E ) . This allowed us to determine changes in the distribution of different genomic regions across the different chromatin types , while correcting for any differences in the amount of material loaded . As expected , sequences in the actively transcribed VSGT3 ES ( blasticidin resistance and VSGT3 genes ) were enriched in the mononucleosomal fraction compared with sequences in the transcriptionally silent VSG221 ES ( puromycin resistance , eGFP or VSG221 genes ) ( 18–22% of total compared with 6–8% , respectively ) ( Fig . 5E ) . This indicates that as expected , active ESs have a relatively open chromatin structure , and are preferentially digested down to mononucleosomes [14] . Interestingly , the Pol II-transcribed actin , γ-tubulin , Pol I large subunit , PFR and URA3 regions , while transcriptionally active , appeared to have a less open chromatin state ( i . e not enriched in the mononucleosomal fraction ) than Pol I transcribed regions . This is similar to as observed for transcriptionally inactive sequences such as VSG118 ( Fig . 5E ) , and is consistent with the distribution of the core histones across these regions as was previously determined [14] . After blocking H1 synthesis for 48 hours , several regions of the T . brucei genome became more accessible to MNase . This effect was most pronounced at the promoter of the silent VSG221 ES ( puromycin resistance and eGFP genes ) ( P = 0 . 0007 and P = 0 . 0043 respectively ) ( Fig . 5E ) . H1 knockdown also resulted in a significant increase in MNase accessibility at the Pol II transcribed actin genes , as well as at the nontranscribed VSG118 in the VSG basic copy array and the regions immediately upstream of the ES and procyclin promoters ( P values<0 . 05 ) . These data indicate that depletion of H1 results in a general opening of chromatin structure , however this effect is particularly clear at the silent ES promoters . Although depletion of H1 caused changes in chromatin structure upstream of the procyclin promoter , we did not detect increases in the levels of procyclin transcript following H1 knockdown ( Sup Fig . S5 ) . Knockdown of histone H1 did not result in obvious changes in the structure or staining intensity of the nucleus as monitored by fluorescence microscopy using DNA staining with DAPI ( data not shown ) . To investigate the effect of histone H1 depletion at the ultrastructural level , we performed transmission electron microscopy ( TEM ) analysis on parental bloodstream form T . brucei , or cells where H1 RNAi had been induced for 48 hours ( Fig . 6 ) . The nucleus of T . brucei has particularly darkly stained areas ( with an electron density comparable to that of the nucleolus ) which presumably correspond to heterochromatin . These dark areas are interspersed with more lightly-stained areas which are likely to contain euchromatin [53] . We find that H1 knockdown results in a dramatic loss of the darkly-stained areas ( black arrows in Fig . 6 ) , resulting in a homogeneously-stained nucleoplasm . This is consistent with our MNase results showing a general increase in chromatin accessibility after H1 knockdown , indicating that depletion of H1 has an effect on the structure of heterochromatin in T . brucei . We next investigated if the changes in chromatin structure observed after H1 depletion had functional consequences for transcription . The bloodstream form T . brucei RYT3 reporter cell line has eGFP immediately downstream of the promoter of the inactive VSG221 ES , allowing ES derepression to be monitored using GFP fluorescence ( Fig . 7A ) [42] . Parental T . brucei RYT3 and two independent RYT3-H1RNAi clones were grown in the presence or absence of tetracycline and monitored by flow cytometry . Both T . brucei RYT3-H1 clones showed 6–8 fold derepression of the silent VSG221 ES after four days of induction of H1 RNAi ( Fig . 7B ) . In procyclic form T . brucei , all ESs are silenced using a mechanism that appears to be different from that used in bloodstream form T . brucei [54]–[59] . Although H1 knockdown in procyclic form T . brucei did not result in a detectable change in accessibility of chromatin to endonucleases , we investigated the effect on ES silencing . Our procyclic form T . brucei 221BsrDsRed cell line has the DsRed reporter gene integrated behind the silent VSG221 ES promoter ( Fig . 7C ) [42] . Induction of H1 RNAi led to a modest but reproducible ∼3-fold derepression of the VSG221 ES promoter in two independent clones , indicating that H1 is required for maximal ES silencing in procyclic form T . brucei ( Fig . 7D ) . In contrast to the observed derepression at ESs , steady state levels of actin and γ-tubulin transcripts remained constant after H1 depletion . We also did not observe a significant increase in precursor transcripts derived from the tubulin array following knock-down of H1 ( data not shown ) . However , since T . brucei relies on post-transcriptional regulation of its mRNAs , it is possible that any putative minor changes in Pol II transcription following H1 knock-down are absorbed by the cell through mRNA degradation/stability pathways . As histone H1 depletion results in transcriptional derepression of silent ESs , we also investigated the consequences of H1 knockdown on VSG switching frequency . We used a similar strategy to [60] , [61] , in which a thymidine kinase ( TK ) gene fused to a drug resistance gene is integrated in the active VSG221 ES between the 70 bp repeat array and the telomeric VSG221 . In addition , eGFP and a puromycin resistance gene are integrated downstream of the VSG221 ES promoter ( Fig . 8A ) . VSG switch events which cause silencing or loss of the TK gene can be selected for using the nucleoside analogue ganciclovir ( GCV ) . GCV resistance can also arise from mutations in the TK gene [62] , however these GCV resistant cells would not have switched their VSG coat as revealed by anti-VSG221 immunofluorescence . The presence of the ES-located single copy marker genes allows the mechanism of VSG switching to be deduced from the genotype and phenotype of a clone after a VSG switch event ( Fig . 8B ) . After a VSG gene conversion ( GC ) , the TK and VSG221 genes are lost , while the switched cell remains GFP positive . After a switch mediated by a telomere exchange , cells are also GFP positive but retain VSG221 . A transcriptional ( in situ ) switch results in GFP negative cells without loss of VSG221 . Finally , a gene conversion can occur which initiates at or near the ES promoter , resulting in the duplication of an entire new ES into the active ES , thereby resulting in its deletion . These switched cells would be GFP negative and would have lost all sequences present in the old VSG221 ES . Alternatively , as previously observed , the same phenotype can result from an ES in situ switch coupled with deletion of the active ES [60] , [61] , [63] , [64] . We first established whether H1 knockdown altered the frequency of GCV resistant clones , which can give an indication of changes in VSG switching frequency ( Fig . 9A ) . Cells were removed from drug selection maintaining transcription of the active VSG221 ES for 48 hours , and H1 RNAi was induced in one culture of TK-expressing cells . Cultures were subsequently serially diluted in 96-well plates in the presence of GCV , and positive wells were scored after 7 days . We observed an average 4-fold increase in the frequency of generation of GCV resistant clones per generation after the induction of a block in H1 synthesis for 48 hours in five independent experiments ( Fig . 9A; P = 0 . 0002 ) . Three cultures of parental cells ( without the H1 RNAi construct ) were also examined in independent experiments . The frequency of generation of GCV resistant clones in the parental cells was ∼3-fold less than in noninduced cells containing the H1 RNAi construct ( P = 0 . 0086 ) , again indicating leaky production of dsRNA in the absence of tetracycline . We next determined the mechanism of VSG switching in clonal cell lines where H1 RNAi had been induced for 48 hours ( n = 7 independent cultures ) or had not been induced ( n = 4 independent cultures ) . Using immunofluorescence microscopy we established if the GCV resistant clones had indeed switched their VSG221 coat , monitored for GFP expression , and determined the presence or absence of the VSG221 gene by PCR . Data showing validation of this strategy to study VSG switching are in Supplementary Figures S6 and S7 . We analysed 33 switched clones derived from uninduced H1 RNAi cells and 64 switched clones from cells where H1 RNAi had been induced for 48 hours . In both cases , we found that the majority of clones had switched by ES gene conversion or an in situ switch plus deletion of the VSG221 ES ( Fig . 9B ) . We also analysed 35 switched clones from four independent cultures derived from an equivalent cell line lacking the histone H1 RNAi construct ( parental , data not shown ) . Here too , the majority of VSG switch events involved deletion of the VSG221 ES . This high frequency of deletion of the VSG221 ES after a VSG switch has been previously observed [60] , [61] , [63] , [64] . We still do not know why VSG ES deletion events are frequently picked up in VSG switching experiments with T . brucei 427 . VSG switching experiments performed with T . brucei have shown variability regarding the predominant VSG switch mechanism used . Some reports have found that VSG gene conversion is the most common switching mechanism [60] , [65] , while others have shown that transcriptional switching between ESs predominates [66] , [67] . However , consistent with both these results and those from other laboratories , deletion of the previously active ES frequently occurs [60] , [61] , [63] , [64] , [66] . This phenomenon has been frequently observed with the VSG221 ES , which is often deleted during switch events involving a switch to another ES . The VSG221 ES is unusually large with extensive duplications and triplications . In addition , it contains unusually short regions of 70 bp repeats which normally facilitate gene conversion [8] , [9] . As the VSG221 ES has been hypothesised to have a particularly low frequency of inactivation , it has been postulated that if the rate of switch off of this ES drops low enough , telomere deletion events resulting in its loss are an alternative way of resolving an unfavourable double-expressor state [64] . This VSG221 ES deletion event is not a consequence of the TK negative selection system , as these events have also been uncovered in experiments using positive selection for ES activation using drugs or negative selection against VSG221 using VSG RNAi [66] . We attempted to investigate the effect of H1 depletion on VSG switch events involving DNA rearrangements near the telomeric VSG . We repeated our VSG switching assays in the presence of puromycin selection to maintain expression of the VSG221 ES . This procedure selects for cells that continue to transcribe the active VSG221 ES , but have silenced ( through telomere exchange ) or lost ( through VSG gene conversion ) the HYGTK fusion gene . As with the previous switching assay , this procedure will also select for cells with mutations in the TK gene [60] , [61] , [63] , [64] . Using these conditions we also observed an increase in GCV-resistant clones in cells containing the histone H1 RNAi construct , although this frequency did not further increase after induction of H1 RNAi , presumably as H1 was already significantly depleted in the noninduced cells containing the H1 RNAi construct ( Fig . 9C ) . We analysed the VSG switching mechanisms that had occurred in clonal cell lines derived from at least three independent cultures each of parental cells ( n = 19 ) , uninduced T . brucei H1 RNAi cells ( n = 21 ) , and cells induced for H1 RNAi for 48 h ( n = 15 ) . Here we found that our protocol had enriched for cells that continue to express VSG221 , and are probably TK mutants ( Fig . 9D ) . However , we also observed that a higher percentage of the clones in which H1 had been depleted had indeed switched their VSG via gene conversions , indicating that histone H1 in T . brucei possibly plays a role in suppressing DNA rearrangements at the telomeric end of the ES . One possibility explaining these data , is that what we observe here are two potentially distinct effects of histone H1 depletion . Leaky H1 RNAi in the uninduced sample results in an increase in the frequency of TK mutants , but does not significantly affect the VSG switching frequency . However , further depletion of histone H1 after tetracycline induction of H1 RNAi causes an increase in VSG switching frequency . We next investigated if the rate of DNA recombination in the Pol II-transcribed URA3 locus was also affected by H1 knockdown . We didn't observe dramatic changes in chromatin structure at the URA3 locus after knockdown of H1 , although H1 depletion did result in an increase in MNase accessibility of chromatin at other Pol II transcription units ( Fig . 5E ) . However , rates of gene conversion at loci other than URA3 are difficult to measure in the absence of negative selectable markers which can select for both alleles . As we could not exclude that minor ( and for us undetectable ) changes in chromatin structure were nonetheless occurring at the URA3 locus which could be affecting DNA recombination , we used an assay described in [60] , [61] to measure the frequency of gene conversion at this locus after H1 knockdown ( Fig . 10A ) . T . brucei expressing URA3 is sensitive to 5-fluoroorotic acid ( FOA ) . We replaced one allele of URA3 with the hygromycin resistance/thymidine kinase fusion gene ( HYGTK ) in bloodstream form T . brucei that either did ( H1 RNAi ) or did not ( Par ) contain the H1 RNAi construct . Removal of cells from hygromycin selection allows gene conversion at the URA3 locus to occur . This results in cells with either two copies of the URA3 gene ( which are sensitive to FOA but resistant to GCV ) , or two copies of the HYGTK gene ( which are sensitive to GCV but resistant to FOA ) . Parental and H1 RNAi URA3/HYGTK T . brucei cells were expanded for 48 hours in the absence of hygromycin selection . During this period , one of the H1 RNAi cultures was induced with tetracycline ( H1 RNAi 48 h ) . We next placed cells under either FOA or GCV selection , and scored for positive wells after eight days . Validation of a selection of these clones by PCR confirmed that , as expected , GCV resistant clones lacked the HYGTK gene while FOA resistant clones lacked the URA3 gene ( data not shown ) . The frequency of positive wells in both FOA and GCV-containing plates was divided by the number of generation times in the absence of hygromycin to obtain the total frequency of gene conversion per generation for each sample ( Fig . 10B ) . We did not observe a significant difference in frequency of gene conversion at the URA3 locus in parental ( Par ) or uninduced cells , compared with cells in which histone H1 had been depleted for 48 hours ( n = 3 ) .
We find that the single-domain histone H1 proteins play an essential role in the maintenance of higher order chromatin structure in bloodstream form T . brucei . H1 depletion results in a general disruption of electron dense material in the nucleus which appears to correspond to heterochromatin . In addition , H1 knock-down results in an increase in the accessibility of chromatin to MNase in bloodstream T . brucei , as well as a marked opening of chromatin structure at the silent ES promoters . In agreement with this , H1 knockdown results in derepression of silent ES promoters , and also appears to lead to an increase in VSG switching . These data all argue that H1 could be one of the layers of control allowing the parasite to tightly control expression of its extensive VSG repertoire . H1 knockdown in procyclic form T . brucei resulted in only moderate derepression of ESs , and no obvious changes in global chromatin structure as assessed using MNase accessibility . It is possible that the chromatin of wild-type procyclic form T . brucei is in a more open state than that found in bloodstream form T . brucei [45] , or alternatively that H1 interacts with chromatin differently in procyclic form compared with bloodstream form T . brucei . T . brucei histone H1 has unusual properties . While H1 proteins normally have a tripartite domain structure including a central globular domain flanked by relatively unstructured N- and C-terminal domains , kinetoplastid H1 proteins have only a single domain rich in lysine , alanine , and proline residues [33] . Here we show that T . brucei H1 proteins , while missing domains important for function in other eukaryotes , can still facilitate higher order chromatin structure . It is possible that the mechanism by which single-domain H1 proteins compact chromatin is different from that used by the tripartite H1 proteins [21] . In T . brucei , both bloodstream and procyclic form life-cycle stages still grow at reduced rates even if histone H1 is depleted down to very low levels . Histone H1 is nonessential in a number of unicellular eukaryotes [16]–[18] . Although histone H1 depletion in T . brucei results in a mild growth phenotype , it is possible that here too these proteins are not essential . However , in the absence of a true genetic H1 knockout ( complicated by the fact that there are at least five histone H1 genes ) it is not possible for us to establish this point definitively . It is also unclear if the different isoforms of T . brucei H1 perform distinct roles or have overlapping functions . While detailed biochemical analysis of H1 proteins has been performed in T . brucei , T . cruzi , L . major , and most recently L . braziliensis , it has not yet been thoroughly investigated how these proteins function in vivo . T . cruzi has three H1 genes , and phosphorylation sites of the encoded proteins have been mapped [68] . Interestingly , phosphorylated H1 seems to be regulated according to both the cell and developmental cycle of T . cruzi [68]–[70] . The phosphorylated H1 also has a distinct localization compared to the non-phosphorylated proteins , and is enriched in the area of the nucleolus [69] . Leishmania species have at least two H1 proteins ( L . brasiliensis has three ) [71] that in L . major have been shown to be developmentally regulated [72] . Overexpression of H1 in this species affects both chromatin structure and parasite virulence [38] , [73] . However , the challenges of working with a multi-gene family as well as the limited availability of tools for genetic manipulation , has precluded loss of function analysis of H1 in these organisms . In addition to providing insight into the role of H1 , this study also provides information on the chromatin structure of T . brucei . Genomic areas that are actively transcribed by Pol I including the rDNA , active ESs , and the procyclin loci are highly depleted of nucleosomes , and have an unusually open chromatin structure [13] , [14] . We now find that the linker histone H1 is also depleted from these highly transcribed Pol I transcription units . Interestingly , histone H1 as well as histone H3 are also depleted from the procyclin promoter even in bloodstream form T . brucei , where procyclin loci are transcriptionally silent . This could indicate the presence of an open chromatin structure associated with ‘paused’ Pol I complexes . Pol I is thought to initiate transcription at procyclin promoters in bloodstream form T . brucei , which is not fully processive [74] . In contrast , in Pol II transcription units including actin , γ-tubulin , the URA3 locus and other single copy genes , the chromatin structure as investigated using histone H1 distribution and MNase accessibility in parental cells appears comparable to that of silent ESs . This indicates that a highly open chromatin structure is not a prerequisite for processive Pol II transcription in T . brucei . It has previously been found that the T . brucei nucleus is heterogeneously stained in EM thin sections [6] , [53] , [75] . The dark and lightly-stained nuclear areas are thought to correspond to heterochromatin and euchromatin respectively , although it is not clear exactly which genomic regions are located in these areas . Here , we show that the darker-stained heterochromatic regions of the nucleus disappear following depletion of H1 in bloodstream form T . brucei . Our ChIP experiments show that H1 is enriched at transcriptionally silent regions , including repeat regions and the VSG basic copy array , which can be assumed to be present in the form of heterochromatin . As our MNase accessibility experiments indicate that H1 knockdown causes the chromatin in these areas to become more open , it is therefore highly likely that the darkly stained regions of the T . brucei nucleus visualized by EM are in fact areas of more tightly packaged DNA . It is possible that H1 knockdown also has an effect on other aspects of subnuclear organization , including telomere clustering and the distribution of minichromosomes , thereby leading to some of the functional consequences we observe . Interestingly , knock-down of NUP-1 , a lamin-like protein important for nuclear structure in T . brucei , has also been shown to affect VSG switching frequency [76] . In addition , another aspect of nuclear organization , the association of sister chromatids during mitosis , has been proposed to play a role in regulation of antigenic variation , as depletion of cohesin subunits results in an increased VSG switching frequency [77] . What is the role of H1 mediated chromatin structure in an organism that has little transcriptional control ? Although H1 knock-down resulted in an increase in accessibility of chromatin to nucleases in bloodstream form T . brucei , this effect was not observed in procyclic form T . brucei . It is clear that one function of H1 could be to facilitate different processes required for efficient antigenic variation in T . brucei . We also find that histone H1 plays a role in silencing inactive ESs particularly in bloodstream form T . brucei . In contrast , depletion of H1 in S . cerevisiae had no effect on telomeric silencing [17] . Our results argue that the higher-order chromatin structure maintained by histone H1 in T . brucei is critical for complete ES silencing in bloodstream form T . brucei . In addition to being enriched on silent VSG ESs , H1 in T . brucei also appears to play a role in maintenance of the closed chromatin state found at the silent basic copy VSGs . We find that H1 is enriched on silent VSGs including the VSG118 basic copy gene , and H1 knock-down results in a more open chromatin structure at this locus . For antigenic variation to be effective , rates of DNA recombination need to be suppressed to levels that prevent unnecessarily rapid depletion of the VSG repertoire . VSG switching rates of the laboratory adapted T . brucei 427 strain are significantly lower than in strains which are closer to field isolates [65] . The factors that contribute to these high VSG switch rates in some isolates have not yet been determined . However , we feel that the first step is to identify proteins which modulate the frequency of VSG switching in experimentally amenable laboratory strains , before verifying their relevance in the field . We find that depletion of H1 not only disrupts VSG ES silencing , but also appears to result in an increase in the frequency of VSG switching . It is possible that depletion of H1 from the silent ESs causes an increase in homologous recombination . This is consistent with a proposed role for H1 in suppressing recombination events and maintaining genome stability in yeast [27] , [78] . Homologous recombination plays a major role in VSG switching [2] , [79] . Depletion of BRCA2 , Rad51 and RAD51-related proteins all decrease VSG switching frequency , presumably through disruption of homologous recombination pathways [80]–[83] . In contrast , knockout of subunits of the T . brucei RTR complex , which is thought to resolve recombination intermediates , causes an increase in the VSG switching frequency [60] , [61] . Possibly higher-order chromatin structures maintained by H1 could suppress the formation of recombination intermediates between the highly similar VSG ESs , as has been found in S . cerevisiae in the rDNA transcription units [27] , [78] . Depletion of H1 had a much more dramatic effect on chromatin structure in the region immediately downstream of the silent ES promoters compared with at the telomeric VSG gene . It is possible that in the absence of H1 other factors ( such as RAP1 ) are still affecting chromatin structure in the region adjacent to the telomere [84] . Depletion of H1 also affected the chromatin structure of a VSG located in a basic copy array . Although our ChIP and MNase accessibility data indicate that H1 could affect chromatin structure in at least some Pol II transcription units , knockdown of H1 did not result in increased MNase accessibility at the URA3 locus . Since H1 knockdown does not alter chromatin structure at the URA3 locus , it is perhaps unsurprising that rates of homologous recombination at this locus do not increase after H1 depletion . Although there was some effect of H1 depletion on the chromatin of other Pol II loci including actin or γ-tubulin , H1 depletion had a less dramatic effect on the chromatin structure at these loci compared with at the ES promoter regions . In S . cerevisiae , histone H1 has been shown to suppress homologous recombination at the repetitive Pol I transcribed rDNA loci , but had no effect on homologous recombination at another locus outside of the rDNA [78] . Here , we show that H1 containing heterochromatin is not only involved in silencing VSG ES promoters , but could also be involved in suppressing VSG switching . Our studies on T . brucei histone H1 provide insights into the unique structure and role of chromatin in these parasites , and these data therefore provide additional evidence for the key role played by chromatin structure and remodeling in the processes involved in antigenic variation in these important human pathogens .
Bloodstream form ( BF ) T . brucei brucei strain 427 was cultured in HMI-9 medium [85] supplemented with foetal calf serum ( FCS ) and the appropriate drugs at 37°C , 5% CO2 . During the course of these experiments the amount of FCS in the HMI-9 was reduced from 20% to 15% , with no observable effect on cell growth or the kinetics of the histone H1 RNAi phenotype . Procyclic form ( PF ) T . brucei was cultured in SDM-79 medium supplemented with 10% FCS , 5 µg/ml hemin , and the appropriate drugs at 27°C . Cell densities were determined using a haemocytometer . The BF T . brucei RYT3-H1 cell line was generated by transfection of the p2T7-H1hy histone H1 RNAi construct into the T . brucei RYT3 cell line . This parental cell line is based on the BF T . brucei ‘single-marker’ [86] cell line , and contains a blasticidin resistance gene in the active VSGT3 ES , and a 221GP1 puromycin/eGFP construct integrated behind the silent VSG221 ES promoter [42] , [87] . This BF T . brucei RYT3-H1 cell line was used for experiments monitoring the general phenotype after the induction of histone H1 RNAi including growth , protein and transcript levels , micrococcal nuclease ( MNase ) sensitivity , and global nuclear architecture as investigated using electron microscopy ( EM ) . For the experiments analysing VSG switching , BF T . brucei ‘single marker’ cells were transfected with the 221GP1 construct ( containing the puromycin and eGFP genes ) which integrates immediately downstream of the active VSG221 ES promoter [87] . Subsequently , the p2T7-H1ph histone H1 RNAi construct ( described below ) was introduced and selected for using the phleomycin resistance gene . Finally , a construct containing a hygromycin resistance-thymidine kinase fusion gene ( HYGTK ) [60] ( see below ) was integrated into the active VSG221 ES between the 70 bp repeat sequences and the VSG221 gene itself . To examine the effect of histone H1 depletion in procyclic form T . brucei , the PF T . brucei DsRed-H1 cell line was created . To do this , the p2T7-H1ph construct was transfected into the 221BsrDsRed parental T . brucei line [42] , which is based on the T . brucei 29-13 line [86] , and contains a blasticidin resistance gene and a DsRed fluorescent protein gene in the silent VSG221 ES . Phleomycin selection was used to select for stable integration of the construct , and clonal cell lines were obtained . ChIP experiments were performed using the BF T . brucei HNI 221+ [64] and PF T . brucei 221BsrDsRed cell lines [42] . Five genes have been annotated as histone H1-like in the T . brucei brucei genome with the TriTrypDB accession numbers: Tb11 . 42 . 0005 , Tb11 . 42 . 0006 , Tb11 . 42 . 0007 , Tb11 . 55 . 0001 , and Tb11 . 39 . 0008 . Protein sequence alignments were performed using ClustalW . To produce double-stranded RNA ( RNAi ) homologous to the different polymorphic members of the histone H1 gene family in T . brucei , two histone H1 fragments ( ∼600 bp each ) corresponding to different histone H1 genes were PCR amplified from the T . b . brucei 427 genome , and cloned in tandem into the p2T7-177 RNAi vector which targets T . brucei minichromosomes [51] . The first histone H1 fragment includes the coding region of Tb11 . 42 . 0005 as well as the intergenic region between Tb11 . 42 . 0005 and Tb11 . 42 . 0006 , and was amplified using primers HisH1Con1_183s 5′-TATGGATTCAGACAACTGCTGTCCCCAAG-3′ ( BamHI link ) and HisH1Con1_884as 5′-TATAAGCTTGAGCAGCAGATGCCTTCG-3′ ( HindIII link ) . The second histone H1 fragment includes the intergenic region between Tb11 . 39 . 0008 and the coding region of Tb11 . 55 . 0001 , and was amplified using primers HisH1Con2_524s 5′-TATAAGCTTCCCGCTATTAGACACGCTATG-3′ ( HindIII link ) and HisH1Con2_1173as 5′-TATCTCGAGGATGCGCTCACGCCTTCT-3′ ( XhoI link ) . These two histone H1 fragments were inserted by triple ligation into the p2T7-177-Hygro or p2T7-177-Phleo constructs to generate the p2T7-H1hy and p2T7-H1ph constructs . These contained a ∼1 . 2 kb fragment producing dsRNA capable of RNAi-mediated knockdown of all five predicted T . brucei histone H1 genes . In order to integrate a construct containing the HYGTK fusion gene into the VSG221 ES , we employed a strategy similar to that of Kim and Cross , 2009 . First , we cut out a ∼2 . 5 kb region of 70 bp repeat sequence from the construct RM3173 [67] and cloned it into pBluescript . We next amplified by PCR a region of the VSG221 ES between the 70 bp repeats and the VSG221 gene using primers VSG221TAR_55928s 5′-TATGGATCCGACGAATACAAACCATAAATAAATGC-3′ ( BamHI link ) and VSG221TAR_56433as 5′-TATGCGGCCGCCAAGACGTGGTGCAATCATC-3′ ( NotI link ) and cloned it into the same vector . We finally inserted the HYGTK fusion gene ( a generous gift of Nina Papavasiliou and George Cross ) flanked by an upstream tubulin intergenic region containing an RNA splice site and a downstream actin intergenic region containing a polyadenylation signal in between the 70 bp repeat fragment and the VSG221 ES targeting fragment . The vector was cut with HindIII and NotI prior to transfection , and correct integration confirmed by PCR linking . To create antibodies specific for T . brucei histone H1 proteins , two histone H1 peptides were designed , synthesized , and injected into rabbits ( Eurogentec ) . The T . brucei histone H1 peptide sequences are: N-KKVAPKKVAGKKAAA-C ( amino acids 59–73 of Tb11 . 55 . 0001 ) and N-AAPKKAVAKKAAPKK-C ( amino acids 6–20 of Tb11 . 55 . 0001 ) . Affinity purification of antibodies specific for these peptides was carried out by Eurogentec . The specificity of the T . brucei histone H1 antibodies was confirmed using Western blots comparing lysates from wild-type T . brucei with lysates from T . brucei after RNAi-mediated knockdown of histone H1 . To create antibodies specific for VSG221 , a ∼500 bp fragment was amplified by PCR using primers VSG221Ab_243s 5′-GCAAGTATATACGCTGAAATAAATCAC-3′ and VSG221Ab_741as 5′-TGTTTGGCTGTTCGCTACTGTGAC-3′ and cloned into the pRSETA vector ( Invitrogen ) for expression of an N-terminally tagged 6×-His-tagged protein . Protein was purified under denaturing conditions and was used for immunisation of rabbits ( Eurogentec ) . Tris-Tricine SDS-PAGE electrophoresis was performed with 16% polyacrylamide gels before being transferred to PVDF membrane ( GE Healthcare ) using standard protocols . Blots were probed with affinity-purified anti-histone H1 peptide antibodies , anti-histone H3 antibody ( ab1791 , AbCam ) , anti-La antibodies ( a gift from Elisabetta Ullu ) , or anti-BiP ( a gift from Jay Bangs ) . Detection of the appropriate peroxidase-coupled secondary antibody was performed using ECL Plus ( Amersham ) detection kit . To determine the affinity of histone H1 for chromatin in T . brucei strain ( BF ) HNI 221+ or ( PF ) 221BsrDsRed , 1×107 cells per sample were centrifuged at 1000 rcf for 7 min . Cell pellets were washed twice and the supernatants removed . For “total” samples , cell pellets were resuspended in 200 µl hot Tricine SDS-PAGE sample buffer . For other samples , cell pellets were resuspended in PBS+1% Triton X-100 containing either 0 , 100 , 200 , 400 , or 800 mM NaCl . Roche protease inhibitors ( −EDTA ) were also included in each lysis solution . Lysates were incubated on ice for 20 min , followed by centrifugation at 16000 rcf for 20 min at 4°C . 25 µl of each supernatant was removed to a fresh tube . Hot 2× Tricine SDS-PAGE buffer ( 25 µl ) was then added to each supernatant and samples were boiled for 5 min . The rest of the supernatant was discarded from each sample , and the pellets resuspended in 200 µl hot Tricine SDS-PAGE sample buffer . All other protein lysates were prepared by resuspension of a washed cell pellet in hot Tricine sample buffer to a final concentration of 105 cell equivalents per µl . RNA isolation , cDNA production and qPCR analysis of steady state transcript levels was performed as described in Narayanan et al 2011 [43] . T . brucei PF 221BsrDsRed cells ( 1×109 ) were centrifuged at 1000 rcf for 10 min and washed twice with 10 ml trypanosome wash solution ( 100 mM NaCl , 3 mM MgCl2 , 20 mM Tris-HCl pH 7 . 5 at 4°C [88] . The cells were then washed once in 10 ml transcription buffer ( 150 mM sucrose , 20 mM L-glutamic acid , 20 mM HEPES-KOH pH 7 . 7 , 3 mM MgCl2 , 1 mM DTT , Roche complete protease inhibitors −EDTA , and Sigma phosphatase inhibitor cocktails 2 and 3 ) , at 4°C [88] . The cell pellet was resuspended in 300 µl transcription buffer and kept on ice for 20 min . Cells were lysed by douncing using short burst of rapid strokes for 45 minutes on ice using a Wheaton 2 ml dounce as modified from [88] . Lysis ( >80% ) was confirmed by light microscopy , then cells were centrifuged at 16000 rcf for 10 min at 4°C . After removal of the supernatant , the pellet was resuspended in 200 µl 5% perchloric acid , and incubated on ice for 40 min [89] . The sample was then centrifuged at 16000 rcf for 20 min at 4°C , and the supernatant transferred to fresh tubes ( 100 µl/1 . 5 ml tube ) . Proteins were precipitated using acetone , and dried pellets were resuspended in Laemmli SDS-PAGE sample buffer . Samples were boiled for 5 min before loading on Tris-glycine SDS-PAGE gels for analysis using standard protocols . For localization of histone H1 proteins , immunofluorescence microscopy was performed essentially as described [43] . T . brucei BF HNI ( 221+ ) and PF 221BsrDsRed cells were used . Histone H1 was detected using the affinity purified T . brucei anti-histone H1 peptide antibody and an Alexa-594-conjugated anti-rabbit secondary antibody ( Invitrogen ) . The T . brucei nucleolus was detected using the monoclonal L1C6 antibody ( Devaux et al 2007 ) ( a gift from Keith Gull ) and an anti-mouse Alexa-488-conjugated secondary antibody ( Invitrogen ) . Slides were visualized on a Zeiss Axio Imager . M1 microscope equipped with a Zeiss AxioCam MRm camera using AxioVision Rel 4 . 8 software . Images were cropped and brightness and contrast uniformly adjusted using Adobe Photoshop . Chromatin immunoprecipitation ( ChIP ) experiments were performed as described [14] . Histone H3 was immunoprecipitated using an anti-H3 antibody ( ab1791 , AbCam ) and served as a positive control for the ChIP procedure . A cross-linked sample in which no antibody had been added served as a negative control . For ChIP of histone H1 proteins , 20 µl of affinity-purified histone H1 antibody was used , with 20 µl of pre-immune serum from the same rabbit used as a negative control . Analysis of the ChIP material was performed using either quantitative PCR ( qPCR ) using the primer sets described in [14] or by slot-blotting and hybridization with radio-labelled probes as described [14] . All qPCR reactions were performed with Agilent Technologies Brilliant II SybrGreen qPCR master mix with low ROX on an Applied Biosystems 7500 Fast Real-Time PCR system . All statistical analyses are unpaired , two-tailed t-tests with a 95% confidence interval , and were performed using GraphPad Prism software . Micrococcal nuclease ( MNase ) treatment of chromatin was performed essentially as described [14] . For small-scale preparations , 2×107 cells per sample were permeabilized with digitonin and treated with either 0 . 0625 , 0 . 125 , 0 . 25 , or 0 . 5 units of MNase ( Worthington Biochemicals ) . Isolated DNA was then loaded on a gel , and bands quantitated using ImageJ . For MNase treatment followed by sucrose gradient fractionation , 2–3×108 cells per sample were permeabilized and treated with 2 . 5 units MNase per 1×108 cells . Solubilised chromatin was then loaded onto a 5–30% discontinuous sucrose gradient , and the different nucleosome species were fractionated by ultracentrifugation at 40 , 000 rpm ( Beckman SW41 Ti rotor ) for 16 h at 4°C . Fractions were treated with proteinase K and RNase A , and DNA isolated by phenol-chloroform extraction followed by ethanol precipitation . After visualization of some of the isolated DNA on a gel , the remaining DNA fractions were pooled into mono- , di- , di/tri- , tri-tetra- , and >tetra nucleosomal fractions . The amount of each DNA target present in each fraction was determined by qPCR using primer sets described in [14] . The summed amount of each target detected in all fractions was designated as the “total” amount , and the amount detected in the mononucleosome fraction was plotted as a percent of that total . Cells were fixed in 2 . 5% glutaraldehyde , post-fixed in 1% osmium tetroxide and stained with 2% uranyl acetate before dehydration through a series of different ethanol concentrations and embedding in Agar 100 epoxy resin . Thin sections ( 90 nm ) were stained with lead citrate and examined using an FEI Tecnai 12 transmission electron microscope at 80 kV according to [90] . Derepression of the silent VSG221 ES in the T . brucei BF RYT3 and PF 221BsrDsRed cell lines was monitored by flow cytometry as described [40] , [42] , [43] with minor modifications . Briefly , histone H1 RNAi was induced with the addition of 1 µg/ml tetracycline . At each time point , ∼1×106 cells were centrifuged , washed , and fixed in 2% paraformaldehyde for one hour at room temperature . Fixed cells were then washed and analysed by flow cytometry using a Becton-Dickinson FACSCalibur and CellQuest software ( BD ) . For each time point , fold derepression was calculated by dividing the mean FL1-H or FL2-H fluorescence value of induced cells by the mean fluorescence value of the uninduced cells . To determine VSG switching frequencies and mechanisms , we employed a strategy similar to that used in [60] , [61] with some modifications . We created the T . brucei 221pGFPhyTK cell line , which contains a construct containing a puromycin and eGFP gene immediately downstream of the promoter in the active VSG221 ES , and a construct containing the HYGTK fusion protein gene downstream of the 70 bp repeats in the same ES . The T . brucei 221pGFPhyTKH1 cell line contains the same marker genes as 221pGFPhyTK with the addition of the p2T7-H1ph RNAi construct . We maintained both T . brucei 221pGFPhyTK ( parental ) and T . brucei 221pGFPhyTKH1 cells under puromycin and hygromycin selection to select for a homogenous population expressing the VSG221 ES . We subsequently removed these cells from drug selection and allowed them to switch for 48 hours . Tetracycline ( 1 µg/ml ) was added to an additional culture of T . brucei 221pGFPhyTKH1 cells to induce RNAi of histone H1 , also for 48 hours . Cells were then plated in 3-fold serial dilutions in the presence of 4 µg/ml ganciclovir ( GCV ) ( Sigma ) , and VSG switching frequencies were calculated based on the number of positive wells ( containing growing cells ) , the number of cells plated for each dilution , and the number of generations undergone in the absence of drugs . We have determined the plating efficiency of the different cell lines using different experimental conditions ( i . e . in the presence or absence of various drugs ) and did not find a clear difference in plating efficiency ( data not shown ) . We therefore did not normalise our data for this variable . However as the T . brucei lines capable of histone H1 RNAi have a longer generation time than parental cells , we expressed our data as switch/recombination events per generation . However , the effect of H1 depletion is also apparent even when the data is not normalised for number of generation times . Individual switched clones were analysed for GFP fluorescence , VSG221 expression ( using immunofluorescence microscopy ) , and PCR for the VSG221 gene ( primers VSG221_243s 5′-GCAAGTATATACGCTGAAATAAATCAC-3′ and VSG221_741as 5′- TGTTTGGCTGTTCGCTACTGTGAC-3′ ) . PCR amplification of the large subunit of RNA polymerase I was performed as a positive control ( Pol1_4102s 5′-CTGGATCCAGCGCCGTTCCACGCGAGA-3′ and Pol1_4554as 5′-GACTCGAGCTATCCCCAATCCGTGCCGTCCCG-3′ ) . Pulsed-field gels were performed essentially as described [63] . The gel was stained and processed for blotting onto Hybond XL membrane ( GE Healthcare ) and probed for VSG221 and VSG1 . 8 using standard protocols . The probes were amplified by PCR using VSG221_121s 5′- TGCCAGGTCTCCGAG-3′ , VSG221_1056as 5′-GCTGCTCGGATATGAGCTTTT-3′ , VSG1 . 8_179s 5′-CAATCTCGAGGCTCACAAAAGTCTG-3′ , and VSG1 . 8_1022as 5′-GCTGGGATCCTAGCCTCGAAAAATG-3′ . Determination of the frequency of gene conversion at the T . brucei URA3 locus was performed as described [60] , [61] . Briefly , the pyrFEKO-HYG construct ( a gift from G . Cross ) containing targeting fragments for the URA3 gene ( Tb927 . 5 . 3810 ) flanking a hygromycin resistance/thymidine kinase fusion gene was transfected into both BF T . brucei 221pGFP cells or BF T . brucei 221pGFPH1 cells ( containing the histone H1 RNAi construct ) . Correct integration of the construct was confirmed by PCR amplification using primers URA3uplink_9228s ( 5′-AGAAAGAACCGTACCGCAGA-3′ ) and Hygro_125as ( 5′-CCTACATCGAAGCTGAAAGCAC-3′ ) for upstream linking and TKlink_2370s ( 5′-TTTACGGGCTACTTGCCAAT-3′ ) and URA3dnlink_12374as ( 5′-AGGGGGAAACAGCGTAAGTT-3′ ) for downstream linking . Cells were grown in the absence of hygromycin selection for 48 hours . Cells were then plated ( 5×105 cells per sample ) in the presence of either 6 µg/ml 5-Flouroorotic acid ( Sigma ) or 30 µg/ml GCV . Plates were scored 8 days later , and the gene conversion frequency was calculated by dividing the frequency of positive wells for each selection method by the number of doubling times undergone by each culture in the absence of hygromycin .
|
Trypanosoma brucei causes African sleeping sickness , endemic to sub-Saharan Africa . Bloodstream form T . brucei is covered with a dense coat of variant surface glycoprotein ( VSG ) . Only one VSG is expressed at a time out of a vast repertoire of ∼1500 VSGs . The active VSG is transcribed in a telomeric VSG expression site ( ES ) , and VSG switching allows immune evasion . Exactly how monoallelic exclusion of VSG ESs operates , and how switching between ESs is mediated remains mysterious , although epigenetics and chromatin structure clearly play a major role . The linker histone H1 is thought to orchestrate higher order chromatin structure in eukaryotes , but its exact function is unclear . We investigated the role of histone H1 in the regulation of antigenic variation in T . brucei . We show that histone H1 is associated with chromatin and is required for higher order chromatin structure . Depletion of histone H1 results in derepression of silent VSG ES promoters , indicating that H1-mediated chromatin functions in antigenic variation in T . brucei .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"cell",
"biology",
"molecular",
"biology",
"gene",
"expression",
"genetics",
"epigenetics",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"chromatin",
"genetics",
"and",
"genomics",
"parasitology"
] |
2012
|
Histone H1 Plays a Role in Heterochromatin Formation and VSG Expression Site Silencing in Trypanosoma brucei
|
Protein kinases C ( PKCs ) and extracellular signal-regulated kinases ( ERKs ) are evolutionary conserved cell signalling enzymes that coordinate cell function . Here we have employed biochemical approaches using ‘smart’ antibodies and functional screening to unravel the importance of these enzymes to Schistosoma mansoni physiology . Various PKC and ERK isotypes were detected , and were differentially phosphorylated ( activated ) throughout the various S . mansoni life stages , suggesting isotype-specific roles and differences in signalling complexity during parasite development . Functional kinase mapping in adult worms revealed that activated PKC and ERK were particularly associated with the adult male tegument , musculature and oesophagus and occasionally with the oesophageal gland; other structures possessing detectable activated PKC and/or ERK included the Mehlis' gland , ootype , lumen of the vitellaria , seminal receptacle and excretory ducts . Pharmacological modulation of PKC and ERK activity in adult worms using GF109203X , U0126 , or PMA , resulted in significant physiological disturbance commensurate with these proteins occupying a central position in signalling pathways associated with schistosome muscular activity , neuromuscular coordination , reproductive function , attachment and pairing . Increased activation of ERK and PKC was also detected in worms following praziquantel treatment , with increased signalling associated with the tegument and excretory system and activated ERK localizing to previously unseen structures , including the cephalic ganglia . These findings support roles for PKC and ERK in S . mansoni homeostasis , and identify these kinase groups as potential targets for chemotherapeutic treatments against human schistosomiasis , a neglected tropical disease of enormous public health significance .
Protein kinases C ( PKCs ) and extracellular signal-regulated kinases/mitogen-activated protein kinases ( ERKs/MAPKs ) are signalling enzymes that play a critical role in regulating cellular processes , such as gene expression , the cell cycle , growth , development and differentiation , cellular motility , survival and apoptosis [1] , [2] . PKC/ERK signalling occurs in response to various stimuli , including ligands that bind receptor tyrosine kinases ( RTKs ) and G-protein coupled receptors ( GPCRs ) [1] , [2] . Putative PKCs and ERKs exist in kinomes of the blood flukes Schistosoma mansoni [3] , [4] , S . japonicum [5] and S . haematobium [6] . These parasites cause human schistosomiasis , a neglected tropical disease ( NTD ) characterised by inflammatory granulomatous reactions in the host organs that occur in response to entrapped eggs from adult female worms [7] . The global significance of human schistosomiasis is huge; more than 200 million people have the disease , and 0 . 8 billion are at risk of infection [8] , [9] . Chemotherapy relies upon praziquantel ( PZQ ) treatment , but this compound kills adult worms and not juvenile stages , and does not prevent re-infection [10] , [11]; emergence of drug resistant strains is also possible [12] . Although anti-schistosome vaccine targets exist [13] , some worms exhibit antigenic polymorphism , making targeting difficult [14] . Thus , there is considerable interest in identifying new anti-schistosome drug targets , and the protein kinases are potential candidate molecules [15] , [16] . In humans 10 PKCs exist , including PKCβI and PKCβII , which arise from alternate gene splicing [17] . PKCs comprise a C-terminal serine/threonine kinase domain linked through a flexible hinge segment to an N-terminal regulatory region that contains diacylglycerol ( DAG ) -sensitive and Ca2+-sensitive domains [18] . Conventional PKCs ( cPKCs: PKCα , PKCβI/βII , and PKCγ ) are DAG sensitive and Ca2+ responsive; novel PKCs ( nPKCs: PKCδ , PKCε , PKCη , and PKCθ ) are DAG sensitive but Ca2+ insensitive; and atypical PKCs ( aPKCs: PKCζ and PKCι , or λ-murine ) are insensitive to both DAG and Ca2+ [18] . Phosphorylation on specific amino acid residues is also crucial to PKC activation [19] . Previously , a DAG/Ca2+-dependent PKC-like activity and a phospholipid/phorbol ester sensitive kinase activity were detected in adult S . mansoni homogenates [20] , [21] , and a PKC ( SmPKC1 ) homologous to human PKCβ was characterised molecularly [22] . Previously , we identified four putative PKCs in the S . mansoni genome with homology to human PKCs , particularly within functional domains [23]; two proteins were similar to human cPKCβI , one to nPKCε and one to aPKCζ [23] , with PKCε also being designated PKCη [4] . Using phospho-specific antibodies , we showed that activated PKCβ associated with the neural mass , tegument , ciliated plates and germinal cells of miracidia , and that PKC activation restricted development to mother sporocysts that parasitize the snail intermediate host [23] . MAPK pathways exist in all eukaryotes , with components being conserved among yeast , invertebrates and mammals [24]–[29] . The ERK pathway features Ras as a monomeric G-protein , Raf as a MAPKKK , MAPK/ERK Kinase ( MEK ) as a MAPKK , and ERK as a MAPK , the last three forming a hierarchical kinase cascade [30] . Humans and many other organisms express ERK1 and ERK2 ( p44 and p42 MAPK ) to varying extents in tissues and more than 150 ERK1/2 substrates exist [2] , including cytosolic , membrane , nuclear and cytoskeletal proteins [30] . Phosphorylation of ERK1/2 on threonine and tyrosine resides within the Thr-Glu-Tyr ( TEY ) motif in the activation loop is essential for activation . In S . mansoni , Ras GTPase activator protein- ( Ras-GAP ) and ERK1/2-like proteins have been detected [31] , and a Ras homologue has been characterised [32] . Activation of the S . mansoni epidermal growth factor receptor ( EGFR; SER ) by human EGF leads to ERK2 phosphorylation in Xenopus oocytes [33] , and hypothetical ERK pathways for S . mansoni and S . japonicum have been reconstructed in silico [3] , [4] , [34] , supporting that ERK signalling is intact in schistosomes . Identifying signalling molecules that play a fundamental role in cellular communication and function of schistosomes represents one of the great challenges of the schistosome post-genomic era [16] . The aims of the current study were to characterise global PKC and ERK signalling in S . mansoni and to determine whether PKC and ERK are critical to schistosome function . We demonstrated several PKC and ERK isotypes , profiling activities in different life-stages . We determined PKC and ERK responses to kinase activators and inhibitors in adult worms , and determined the localization of active kinases in intact worms in situ through functional kinase mapping . Finally , we showed physiological roles for PKC and ERK in schistosomes through the modulation of kinase activities using pharmacological agents . The findings , together with those also presented for the effects of PZQ on PKC/ERK signalling , establish a role for PKC and ERK in worm homeostasis and behaviour , and identify these kinase groups as potential anti-schistosome drug targets .
Laboratory animal use was within a designated facility , regulated under the terms of the UK Animals ( Scientific Procedures ) Act , 1986 , complying with all requirements therein; regular independent Home Office inspections occurred . The Natural History Museum Ethical Review Board approved experiments involving mice and work was carried out under Home Office project license 70/6834 . Anti-phospho antibodies ( Ab ) used to detect phosphorylated PKC and ERK in S . mansoni were anti-phospho-PKC ( pan ) ( βII Ser660 ) rabbit Ab ( #9371 ) , anti-phospho-PKC ( pan ) ( ζ Thr410 ) ( 190D10 ) rabbit mAb ( #2060 ) , immobilized anti-phospho p44/p42 MAPK ( ERK1/2 ) ( Thr202/Tyr204 ) ( D13 . 14 . 4E ) XP rabbit mAb ( #3510 ) , and anti-phospho-p44/42 MAPK ( ERK1/2 ) ( Thr202/Tyr204 ) rabbit mAb ( #9101 ) ( Cell Signalling Technology , New England Biolabs ) . The p44/p42 MAPK ( ERK1/2 ) ( non-radioactive ) immunoprecipitation/kinase assay kit , phorbol 12-myristate 13-acetate ( PMA ) , cell lysis buffer , RIPA buffer , lambda phosphatase , and anti-rabbit horseradish peroxidase ( HRP ) -linked secondary antibodies were also purchased from Cell Signalling Technology . The Omnia S/T peptide 8 kinase assay kit ( KNZ1081 ) , RPMI-1640 medium , foetal bovine serum , antibiotics/antimycotics , and anti-rabbit Alexa Fluor 488 secondary antibodies were purchased from Invitrogen . Pre-cast Precise 10% polyacrylamide gels , West Pico chemiluminescence substrate , Restore Western blot stripping buffer and Halt protease/phosphatase inhibitor cocktail were from Pierce , whereas nitrocellulose membrane was from GE Health . U0126 , GF109203X , and PKC catalytic subunit derived from rat brain were purchased from Merck . Vectashield mounting medium was from Vector Laboratories . PKC and ERK gene candidates were identified from the S . mansoni genome assembly , relying on existing annotations ( http://www . genedb . org/genedb/smansoni ) , Andrade et al . [4] , and our existing analysis [23] . Protein sequences were assessed for similarity with other organisms using pBLAST ( http://www . uniprot . org ) . The detection site of the anti-phospho antibodies was identified within the putative S . mansoni PKC and ERK sequences and was aligned to human PKC and ERKs using MUSCLE ( www . ebi . ac . uk/Tools/msa/muscle ) . The Belo Horizonte strain of S . mansoni was maintained in Biomphalaria glabrata and albino female mice ( BKW strain ) . Miracidia and in vitro transformed sporocysts were collected , concentrated , and processed for Western blotting as previously described [35] , [36] . Briefly , cercariae were collected after they emerged from patent snails and were transferred to 15 ml conical tubes and cooled on ice for 15 min prior to centrifugation ( 200× g ) . Concentrated cercariae were transferred to 1 . 5 ml microfuge tubes , pelleted by pulse centrifugation , and homogenized on ice in 25 µl RIPA buffer containing Halt protease/phosphatase inhibitors . An aliquot was removed for protein quantification ( Bradford assay with Bovine serum albumin ( BSA ) as protein standard ) , 5× SDS-PAGE sample buffer added and samples heated at 95°C for 5 min . Extracts were then either electrophoresed immediately or stored at −80°C . Adult worms were collected by hepatic portal perfusion of mice 40–42 days post infection and were gently washed with pre-warmed RPMI 1640 and either frozen in liquid nitrogen and stored at −80°C for immunoprecipitation/kinase assay , fixed on ice in acetone and stored at 4°C for immunohistochemistry , or placed in RPMI 1640 at 37°C . Following equilibration at 37°C for 30 min in RPMI 1640 , live adult worms were treated for various durations ( 0 , 15 , 30 , 60 , or 120 min ) with either 20 µM GF109203X , 1 µM PMA , 1 µM U0126 , dimethyl sulphoxide ( DMSO; vehicle for PMA or U0126 , 0 . 2% or 0 . 1% , respectively ) , or were left untreated ( RPMI 1640 only ) . In addition , live adult worms were also incubated in 1 µM PMA or DMSO for 24 h at 37°C/5% CO2 in RPMI containing glutamine , glucose , antibiotic/antimycotic mixture ( 100 U penicillin , 100 µg streptomycin and 0 . 25 µg amphotericin B/ml ) and 10% FBS . GF109203X is a PKC inhibitor that competes for the ATP binding site and therefore only catalytically inactive proteins are inhibited . The effect of GF109203X on PKC was thus established by pre-incubating worms in 20 µM GF109203X or RPMI 1640 for 120 min prior to exposure to 1 µM PMA for 30 min at 37°C . The MEK inhibitor U0126 inhibits active and inactive MEK1/2 blocking ERK phosphorylation . Immediately after treatment , medium was removed and worms were either homogenized in 30 µl RIPA buffer and processed for Western blotting or were acetone-fixed for immunohistochemistry in similar ways to that for cercariae ( above ) . Perfused adult worms ( 100 or 20 pairs for PKC or ERK assay , respectively ) were transferred to microfuge tubes and washed twice with RPMI 1640 at 37°C . Worms were then exposed to 1 µM PMA or DMSO ( vehicle ) in RPMI 1640 for 30 min at 37°C prior to being snap frozen in liquid nitrogen and stored at −80°C . When required , worm pairs were defrosted and homogenized on ice in 100 µl cell lysis buffer with Halt protease/phosphatase inhibitor cocktail . The homogenate was then centrifuged for 15 min at 4°C , the supernatant recovered , and the remaining pellet re-homogenized in 50 µl cell lysis buffer with inhibitors and centrifuged for a further 10 min . Supernatants were pooled and equal quantities of protein from each sample incubated overnight at 4°C in either anti-phospho PKC ( pan ) ( ζ Thr410 ) , anti-phospho PKC ( pan ) ( βII Ser660 ) antibodies , or immobilized anti-phospho p44/p42 MAPK ( ERK1/2 ) ( Thr202/Tyr204 ) ( D13 . 14 . 4E ) XP primary antibodies ( each at 1/25 dilution ) . The next day , 50 µl protein A agarose beads were added to homogenates and agitated for 5 min at 4°C except when the pre-immobilized anti-phospho p44/p42 MAPK ( ERK1/2 ) antibody was used; such rapid immunocapture reduces non-specific binding while permitting efficient immunocomplex adsorption [35] , [37] . Following brief centrifugation , samples were washed twice each in 500 µl ice-cold lysis buffer and 500 µl ice-cold kinase buffer ( 25 mM Tris ( pH 7 . 5 ) , 5 mM β-glycerolphosphate , 2 mM DTT , 0 . 1 mM Na3VO4 and 10 mM MgCl2 ) . Negative control immunoprecipitations lacking primary antibodies were also performed . The Omnia ( Ser/Thr 8 peptide ) PKC assay was performed as per the manufacturer's instructions to detect immunoprecipitated PKC activity . Master mix , containing ATP , Ser/Thr 8 substrate , DTT , kinase buffer and water was added to individual wells of black 96-well microtitre plates ( Nunc ) and reactions started by adding immunocomplex . Accumulation of phosphorylated substrate was measured ( excitation 355 nm , emission 460 nm ) every 30 s for up to 200 min at 30°C using a FLUOstar OPTIMA ( BMG Labtech ) microplate reader . Positive control reactions contained 2 ng recombinant human PKC; negative controls either lacked the substrate or comprised samples prepared without the immunoprecipitation antibody . Immunoprecipitated ERK activity was detected using the p44/p42 MAPK ( ERK1/2 ) assay kit . Immunocomplexes were re-suspended in 20 µl kinase buffer supplemented with 200 µm ATP and 2 µg Elk-1 fusion protein and incubated for 30 min at 30°C . Reactions were terminated with 10 µl 3× SDS-PAGE sample buffer and samples heated at 95°C for 5 min in preparation for SDS-PAGE and Western blotting with anti-phospho Elk-1 antibodies to reveal ERK1/2 activity . Negative controls comprised samples without immunoprecipitation antibody . Adult worms were placed in individual wells of a 24-well culture plate ( Nunc; 3–6 worm pairs per well ) in 1 . 5 ml RPMI 1640 containing glutamine , glucose , antibiotic/antimycotic mixture and 10% FBS at 37°C/5% CO2 [22] for 1 h to equilibrate . To determine effects of PKC inhibition/activation or ERK inhibition on adult worms , 1 µM , 5 µM , 20 µM or 50 µM GF109203X or U0126 , or 1 µM PMA were added to cultures . Control groups containing RPMI 1640 only or 0 . 5% DMSO ( vehicle for PMA/U0126 ) were also included and all media and components were replenished daily . Worms were observed at various times over 96 h using an Olympus SZ54045 binocular dissecting microscope and avi-format movies captured using a JVC TK-1481 composite colour video camera linked to Studio Launcher Plus for Windows software . Worm behaviour including pairing status and ventral sucker attachment to the base of the culture plate was determined . Egg release by worms was also enumerated . Detailed analysis of worm movement was done using image J [38] ( http://rsbweb . nih . gov/ij/ ) ; the diameter of the well in pixels was calibrated to mm and the distance travelled by the posterior tip of each worm in 10 s was manually tracked and measured enabling translation into speed of movement ( velocity ) of pixels/s to mm/s . Worm coiling ( Figure 10A ) was determined by counting the number of coils that persisted during 10 s visualization . Results are representative of four independent experiments with a minimum of three replicates each; 30 or more parasites per treatment were scored , except for DMSO controls ( n = 24 ) . Racemic PZQ powder ( Shin Poong Pharmaceutical ) was dissolved in DMSO . Freshly perfused adult worm pairs were incubated in RPMI 1640 containing 0 . 2 µg/ml PZQ , 0 . 1% DMSO ( vehicle ) , or RPMI 1640 alone at 37°C for 15 , 30 or 120 min . Next , worms were either processed for Western blotting or immunohistochemistry as detailed above . The final concentration of PZQ used was similar to a study [39] that reported 0 . 2 µg/ml PZQ to be the lowest concentration needed to induce maximal muscular contraction in worms , a phenotype observed with PZQ treatment . Equal amounts of sample protein ( 12 µg ) were separated on 10% Precise SDS–PAGE gels , semi-dry transferred to nitrocellulose membranes and stained with Ponceau S to confirm homogeneous transfer . After blocking for 1 h in 5% non-fat dried milk , membranes were washed three times in 0 . 1% Tween-Tris-buffered saline ( TTBS ) and incubated in anti-phospho PKC ( pan ) ( ζ Thr410 ) , anti-phospho PKC ( pan ) ( βII Ser660 ) or anti-phospho p44/p42 MAPK ( ERK1/2 ) ( Thr202/Tyr204 ) antibodies ( 1∶1000 in TTBS ) overnight at 4°C . After TTBS wash , blots were incubated for 2 h at room temperature with HRP-conjugated secondary antibodies ( 1∶3000 in TTBS ) and exposed to West Pico chemiluminescence substrate . Immunoreactive bands were visualized with a GeneGnome imaging system ( Syngene ) and relative band intensities quantified using GeneTools software . Protein loading was determined by incubating blots with anti-actin antibodies ( 1∶1000 ) . When necessary blots were stripped with Restore Western blot stripping buffer and incubated in an alternative antibody . In addition , to confirm that the anti-phospho PKC and anti-phospho ERK antibodies detected only the phosphorylated kinases , blots were treated with lambda phosphatase ( 400 U/ml in TTBS containing 1% BSA and 2 mM MnCl2 ) for 4 h before incubation in primary antibodies; secondary antibody labeling and detection were then performed . Acetone-fixed worms were washed twice with PBS , treated with 100 mM glycine for 15 min and blocked in 10% goat serum for 1 h . After a further wash , worms were incubated in anti-phospho PKC ( pan ) ( ζ Thr410 ) , anti-phospho PKC ( pan ) ( βII Ser660 ) or anti-phospho p44/p42 MAPK ( ERK1/2 ) ( Thr202/Tyr204 ) antibodies ( 1∶50 dilution in PBS with 5% BSA ) for 3 days on a rotator . Worms were then washed twice in PBS for 1 h each and incubated in Alexa Fluor 488 secondary antibodies ( 1∶500 in 5% BSA ) and 2 µg/ml rhodamine phalloidin for 24 h followed by two 1 h washes in PBS . Next , specimens were mounted on slides in Vectashield . All incubations were carried out at room temperature and washes were done in microfuge tubes . Adult worms were visualized on a Leica TCS SP2 AOBS confocal laser scanning microscope using 20× dry , 40× or 62× oil immersion objectives and images captured . The signal received from negative controls ( i . e . worms incubated with secondary antibody only ) was negated from that of the positive samples . This was achieved by reducing the power level of the photomultiplier tube , which was then kept constant for all observations . Statistical comparisons were performed with one-way analysis of variance ( ANOVA ) using Minitab ( version 16 ) . All data were expressed as mean ±SEM , and statistical significance was determined by a Fisher's multiple pair-wise comparison .
The phospho-specific antibodies used in this study were anti-phospho PKC ( pan ) ( ζ Thr410 ) , anti-phospho PKC ( pan ) ( βII Ser660 ) and anti-phospho p44/p42 MAPK ( ERK1/2 ) ( Thr202/Tyr204 ) . Using the S . mansoni genome , four putative PKCs [23] and five putative ERKs displaying homology to human isotypes were identified from full-length sequences . Comparative analysis revealed that , except for Smp_131700 ( ∼95 kDa PKC ) , the antibody recognition sequence was conserved in the putative PKCs and ERK1/2 , with the key phosphorylation motif retained; similar levels of sequence homology exist between the human isotypes recognized by these antibodies ( Figure 1 ) . Such motif conservation is common , because of the critical nature of the phosphorylation site in activation of the respective kinases [1] , [2] , [18] , [40] . As with human PKCs , in S . mansoni , the sequence surrounding the phosphorylated Thr residue within the PDK1 consensus motif was more conserved than that surrounding the Ser phosphorylation site in the bulky ring motif ( Figure 1 ) . These antibodies were then used to detect phosphorylated ( activated ) PKCs and ERK1/2 in homogenates ( 12 µg protein ) of four different life stages ( miracidium , sporocyst , cercaria and adult worm ) of S . mansoni by Western blotting . The anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies have been used previously to detect activated PKCs in invertebrates other than S . mansoni ( e . g . [41] ) . They detect PKCα , βI , βII , γ , δ , ε , η , θ , and ζ/ι isotypes only when phosphorylated at a site homologous to Thr410 of human PKCζ ( Figure 1 ) that is critical to catalytic competency and PKC activation [42] . In adult worms and cercariae , three immunoreactive bands were consistently detected , namely an ∼78/∼81 kDa doublet and a ∼132 kDa band with greater immunoreactivity in adult males than cercariae ( Figure 2A ) . A weak band was also seen just above the ∼132 kDa band; however , this was not quantified in later experiments with worm pairs , as it was not always evident . Occasionally , a weak immunoreactive band was also detected at ∼116 kDa in male worms and worm pairs ( Figure 2A ) . In sporocysts and miracidia , a single ∼81 kDa band was detected which appeared highly phosphorylated in sporocysts compared to the other life stages studied ( Figure 2A ) . Previously , we validated anti-phospho PKC ( pan ) ( βII Ser660 ) antibodies to detect phosphorylated ( activated ) PKC in S . mansoni miracidia and sporocysts [23]; they have also been used on other invertebrates , such as the snail Lymnaea stagnalis [43] , [44] . These antibodies recognize PKCα , βI , βII , δ , ε , η , and θ isotypes only when phosphorylated at a residue homologous to Ser660 of human PKCβII ( Figure 1 ) that is crucial to activation [45] . Anti-phospho PKC ( pan ) ( βII Ser660 ) antibodies detected up to three immunoreactive bands in homogenates of adult S . mansoni ( Figure 2B ) . Two faint bands were observed at ∼78 kDa and ∼132 kDa , similar to with anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies; in addition , a more immunoreactive band of ∼116 kDa was seen , occasionally also detected with anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies . This ∼116 kDa protein was only detected in cercariae and adult worms , with greater immunoreactivity in worm pairs ( Figure 2B ) . No immunoreactive proteins were detected in sporocyst homogenates with anti-phospho PKC ( pan ) ( Ser660 ) antibodies , and in miracidia only one protein was detected ( ∼78 kDa ) ( Figure 2B ) . This finding is in accord with our previously published work that demonstrated that this PKC became inactive during the miracidium-to-mother-sporocyst transition [23] . Finally , anti-phospho p44/42 MAPK ( ERK1/2 ) ( Thr202/Tyr204 ) antibodies have been used extensively to detect ERK phosphorylation ( activation ) in invertebrates including flies , snails and nematodes [44] , [46]–[49] , but not in S . mansoni . These antibodies detect ERK1/2 when phosphorylated at residues homologous to Thr202/Tyr204 of human ERK1 ( Figure 1 ) with phosphorylation critical to activation . Three proteins were detected using these antibodies across the four life stages of S . mansoni investigated ( Figure 2C ) . Proteins of ∼48 kDa and ∼43 kDa were detected in adult worm homogenates , with greater immunoreactivity consistently observed in males . In cercariae and sporocysts , a ∼43 kDa band and a weaker ∼38 kDa band were evident but the ∼48 kDa protein was not seen . The ∼43 kDa band was more intense in sporocysts than in cercariae , female worms or adult worm pairs ( Figure 2C ) . Strikingly , there were no visible immunoreactive ERKs in miracidia homogenates ( Figure 2C ) , when probing equal protein amounts from each life stage . Treatment of blots containing protein extracts ( 20 µg ) of adult worm pairs with lambda phosphatase for 4 h prior to exposure to each of the anti-phospho antibodies resulted in either a total loss ( Figures 2B , 2C ) or substantial reduction ( Figure 2A ) in immunoreactivity , demonstrating that the antibodies react specifically with the phosphorylated forms of these proteins . Next , we determined whether the phosphorylation ( activation ) state of the bands detected with the anti-phospho-PKC antibodies could be modulated using the PKC activator PMA or inhibitor GF109203X , reagents that we used previously to characterise PKC signalling in S . mansoni miracidia [23] and L . stagnalis haemocytes [49] . Phosphorylation of all bands increased significantly ( p≤0 . 05 ) when live adult worms were exposed to 1 µM PMA for 30 min ( Figure 3A ) , demonstrating that the endogenous PKC-like proteins are not fully activated in adults under physiological conditions . The ∼116 kDa and ∼132 kDa proteins showed the greatest increases in phosphorylation and are therefore likely cPKCs , which are characteristically highly responsive to DAG , Ca2+ and phorbol esters such as PMA [19] . A biphasic effect of PMA on PKC phosphorylation was also evident with increased phosphorylation observed between 15 and 30 min , followed by a decrease to basal levels at 60 min increasing again at 120 min ( data not shown ) . GF109203X inhibits PKC activity via competition with the ATP binding site and does not directly inhibit active forms of the enzyme . Therefore , the inhibitory effect of GF109203X on S . mansoni PKCs was assessed by pre-incubating worms in GF109203X for 120 min , followed by exposure to 1 µM PMA for 30 min . Under these conditions , the PMA-induced phosphorylation of PKCs was blocked ( Figure 3A ) . Then , to determine whether the immunoreactive phosphoproteins possessed PKC activity an immunoprecipitation/kinase assay was conducted . Adult worm proteins immunoprecipitated with either anti-phospho PKC antibody were able to phosphorylate the PKC substrate ( Ser/Thr 8 peptide ) , with higher activity achieved with anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies than with anti-phospho PKC ( pan ) ( βII Ser660 ) antibodies; activity also increased when worms were exposed to 1 µM PMA for 30 min ( Figure 3B ) . The differences in kinase activity , seen as a result of the different antibodies having been used , is most likely due to the antibody affinity towards , and access to , the various epitopes in the native protein , together with the differential recognition of PKC isoforms displayed in Figure 3A . Activated MEK1/2 is known to phosphorylate ERK1/2 in the TEY motif [30] . Therefore , to evaluate the effect of MEK inhibition on ERK phosphorylation , live adult worms were exposed to 1 µM of the MEK inhibitor U0126 for 30 , 60 or 120 min . Immunoblotting revealed that U0126 significantly attenuated ERK phosphorylation over time , compared to controls ( p≤0 . 05; Figure 4A ) . After 30 min , phosphorylation of the ∼48 kDa and ∼43 kDa proteins was attenuated by ∼27% and 72% , respectively ( p≤0 . 01 and p≤0 . 001 ) , and after 120 min ERK phosphorylation was almost completely blocked ( Figure 4A ) . Furthermore , adult worm proteins immunoprecipitated with anti-phospho p44/42 MAPK ( ERK1/2 ) ( Thr202/Tyr204 ) antibodies phosphorylated the ERK substrate Elk-1 ( Figure 4B ) , demonstrating that the detected ERK proteins display ERK activity . Collectively these findings , when combined with knowledge of molecular weights ( below ) and antibody recognition sites , are entirely consistent with the detected proteins being S . mansoni PKCs and ERK1/2 , with MEK acting upstream of ERK as in other organisms . Having determined that the immunoreactive proteins behave like PKCs and ERKs , it was possible to consider these proteins in relation to the S . mansoni genome data sets . Although the precise assignment of Smp identifiers to immunoreactive bands was beyond the scope of this study , antibody recognition sites support a tentative assignment to the PKC class ( Figure S3 ) . As we have previously reported [23] , the S . mansoni genome contains four annotated PKCs , two similar to human cPKCβI ( Smp_128480 , 75 . 6 kDa; Smp_176360 , 114 . 9 kDa ) , one to nPKCε ( Smp_131700 , 94 . 9 kDa ) and one to aPKCζ/ι ( Smp_096310 , 76 . 7 kDa ) , with PKCε also being similar to PKCη [4] . Of these molecules , the ∼78 kDa and ∼116 kDa phosphorylated PKCs detected by anti-phospho PKC ( Ser 660 ) antibodies are most likely the β-type cPKCs ( Smp_128480 and Smp_176360 ) , because the crucial Ser residue recognized by these antibodies is conserved in these Smps and because the remaining S . mansoni nPKCε/aPKCι proteins do not possess this Ser autophosphorylation site ( Figure 1 ) . Moreover , the ∼81 kDa band detected exclusively with anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies most likely corresponds to the atypical ι type PKC ( Smp_096310 ) ( Figure S3 ) as the Thr phosphorylation site ( Thr489 ) is conserved but , as with human PKCι , the Ser phosphorylation site is not ( Figure 1 ) . The lack of strong activation of this PKC by PMA ( Figure 3A ) further supports this finding , given that aPKCs do not respond directly to DAG/PMA . Available transcriptomic data for cercariae , schistosomules and adult worms [50] ( www . genedb . org ) also confirm that these PKCs are expressed in these life stages , with expression being developmentally regulated ( cf . Figure S3 ) . The large ∼132 kDa PKC-like protein consistently detected in adult worms and cercariae using anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies was activated by PMA and inhibited by GF109203X , confirming its PKC-like nature . Although absent from mammals , such high molecular weight PKCs are common in “lower” animals and , using similar antibodies , have been detected in the sea urchin Lytechnus pictus ( 135–140 kDa ) [51] and Caenorhabditis elegans ( 122 kDa ) [52] . PKC activation profiles were notably different among the four life stages studied , with four , two , and one PKC detected consistently in cercariae and adult worms , miracidia and mother sporocysts , respectively . This information suggests more complex roles for activated PKCs in the human host-infective/-dependent life stages . Wiest et al . [21] found nine-fold greater total PKC activity in adult S . mansoni than in larval stages , and it is established that PKC expression is developmentally regulated in other invertebrates [53] . We have shown that the ∼78 kDa β-type PKC of miracidia becomes inactive during development to the asexually reproducing , parasitic mother sporocyst stage and that PKC activity restricts this transformation [23] . Here , using anti-phospho ( pan ) ( ζ Thr410 ) antibodies , we see that the ∼81 kDa PKC is substantially activated in 48 h mother sporocysts compared to miracidia , highlighting a possible role for this PKC in asexual reproduction and signifying the dynamic nature of PKC activation during schistosome development . The ∼48 kDa and ∼43 kDa proteins detected with anti-phospho p44/42 MAPK antibodies likely correlate to genome sequences Smp_142050 ( 45 . 3 kDa , 70% identity to human ERK1 ) and Smp_047900 ( SmERK2 , 40 . 83 kDa , 68% identity to human ERK1 ) , due to molecular weight and antibody detection site similarity ( Figures 1 , 2C and S3 ) ; these proteins are also expressed in the requisite life-stages [50] ( www . genedb . org ) ( Figure S3 ) . The recognition site of the p44/42 MAPK antibodies was conserved among three other putative S . mansoni ERK-like proteins ( Smp_133490 , 58 . 7 kDa; Smp_133500 , 82 . 7 kDa; Smp_134260 , 70 kDa ) , however , phosphorylated ERK-like proteins of this size were not detected on blots . The ∼38 kDa protein detected in mother sporocysts and weakly in cercariae is possibly a non-specific band , because ∼38 kDa ERK-like proteins could not be detected in the S . mansoni genome . The ERKs also appear differentially activated in the different S . mansoni life stages studied ( Figure 2C ) , with ERK1/2 ( ∼48/∼43 kDa ) detected only in adult worms . No ERK activation was detected in miracidia , and only the ∼43 kDa ERK was phosphorylated ( using 12 µg protein ) in mother sporocysts and cercariae . Thus , the ∼48 kDa ERK in male worms may play a specific role in growth and development and/or host-interactions , particularly given that human EGF activates EGFR in S . mansoni [33] and influences ERK signalling and proliferation in other parasites , including Echinococcus multilocularis [54] and Trypanosoma cruzi [55] . The ERK pathway also interacts with the transforming growth factor β ( TGFβ ) pathway in schistosomes , possibly restricting interaction of SmSmad4 with receptor-activated Smad2 [56] . Therefore , because TGFβ signalling plays a part in mitotic activity , parasite development and egg embryogenesis [57] , [58] , ERK activity in particular cell types might suppress TGFβ signalling , with concomitant effects on development and reproduction . Cross talk between PKC and ERK signalling , with PKC upstream of ERK is common in many organisms , including invertebrates such as C . elegans [59] . To determine whether ERK and PKC signalling are connected in S . mansoni , ERK phosphorylation was determined after live adult worms were exposed to PKC modulators , and vice-versa . PMA ( 1 µM ) increased phosphorylation of the ∼43 kDa ERK ( p≤0 . 05; 1 . 8 fold ) after 30 min , however , the ∼48 kDa ERK was unaffected ( Figure 5A ) . This activation was blocked by treatment with GF109203X for 120 min prior to PMA stimulation , supporting ERK's dependency on PKC in response to PMA ( Figure 5A ) . Treatment with GF109203X alone did not inhibit the phosphorylation of either ERK after 120 min incubation ( Figure 5A ) , likely because this inhibitor only prevents activation of inactive PKC and because ERK may also be activated independently of PKC . On the other hand , exposure of worms to 1 µM U0126 for 60 or 120 min stimulated phosphorylation of the ∼116 kDa PKC ( p≤0 . 05; ∼1 . 9-fold ) , and of the ∼132 kDa PKC after 120 min ( p≤0 . 01; ∼2 . 5-fold ) ( Figure 5B ) . Interestingly , phosphorylation of the 78 kDa PKC was suppressed after 60 min U0126 treatment but recovered after 120 min ( Figure 5B ) . Collectively , these findings imply that PKC lies upstream of ERK in S . mansoni and suggest that complex temporally regulated feedback to PKC can occur in response to MEK inhibition . Identifying where cell signalling pathways in S . mansoni are predominantly activated is valuable in helping to unravel their function . Therefore , we performed functional mapping of phosphorylated PKC and ERK in freshly perfused , paired , intact S . mansoni adults using our recently published approaches [60] . We preferred this approach to in situ hybridization with gene specific probes , because we were interested in localizing functionally active kinases rather than protein expression per se . In all cases , negative control worms displayed minimal background staining ( Figure S1 ) . Staining with anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies revealed activated PKC associated with the tegument and tubercles of the tegument ( Figures 6A , 6D , 6F and 6G ) , particularly in the central dorsal-lateral region of males where tubercles are most prominent . In addition , activated PKC was associated with the oesophagus ( Figure 6B ) , oesophageal glands ( Figure 6C ) , and the uterus of the female in the vicinity of eggs ( Figure 6E ) . In males , deeper confocal scanning revealed PKC activation in the muscle layers , neural plexus , and myocytons ( Figures 6F–H ) . Worms labelled with anti-phospho PKC ( pan ) ( β Ser660 ) antibodies displayed generally weaker fluorescence; however , areas with notable PKC activation included the oesophagus ( Figure 6I ) , tubercles on the dorsal side of the male ( Figure 6J ) , and muscle layers under the tegument ( Figure 6K ) . Live adult worms were also exposed to 1 µM PMA for 30 min in an attempt to identify other regions displaying PKC activation . In these specimens , activation was evident in similar regions to those observed without PMA treatment; however , generally more activation was observed at the tegument surface ( Figure 6L ) , and evidence of tegument disruption was apparent in some specimens in response to PMA , as has been reported by other authors [61] . Activated PKC was also found to be associated with the lumen of the vitellaria , more extensive portions of the oesophagus and the muscular ventral sucker ( Figures 6M–O ) . Although , in a recent study , the transcription of the β-type PKC-encoding gene ( Smp_176360 ) was found to be upregulated ∼3 . 2 times in the testes of adult males compared to whole worms [62] , no significant PKC activation was seen in this tissue in the current study . In untreated male worms , ERK activation was particularly evident in excretory tubules and flame cells ( Figure 7A , 7B ) , the oesophagus ( Figure 7C ) and tubercles of the tegument ( Figure 7D ) . In females , regions displaying notable ERK activation included the seminal receptacle ( Figure 7E ) , uterus surrounding the egg during expulsion ( Figure 7F ) , ootype wall and the region of Mehlis' gland ( Figures 7G–H ) . Upon exposure to 1 µM PMA for 30 min , activated ERK was observed in regions similar to those of untreated samples; however , stronger immunoreactivity was apparent in the tegument and tubercles ( Figures 7I–K ) . Additionally , activated ERK was evident in the male ventral sucker ( Figures 7I–J ) , the testicular lobes ( Figure 7L ) , and ovary ( Figure 7M ) , possibly due to the ∼43 kDa ERK , as only this isotype responded to PMA ( Figure 5A ) . Interestingly , ERK1 ( Smp_142050 ) expression was recently found to be ∼2 . 7 and ∼3 . 9 times higher in the ovary and testes of adult female and male worms , respectively , when compared to whole worms [62] . Activated ERK has also been reported in the tegument of the cestode T . crassiceps [63] , where it is thought to interact with host factors . That PKC and ERK reside within the tubercles and tegument of S . mansoni and that cross talk many occur between these pathways is exciting , because the tegument of schistosomes is in constant contact with host blood . The tegument is a dynamic structure , where considerable host interplay is expected to take place [64] . EGFRs [65] and GPCRs [66] localize to the tegument and could activate PKCs and ERKs , possibly in response to host growth factors/hormones , which might influence schistosome development/survival . Localization of activated PKC to the musculature , neural plexus and myocytons supports a role for PKC in motor activity . Recently , we reported that activated protein kinase A ( PKA ) localizes to the central and peripheral nervous system of S . mansoni and that PKA activation results in hyperkinesia , demonstrating a role for PKA in neuromuscular co-ordination [60] . Activated PKA also localizes to the tegument , tubercles and canyons between the tubercles [60] . This finding , coupled with results described here , supports that complex signalling occurs at the schistosome surface . Finally , the localization of activated PKC/ERK to the oesophageal glands/oesophagus , uterus during egg expulsion , the ootype , Mehlis' gland , seminal receptacle and the excretory system highlights the likely importance of these kinases to schistosome reproduction and homeostasis . Increased pumping of the pharynx/oesophagus muscle during C . elegans starvation is dependent on ERK activity through PKC phosphorylation [67] . We hypothesize , therefore , that PKC/ERK regulate pumping of the schistosome oesophagus and that interference of these pathways may restrict blood feeding . To discover physiological roles for PKC and ERK in S . mansoni the effects of kinase modulation on adult worm phenotype were evaluated over 96 h . A range of inhibitor concentrations was used with doses chosen to reflect that whole , large , worms were exposed to each inhibitor rather than single cells . Five parameters were quantified: separation of worm pairs , male ventral sucker detachment ( in paired and unpaired males ) , egg output ( per couple ) , speed of movement ( velocity , mm/s ) of the worm posterior tip ( a proxy for muscular activity ) , and persistent coiling . In RPMI 1640 , worms remained paired releasing 103–110 eggs/day/couple during the first 72 h , decreasing to 93 thereafter ( data not shown ) . Males remained attached to culture plates during the first 24 h and only 5–7% detached between 24 h and 96 h . The worm posterior tip velocity of couples was 3 . 26 mm/s ( +/−0 . 33 mm/s ) after 1 h increasing to 3 . 68 mm/s ( <+/−0 . 55 mm/s ) thereafter . Extensive and/or sustained coiling was seldom observed . There were no significant differences between DMSO and RPMI 1640 control groups for these parameters . Worms displayed normal physiology ( Movie S1 ) . In contrast , U0126 , which was shown to attenuate ERK activation , increased worm unpairing with 50 µM causing 96% separation within 1 h ( p≤0 . 001 ) ( Figure 8A; Movies S2 and S3 ) . This coincided with rapid detachment of males from culture plates ( p≤0 . 001 ) ( Figure 8E , Movie S3 ) . Lower U0126 concentrations had more subtle but still significant effect; after 96 h , 26% ( p≤0 . 01 ) and 35 . 5% ( p≤0 . 001 ) of adult worms separated , and 44 . 7% ( p≤0 . 001 ) and 29% ( p≤0 . 01 ) of males detached from plates with 5 µM and 20 µM U0126 , respectively ( Figures 8A , 8E ) . With 20 µM U0126 egg output declined at 24 h but increased between 24 and 48 h ( p≤0 . 01 ) ( Figure 8C ) . This early suppression of oviposition is interesting given that activated ERK was observed in the seminal receptacle , uterus , ootype region and Mehlis' gland . Egg production essentially ceased at 50 µM U0126 with only a few ( 2–3 ) small immature eggs sporadically found , possibly a consequence of worm separation ( Figure 8C ) . U0126 also affected worm motion . At 1 h , 1 , 5 , 20 or 50 µM U0126 increased tip velocity ( when worms were paired ) by 265% ( to 9 . 37 mm/s; p≤0 . 001 ) , 203% ( to 7 . 62 mm/s; p≤0 . 01 ) , 251% ( to 8 . 87 mm/s; p≤0 . 001 ) and 285% ( to 10 . 66 mm/s ( p≤0 . 001 ) , respectively , compared to DMSO control speeds of 3 . 54 mm/s . However , movement decreased temporally and at 96 h velocities were between 3 . 17 and 5 . 85 mm/s ( Figure 9B ) . Because worms separated rapidly in 50 µM U0126 , tip velocities for the resultant unpaired worms were also determined for the remainder of the assay while in the inhibitor . These single males and females displayed high posterior tip velocities after 1 h with speeds of 9 . 74 and 8 . 5 mm/s , respectively ( Figure 9D , 9F ) ; thereafter , movement decreased and at 96 h males were 50% and females 90% slower . An uncoordinated ‘jerky’ behavior was observed at all U0126 doses ( Movie S3 ) . GF109203X inhibited worm pairing , egg output , and ventral sucker attachment ( Figure 8B , 8D , 8F ) and also induced extensive and persistent worm coiling ( Figure 10; Movie S4 ) . During the first hour , a proportion of worms remained paired at all GF109203X concentrations , and at >1 µM pairs coiled persistently which seemed to be caused by males , sometimes resulting in expulsion of female gut contents ( Movies S4 and S5 ) . After 1 h in 20 µM GF109203X , 94% of males had detached from plates ( p≤0 . 001 ) ( Figure 8F ) . Between 24 h and 96 h all worms separated in 50 µM GF109203X , with separation also observed at lower concentrations ( p≤0 . 001 ) ( Figure 8B ) . Egg output was also completely blocked within 24 h ( Figure 8D ) with 20 µM or 50 µM GF109203X ( p≤0 . 001 ) ( Figure 8D ) . This reduction in egg output was sustained over 96 h despite couples remaining . This could be due to PKC inhibition affecting uterine muscular movement as proposed for PKA [60] , or to direct effects on egg fertilization/formation . In the context of coiling , although worms displaying three coils or more were most prevalent during the initial 48 h GF109203X treatment ( Figure 10 ) worms had an increased propensity to coil throughout the experiment . For instance after 1 h in 5 , 20 or 50 µM GF109203X only 52% , 26% and 30% worms exhibited their normal shape , respectively ( p≤0 . 001 ) , with the remainder possessing one or more sustained coils ( Figure 10B ) . Worm tip velocity of couples was inhibited by 50 µM GF109203X only , to 1 . 9 mm/s ( p≤0 , 001 ) or 60% that of controls ( Figure 9A ) . However after worms separated in response to GF109203X movement reduced , with striking effects seen at 96 h ( p≤0 , 001 ) when the single males displayed velocities of just 0 . 06 mm/s and 0 . 09 mm/s , and the single females 0 . 3 mm/s and 0 . 17 mm/s for 20 µM and 50 µM GF109203X , respectively ( Figures 9C , 9E; Movie S4 ) . PMA ( 1 µM ) also induced rapid worm separation and only 10% remained paired at 24 h ( p≤0 . 001 ) and none at 48 h ( Figure 8B; Movie S6 ) . At least 75% of worms detached from plates after 24 h rising to over 97% after 96 h . Initially , the separated males showed reduced movement in PMA and separated female velocities were similar to those of remaining control pairs , however , over time the male worm tip velocities increased to 5 . 3 mm/s ( Figure 9C ) . No eggs were produced in PMA by worms ( Figure 8D ) , persistent coiling was absent and worms appeared rigid , although this became less prominent over time ( data not shown ) . Blair et al . [20] first alluded to a role for PKC in S . mansoni muscle contraction . Our detailed analyses substantiate this role , and link aberrant phenotypes induced by disrupted signalling to anatomical findings for activated PKC ( and ERK ) identified using functional kinase mapping . Extensive labelling of activated PKC in the muscle with activation also apparent in the neural plexus and myocytons is in accordance with effects of PKC modulation on worm motility , coiling and male ventral sucker detachment . PKCs , particularly PKCβ , mediate smooth muscle contraction in vertebrates [68] , [69] , and , in C . elegans , PKCα/PKCβ-like PKC2 isotypes are expressed in neurons and muscle cells [40] . In addition , PKC inhibitors block FMRF-amide induced muscle contraction in S . mansoni [70] . Despite having opposing effects on PKC phosphorylation , PMA and the higher doses of GF1090203X affected ventral sucker attachment and worm pairing similarly over 96 h ( Figures 8B , 8D , 8F ) and tip velocity within the initial 24 h ( Figures 9A , 9C , 9E ) ; however , PMA did not induce coiling . Although PMA activates PKC , overnight phorbol ester treatment is used to down-regulate cPKC expression in mammalian cells [71] . Therefore , we exposed live adult worm pairs to PMA ( 1 µM ) for 24 h . Western blotting with anti-phospho PKC ( pan ) ( ζ Thr410 ) and anti-phospho PKC ( pan ) ( βII Ser660 ) antibodies revealed that such treatment suppressed activation of PKCs to between 80% ( ±0 . 06%; p≤0 . 001 , n = 3; for the ∼132 KDa PKC ) and 64% ( ±0 . 1%; p≤0 . 05 , n = −3; 116 kDa PKC ) ( Figure S2 ) . Although we could not ascertain whether PKC protein expression was reduced , because we lacked an antibody that satisfactorily detects total PKC protein levels , the significant reduction in phosphorylation of treated worms demonstrates pathway deactivation , the critical parameter in terms of function . This information also suggests that decreased expression likely occurred . Therefore , the effects of PMA treatment on worm phenotype after 24 h might be expected to reflect those seen with GF109203X , which they do in part . However , because PMA causes tegumental disruption [61] stress-related effects on phenotype are possible . The role of ERK in schistosome muscle contraction requires further investigation . Whereas MEK inhibition increased motility of worm pairs , motility of single females reduced considerably over 72 h and was lower than that of single males ( Figures 9B , 9D , 9F ) . Nevertheless , that U0126 affected worm motility , attachment and pairing , highlights the ERK pathway , together with the PKC pathway , as an important regulator of schistosome muscle homeostasis . Because PZQ affects cellular calcium dynamics [72] , [73] and cPKCs are calcium responsive [19] we hypothesized that PZQ would affect the activation of S . mansoni PKCs and ERKs . After 15 min PZQ treatment , worm pairs became contracted and immobile and assumed a shrunken appearance ( data not shown ) ; coincident with this , the ∼78 , ∼81 , and ∼132 kDa PKCs detected with anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies showed increased phosphorylation ( p≤0 . 05; Figures 11A , 11B ) . Although the ∼116 kDa PKC detected by anti-phospho PKC ( pan ) ( βII Ser 660 ) antibodies was unaffected by PZQ at 15 min , increased phosphorylation was observed at 30 min and 120 min ( p≤0 . 001; Figures 11A , 11B ) . The ∼43 kDa ERK was also activated in response to PZQ treatment displaying a 2 . 9-fold increase in phosphorylation at 15 min ( p≤0 . 001; Figure 11A , 11B ) whereas the ∼48 kDa ERK was less affected . Confocal laser scanning microscopy revealed that PZQ-treated worms possessed strong ERK and PKC activation , with generally more signal in males than females ( Figures 11C–K ) . Activated ERK was associated with the tegument of paired male and female worms , male sub-tegumental regions , musculature , and cephalic ganglia on the male head ( Figures 11G–K ) . Anti-phospho PKC ( pan ) ( ζ Thr410 ) antibodies revealed activated PKC associated with tegumental and subtegumental regions , the oral sucker and oesophagus and structures along the dorsal side of male and female worms resembling collecting ducts ( Figures 11C–F ) which were also faintly seen in some PMA treatments ( data not shown ) . The pattern of activation revealed using anti-phospho PKC ( pan ) ( βII Ser 660 ) antibodies was similar that in the absence of PZQ , however , low levels of PKC activation were also seen in the testicular lobes ( data not shown ) . Despite decades of use the precise action of PZQ remains unresolved . Adult worms exposed to PZQ become paralyzed , lose ion homeostasis and take up calcium [72]–[74]; tegument disruption ensues with an associated exposure of worm antigens at the surface [75] . Mature males are more sensitive towards PZQ than females; additionally juvenile 28-day old worms , which are refractory to PZQ , take up large quantities of calcium , implying that PZQ sensitivity is not entirely due to calcium influx [39] . Possible targets of PZQ include a Cavβ ion channel subunit [76] , myosin light chain [77] and actin [78] . Sustained worm contraction typical of PZQ treatment was observed here when worms were exposed to PMA for 1 h ( Figure 9A ) ; thus , PZQs affect on worm motility is likely mediated through PKC and ERK . Interestingly , 28-day old worms exposed to PZQ for 20 h displayed a 6-fold increase in the gene coding for a suppressor of MEK [79]; this could serve to dampen PZQ-dependent ERK activation . Moreover , recently , PKC expression was found to be upregulated in adult male and female S . japonicum when exposed in vivo to a sub-lethal PZQ dose for 12 h or more , possibly in response to the increased calcium influx [80] . Therefore , whether the activation of PKC and/or ERK influences the outcome of worm survival in the face of PZQ treatment should be further investigated . To conclude , we have , for the first time , functionally characterised global signalling by PKCs and ERKs in adult S . mansoni and , using functional mapping , have localized PKC and ERK activities to numerous anatomical regions important to schistosome motor activity , excretion , reproduction and host-parasite interactions . Detailed phenotypic analysis has revealed that these kinases perform a role in adult worm movement , attachment , pairing and oviposition with pairing , worm tip velocity , ventral sucker attachment , and egg output reduced . Some of these outcomes ( e . g . , muscular movement and oviposition ) might be interconnected . Intracellular signal transduction is complex and signalling pathways are also interconnected . Therefore , while it is possible that some of the observed phenotypic effects might result from indirect effects of kinase inhibition on associated pathways/mechanisms , the results of this paper indicate that PKCs and ERKs play an important role . Nevertheless , a goal of future research would be to define the importance of any indirect mechanisms on these phenotypes . Exploiting existing anti-cancer drugs that target protein kinases for the control of human parasites is an attractive approach , one that has been proposed previously for schistosomes [16] . Ras-Raf-MEK-ERK signalling represents a potential target in cancer therapy and several small molecule inhibitors of this pathway are being tested in clinical trials [81] . The PKC pathway also represents a potential target [82] , [83] . Moreover , the insulin receptor , which signals through to PKC and ERK pathways has been proposed recently to be a viable vaccine target against schistosomes [84] , [85] . Therefore , given the effects of global PKC and ERK inhibition on schistosome homeostasis and egg production , we propose that these pathways , including specific isotype analysis and isotype knockdown by RNAi , are investigated further as they represent potential anti-schistosome drug targets . Currently , we are studying PKC and ERK pathways in schistosomules , a stage that is PZQ-insensitive , to appreciate signalling by these kinases during host-parasite interactions and during early schistosome development in the human host .
|
Parasitic blood flukes , also called schistosomes , cause human schistosomiasis , a neglected tropical disease and major public health problem in developing countries , especially sub-Saharan Africa . Sustainable control of schistosomiasis is difficult , mainly because the complex life cycle of the parasite involves a freshwater snail host , and the ability of the parasite to evade the immune response of the human host and to survive for many years . Little is yet known about the cellular mechanisms in schistosomes and how they regulate parasite homeostasis , development and behaviour . In this paper , the nature of intracellular signalling by protein kinases C ( PKCs ) and extracellular signal-regulated kinases ( ERKs ) in schistosomes is studied and these proteins are found to be vital for the coordination of processes fundamental to parasite survival , such as muscular activity and reproductive function . Our results contribute to an understanding of molecular events regulating schistosome function and identify PKCs and ERKs as possible targets for the development of new chemotherapeutic treatments against schistosomiasis .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
] |
[
"erk",
"signaling",
"cascade",
"protein",
"kinase",
"signaling",
"cascade",
"protein",
"kinase",
"c",
"signaling",
"signal",
"transduction",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"helminth",
"infections",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"ras",
"signaling",
"parasitic",
"diseases",
"molecular",
"cell",
"biology",
"zoology",
"cell",
"signaling",
"helminthology",
"signaling",
"cascades"
] |
2014
|
Protein Kinase C and Extracellular Signal-Regulated Kinase Regulate Movement, Attachment, Pairing and Egg Release in Schistosoma mansoni
|
Extracellular guidance cues steer axons towards their targets by eliciting morphological changes in the growth cone . A key part of this process is the asymmetric recruitment of the cytoplasmic scaffolding protein MIG-10 ( lamellipodin ) . MIG-10 is thought to asymmetrically promote outgrowth by inducing actin polymerization . However , the mechanism that links MIG-10 to actin polymerization is not known . We have identified the actin regulatory protein ABI-1 as a partner for MIG-10 that can mediate its outgrowth-promoting activity . The SH3 domain of ABI-1 binds to MIG-10 , and loss of function of either of these proteins causes similar axon guidance defects . Like MIG-10 , ABI-1 functions in both the attractive UNC-6 ( netrin ) pathway and the repulsive SLT-1 ( slit ) pathway . Dosage sensitive genetic interactions indicate that MIG-10 functions with ABI-1 and WVE-1 to mediate axon guidance . Epistasis analysis reveals that ABI-1 and WVE-1 function downstream of MIG-10 to mediate its outgrowth-promoting activity . Moreover , experiments with cultured mammalian cells suggest that the interaction between MIG-10 and ABI-1 mediates a conserved mechanism that promotes formation of lamellipodia . Together , these observations suggest that MIG-10 interacts with ABI-1 and WVE-1 to mediate the UNC-6 and SLT-1 guidance pathways .
Axons navigate to their targets in the developing nervous system by making a series of responses to extracellular guidance cues [1]–[6] . Several conserved families of guidance cues have been identified , including the netrins and slits . These guidance cues activate receptors on the growth cone at the tip of the growing axon , causing a directional response that steers the axon either towards or away from the source of guidance cue . A key component of the mechanism that drives the directional response to guidance cues is the asymmetric accumulation of f-actin within the growth cone . For instance , in vitro growth cone turning assays have shown that actin is asymmetrically polymerized to the side of the growth cone closest to a source of netrin , which is thought to cause the growth cone to turn towards the source of netrin [7] . Likewise , asymmetric actin distribution has also been observed in growth cones migrating in vivo [8] , [9] . Although many actin regulatory proteins have been implicated in the control of growth cone morphology , we do not understand how these proteins are coordinated to cause a directional response to guidance cues [10] . For example , the WVE-1 ( Wave ) complex activates the ARP2/3 actin-nucleating complex to control the formation of growth cone filopodia [11] , [12] . Furthermore , loss of function of WVE-1 or ARP2/3 components causes defects in axon guidance [11] , [13] , [14] . However , we do not know how the activity of this complex is controlled to give rise to the asymmetry that underlies growth cone guidance . In particular , it is difficult to understand how shallow gradients of guidance cues could be transformed into the sharply polarized outgrowth-promoting activity that is required for a directional response . MIG-10 may provide a key to understanding how guidance signals are transformed into sharply localized actin-based outgrowth activity . MIG-10 is a cytoplasmic outgrowth-promoting protein that becomes sharply localized in response to the UNC-6 guidance cue [15]–[19] . The role of MIG-10 in guidance has been studied in the HSN neuron of C . elegans , which extends an axon ventrally , towards a source of UNC-6 guidance cue . In response to UNC-6 , the UNC-40 ( DCC ) receptor becomes asymmetrically localized to the side of the cell closest to the source of UNC-6 . This in turn , leads to the asymmetric localization of MIG-10 to the side of the cell closest to the source of UNC-6 . MIG-10 has an outgrowth-promoting activity , thereby causing axon growth towards the source of UNC-6 . However , the mechanisms that mediate the outgrowth-promoting activity of MIG-10 are not understood . Although MIG-10 and its homolog lamellipodin are thought to play a major role in inducing actin polymerization , the mechanisms that mediate this effect are not known [20] . Knockdown of lamellipodin in fibroblasts results in a severe reduction in polymerized f-actin , with large areas devoid of the normal meshwork of f-actin . The Ena/VASP actin regulatory proteins can physically interact with lamellipodin . However , loss of all Ena/VASP function in fibroblasts does not produce the severe defects in actin polymerization that occur with knockdown of lamellipodin . Likewise , in C . elegans axon guidance , complete loss of Ena/VASP ( UNC-34 ) function results in axon guidance defects that are far weaker than those observed after complete loss of MIG-10 function [18] . Furthermore , complete loss of UNC-34 function does not reduce the outgrowth-promoting activity of MIG-10 [16] . Together , these observations indicate that the actin-polymerizing activity of lamellipodin and MIG-10 must be explained primarily through interaction with an effector other than UNC-34 ( Ena/VASP ) . Here , we present evidence that the outgrowth-promoting activity of MIG-10 is mediated through interactions with ABI-1 and WVE-1 . Furthermore , our genetic data indicate that MIG-10 functions with both ABI-1 and WVE-1 to mediate the UNC-6 and SLT-1 guidance signaling pathways . ABI-1 and WVE-1 are part of a well-characterized complex that promotes actin polymerization by activating the actin-nucleating activity of the ARP2/3 complex [21]–[26] . Thus , our observations suggest a model where MIG-10 interacts with ABI-1 and WVE-1 to direct actin polymerization in response to guidance cues .
MIG-10 is hypothesized to serve as a scaffold that can asymmetrically localize outgrowth-promoting proteins . However , the identities of these outgrowth-promoting proteins are unknown . We hypothesized that these outgrowth-promoting proteins are likely to bind to the polyproline motifs in MIG-10 . Therefore , we searched for proteins that include polyproline-binding domains ( SH3 , WW , and EVH1 domains ) for potential function in the MIG-10 pathway . To do this , we constructed an RNAi sublibrary for genes that encode polyproline-binding domains and screened for RNAi clones that phenocopy loss of mig-10 function in the HSN neuron ( Figure 1 ) . The HSN axon normally migrates ventrally to the ventral nerve cord ( Figure 1B ) . In mig-10 null mutants , 44±4 . 4% of the HSN axons migrate anteriorly before turning ventrally [18] . We observed this same phenotype in mig-10 ( RNAi ) animals , but with a penetrance of 21 . 6±4 . 6% ( see Figure 1C ) . From our screen , we identified abi-1 ( RNAi ) , which had a penetrance of 15 . 1±3 . 6% HSN guidance defects ( see Figure 1D ) . The migration of the HSN cell body was also affected by mig-10 ( RNAi ) and abi-1 ( RNAi ) . However , previous analysis has indicated that HSN axon guidance defects are not secondary consequences of defects in cell body migration [18] , [27] . Since ABI-1 has an SH3 domain , and MIG-10 has consensus-binding sites for SH3 domains , we tested for a physical interaction between the SH3 domain of ABI-1 and MIG-10 ( Figure 1E ) . We found that MIG-10 binds to the SH3 domain of ABI-1 fused to GST ( GST::ABI-1-SH3 ) . By contrast , MIG-10 did not bind to GST alone . Two concurrent studies have also found that MIG-10 can bind to full length ABI-1 , using yeast 2 hybrid and also co-immunoprecipitation [28] , [29] . Together , these observations identify ABI-1 as a potential member of the MIG-10 outgrowth-promoting complex . The AVM and PVM neurons are ideal for studying ventral guidance because their axons are guided ventrally by both attraction towards a source of UNC-6 guidance cue and by repulsion from a source of the SLT-1 guidance cue . Previous work with these neurons has indicated that MIG-10 functions in both the UNC-6 and SLT-1 signaling pathways [16] , [17] . In these experiments , null alleles were used to remove function of either the UNC-6 or SLT-1 guidance cues . Since unc-6; slt-1 double null mutants exhibit guidance defects that are nearly fully penetrant , UNC-6 and SLT-1 are thought to be the predominant guidance cues responsible for AVM and PVM axon guidance [16] , [17] , [30] , [31] . Therefore , the function of either guidance cue can be assayed by removing the function of the other cue . For example , a null mutation in mig-10 can enhance guidance defects in the unc-6 null mutant background , indicating that mig-10 functions in SLT-1 signaling . Likewise , a null mutation in mig-10 can also enhance defects in the slt-1 null mutant background , indicating that mig-10 functions in UNC-6 signaling . To determine if ABI-1 also functions in UNC-6 and SLT-1 signaling , we repeated these experiments with an abi-1 loss of function mutation . To determine if ABI-1 functions in the UNC-6 signaling pathway , we examined abi-1 ( tm494 ) ; slt-1 ( eh15 ) double mutants . The abi-1 ( tm494 ) mutation is a hypomorphic loss of function allele [32] , whereas the slt-1 ( eh15 ) is a null allele . In this slt-1 null mutant background , the AVM and PVM axons are guided by attraction towards UNC-6 . Thus , any enhancement of guidance defects in the abi-1; slt-1 double mutant would indicate that ABI-1 functions in the UNC-6 pathway . Indeed , we found that in abi-1; slt-1 double mutants , both AVM and PVM ventral guidance errors were enhanced relative to slt-1 single mutants , indicating that ABI-1 is involved in the UNC-6 attractive signaling pathway ( Figure 2A–2D ) . To determine if ABI-1 functions in the SLT-1 signaling pathway , we examined abi-1 ( tm494 ) ; unc-6 ( ev400 ) double mutants . In this unc-6 null mutant background , these axons are guided by repulsion from SLT-1 . Both AVM and PVM axon guidance errors were significantly enhanced by loss of abi-1 function in abi-1; unc-6 double mutants , indicating that ABI-1 is involved in the SLT-1 repulsive signaling pathway ( Figure 2A–2D ) . Despite being involved in both UNC-6 and SLT-1 signaling , we did not observe any AVM or PVM guidance defects in abi-1 single mutants , suggesting that ABI-1 , like MIG-10 , functions redundantly with other proteins to mediate UNC-6 and SLT-1 signaling . Although double mutant analysis suggests that the AVM and PVM axons are guided predominately by UNC-6 and SLT-1 , we can not exclude the possibility of a third guidance pathway . To determine if ABI-1 might function in a third pathway , we constructed an unc-6; slt-1; abi-1 triple mutant . Neither AVM nor PVM axon guidance defects were enhanced in the unc-6; slt-1; abi-1 triple mutant relative to the unc-6; slt-1 double mutant ( Figure 2E and Figure S1 ) . These observations do not support a role for ABI-1 in a third guidance pathway . To determine if ABI-1 functions cell autonomously to mediate the axon guidance , we conducted a transgenic rescue experiment in abi-1; slt-1 double mutants ( Figure 2F ) . The mec-4 promoter was used to drive expression of ABI-1 in the six touch neurons , including the PVM neuron . Transgenic expression of ABI-1 in the PVM neuron partially rescued the ventral guidance defects in the abi-1; slt-1 double mutants . By contrast , siblings that had lost the transgenic array did not show rescue of the ventral guidance defects . These observations indicate that ABI-1 functions cell autonomously to mediate the UNC-6 signaling pathway . Since we found that ABI-1 functions in the UNC-6 signaling pathway , we also wanted to determine if ABI-1 functions downstream of UNC-40 , the receptor for UNC-6 . To determine if ABI-1 functions downstream of UNC-40 , we used a mec-7::unc-40 transgene to create an UNC-40 gain of function phenotype . The mec-7::unc-40 transgene caused ventral axon guidance defects in the AVM and PVM neurons ( Figure 2G and Figure S2 ) . These guidance defects are suppressed by loss of abi-1 function in both the AVM and PVM neurons . These observations suggest that ABI-1 functions downstream of UNC-40 . The physical association between MIG-10 and ABI-1 suggests that they could function together to regulate axon guidance . To study genetic interactions between mig-10 and abi-1 , we used the HSN ventral axon migration , since single mutants in abi-1 or mig-10 produce guidance defects in this neuron . Homozygous mig-10 ( ct41 ) null mutants have guidance defects with a penetrance of 44±4 . 4% [18] . Homozygous abi-1 ( tm494 ) hypomorphic loss of function mutants have HSN guidance defects with a penetrance of 14±3 . 4% ( n = 100 ) . To ask if ABI-1 and MIG-10 function together , we used dosage sensitive genetic analysis ( Figure 3A ) . Both the abi-1 and mig-10 mutations were recessive , as neither mig-10 heterozygotes ( mig-10/+ ) nor abi-1 heterozygotes ( abi-1/+ ) had guidance errors in excess of wild-type animals . To test for a genetic interaction between mig-10 and abi-1 , we examined HSN axon guidance in animals transheterozygous for mutations in mig-10 and abi-1 , that is containing one mutant and one wild-type copy of each of these genes ( mig-10/+; abi-1/+ ) . These transheterozygous mutants had HSN guidance defects with a penetrance of 10 . 7±1 . 8% ( Figure 3A ) . Consistent with the physical association between MIG-10 and ABI-1 , these observations indicate that ABI-1 functions with MIG-10 to regulate axon guidance . Since mig-10 interacts genetically with abi-1 , we also wanted to test for interaction between mig-10 and wve-1 . Studies of the mammalian homologs of ABI-1 and WVE-1 , known as Abi1 and Wave , have indicated that Abi1 binds to Wave to enhance its ability to promote lamellipodial protrusion by activating the ARP2/3 complex to promote nucleation and branching of f-actin [21] , [22] , [24] , [25] . Likewise , genetic studies in C . elegans have indicated that ABI-1 and WVE-1 are required for subcellular enrichment of f-actin and for the formation of cellular protrusions during cell migration [23] . To determine if WVE-1 is involved in HSN ventral guidance , we examined homozygous wve-1 ( ok3308 ) mutants that had been maternally rescued . We found that these wve-1 mutants had HSN guidance defects with a penetrance of 13±3 . 3% ( n = 150 ) , suggesting that WVE-1 is involved in HSN ventral guidance . To determine if MIG-10 functions with WVE-1 , we tested for dosage-sensitive genetic interaction between mig-10 and wve-1 ( Figure 3B ) . Both wve-1 ( ok3308 ) and mig-10 ( ct41 ) are recessive , as neither mig-10 heterozygotes ( mig-10/+ ) nor wve-1 heterozygotes ( wve-1/+ ) had guidance errors in excess of wild-type animals . Animals transheterozygous for mig-10 ( ct41 ) and wve-1 ( ok3308 ) , ( mig-10 ) /+; wve-1/+ ) , had HSN guidance errors with a penetrance of 20 . 7±3 . 3% ( Figure 3B ) . We repeated this experiment with the wve-1 ( ne350 ) allele [23] and found that animals transheterozygous for mig-10 ( ct41 ) and wve-1 ( ne350 ) had HSN guidance errors with a penetrance of 18±2 . 7% ( n = 200 ) . Together , these results indicate that MIG-10 functions with ABI-1 to regulate guidance . Previous work has indicated that MIG-10 has an outgrowth-promoting activity and that the actin regulatory protein UNC-34 can interact with MIG-10 [16] , [17] . However , complete loss of UNC-34 function does not reduce the outgrowth-promoting activity of MIG-10 , indicating that UNC-34 does not account for MIG-10's outgrowth promoting activity . Since ABI-1 and WVE-1 are part of a complex that can promote lamellipodial protrusion , we asked if ABI-1 and WVE-1 can mediate the outgrowth-promoting activity of MIG-10 . To test this hypothesis , we determined if loss of WVE-1 function or loss of ABI-1 function could suppress the outgrowth-promoting activity of MIG-10 . The ALM neuron normally has a single axon that grows towards the anterior ( Figure 4A ) . Transgenic expression of MIG-10 in the ALM neuron causes the growth of a second posterior process ( Figure 4B ) . The growth of this second process is suppressed by loss of function mutations in abi-1 or wve-1 ( Figure 4C ) . By contrast , max-2 ( nv162 ) , a likely null mutation , had no effect on the growth of the second process . The lack of an effect of the max-2 mutation is expected because MAX-2 is thought to regulate axon guidance by functioning in a pathway that is parallel to MIG-10 [18] . Together , these observations indicate that ABI-1 and WVE-1 function downstream of MIG-10 to mediate its outgrowth-promoting activity . We also examined a role for ABI-1 in mediating the MIG-10 outgrowth-promoting activity in the AVM and PVM neurons . In these neurons , the outgrowth-promoting activity of MIG-10 can be oriented by either the UNC-6 or SLT-1 guidance cues [17] . Thus , transgenic expression of MIG-10 does not cause multipolar outgrowth in the wild-type , unc-6 null , or slt-1 null backgrounds . However , transgenic expression of MIG-10 does produce multipolar outgrowth in the unc-6; slt-1 double null background . We found that the abi-1 ( tm494 ) loss of function mutation significantly suppresses the MIG-10 outgrowth activity in the unc-6; slt-1 double null mutant background in the AVM and PVM neurons ( Figure 4D and Figure S3 ) . These results further support our conclusion that ABI-1 mediates the outgrowth-promoting activity of MIG-10 . Overexpression of C . elegans MIG-10 in cultured mammalian cells induces the formation of lamellipodia , suggesting that MIG-10 can interact with a conserved pathway to promote the formation of lamellipodia [17] . To determine if ABI-1 might be a part of that conserved pathway , we asked if the mammalian homolog of ABI-1 ( Abi1 ) is required for the lamellipodia-forming activity of MIG-10 in cultured mammalian cells . To address this question , we knocked down expression of mammalian ABI-1 ( Abi1 ) in HEK293 cells expressing MIG-10::GFP ( Figure 5 ) . Control cells expressing GFP have a round morphology with only minimal lamellipodia ( Figure 5A , 5D ) . Expression of MIG-10::GFP induces the formation of lamellipodia ( Figure 5B , 5D ) , which is significantly reduced by co-expression of an Abi1 shRNA ( Figure 5C , 5D ) . By contrast , co-expression of a scrambled control shRNA had no effect on lamellipodia formation ( Figure 5D ) . These results indicate that MIG-10 promotes the formation of lamellipodia in mammalian cells by functioning with a conserved pathway that includes mammalian Abi1 .
Several actin regulatory proteins have been implicated in the control of growth cone morphology [10] . A few of these have been implicated in the directional response to specific guidance cues . However , little is known about how actin regulatory proteins interact with one another to establish a directional response to guidance cues . Here , we uncover an interaction between the MIG-10 cytoplasmic scaffold and the ABI-1 actin regulatory protein . This interaction could help to explain how actin polymerization is spatially regulated during guidance , since MIG-10 is asymmetrically localized in response to guidance cues [15] , [18] . Concurrent work has indicated that the interaction between MIG-10 and ABI-1 can also regulate excretory canal morphogenesis and synapse formation [28] , [29] , Moreover , recent work has also shown that axon guidance cues and receptors are involved in regulating the WVE-1 complex during embryonic morphogenesis [33] . Together , these observations suggest that the interactions between axon guidance signaling components and the WVE-1 actin regulatory complex may be important in multiple aspects of actin-dependant developmental processes . Genetic studies of UNC-6 ( netrin ) signaling in C . elegans have revealed a direction-sensing module that includes the UNC-40 receptor , PI 3-Kinase , Rac and MIG-10 [15]–[19] . UNC-6 is secreted from ventrally localized cells and is thought to form a gradient that causes the HSN axon to migrate ventrally . In response to the UNC-6 gradient , UNC-40 becomes asymmetrically localized to the ventral side of the HSN cell . UNC-40 is thought to initiate signaling events that involve activation of Rac and production of PI ( 3 , 4 ) P2 by PI 3-Kinase . MIG-10 binds to both Rac and PI ( 3 , 4 ) P2 and thus becomes asymmetrically localized to the ventral side of the cell . This direction-sensing module is reminiscent of chemotaxis in neutrophils , where Rac and PI 3-Kinase are thought to form a positive feedback loop that transforms directional information from shallow gradients of chemotactic cues into sharply localized directional signal that promotes actin-based motility [34] , [35] . Likewise , we propose that UNC-40 , Rac , PI 3-Kinase and MIG-10 form a direction-sensing module that can transform a shallow gradient of UNC-6 into a sharply localized outgrowth-promoting activity . Our current results provide a link that connects MIG-10 to an actin polymerization-promoting complex , thereby explaining how the directional information encoded by asymmetric localization of MIG-10 can be transformed into directed axon outgrowth . We have found that MIG-10 interacts with ABI-1 and WVE-1 to mediate axon guidance . Previous work has indicated that ABI-1 binds to WVE-1 to promote its ability to activate the ARP2/3 complex [21] . The ARP2/3 complex can nucleate actin branches , thereby producing the meshwork of actin that is thought to provide the force that drives motility [36] . Therefore , the interaction between MIG-10 and ABI-1 can explain how MIG-10 is able to spatially direct outgrowth activity . We have been unable to visualize ABI-1::GFP in the HSN neuron . However , studies of Abi1 in mammalian cells indicate that it is located at the leading edge of migrating cells [37] . Since we have found that ABI-1 functions with MIG-10 , it is likely that ABI-1 localizes to the leading edge of the HSN neuron . Alternatively , ABI-1 might be localized throughout the cell . However , since MIG-10 is localized to the leading edge , the functional interaction between MIG-10 and ABI-1 would still be confined to the leading edge . Previous work has implicated ABI-1 and WVE-1 in axon guidance , but the guidance cues were not known [11] , [13] , [14] . Our current work indicates that ABI-1 and WVE-1 are involved in both the attractive UNC-6 signaling pathway and the repulsive SLT-1 signaling pathway . Despite being involved in both UNC-6 and SLT-1 signaling , the single abi-1 loss of function mutant does not have any defects in AVM or PVM guidance , suggesting that ABI-1 functions redundantly with other proteins to mediate guidance in these neurons . The lack of AVM and PVM guidance defects in abi-1 ( tm494 ) mutants might also be due to the fact that this is a hypomorphic allele [32] . In fact , recent work has indicated that RNAi depletion of GEX-3 , another member of the WVE-1 complex , can cause mild guidance defects in the AVM [33] . In our study , the role of ABI-1 in axon guidance is revealed in abi-1; unc-6 or abi-1; slt-1 double mutants , in which guidance information has been reduced to create a sensitized genetic background . Mutations in several other axon guidance genes ( including null alleles ) also give only weak or non-existent phenotypes as single mutants , but are enhanced by unc-6 or slt-1 null mutations . These mutations include mig-10 ( null ) , age-1 ( maternally rescued ) , unc-34 ( null ) , ced-10 ( hypomorphic ) , unc-115 ( null ) and madd-2 ( null ) 16–18 , 30 , 31 . Together , these observations suggest that guidance signaling functions with a great deal of redundancy in these neurons . Our results , when considered with previous findings , suggest that CED-10 , MIG-10 , ABI-1 and WVE-1 may be organized into a complex or complexes that features redundant physical interactions ( see Figure S4 ) . Our previous and current results suggest that CED-10 can interact with MIG-10 and that MIG-10 interacts with ABI-1 [18] . Previous studies have indicated that CED-10 can interact with a subcomplex that contains Sra1 ( GEX-2 ) and Nap1 ( GEX-3 ) [21] . This GEX-2/GEX-3 subcomplex interacts with ABI-1 . Thus , ABI-1 could be redundantly bound by both MIG-10 and the GEX-2/GEX-3 subcomplex . This redundant binding could occur within a single complex or could occur within separate complexes ( see Figure S4 ) . Discrimination between these two possible configurations will require biochemical and structural studies . We propose that guidance signaling molecules may be organized into networks that include redundant physical interactions . This redundancy could provide a more robust mechanism for the control of the actin polymerization . This model is consistent with the observation that mutations in genes that encode guidance signaling proteins ( such as mig-10 , age-1 , unc-34 , ced-10 , unc-115 and madd-2 ) , generally lead to only weak or nonexistent guidance defects [16]–[18] , [30] , [31] . In our study , ABI-1 is shown to act in both the UNC-6 and SLT-1 pathways . Several other intracellular outgrowth-promoting proteins have also been implicated in both attractive and repulsive signaling pathways including MIG-10 , Rac , Pak , UNC-34 and Abl [16] , [17] , [38]–[41] . In addition , overexpression of the repulsive UNC-5 receptor can promote neurite outgrowth in cultured cells [42] . Together , these observations suggest that a common set of outgrowth-promoting proteins are involved in both attractive and repulsive responses . A likely explanation for the dual roles of outgrowth-promoting proteins is that they could be oriented by both attractive and repulsive signals , thereby promoting growth towards or away from the source of guidance cue , respectively . Thus , the difference between attraction and repulsion could be in where and how a common set of outgrowth-promoting proteins are localized . Finally , ABI-1 may link to multiple upstream signaling modules to mediate distinct aspects of axon growth . For instance , recent work has found that the UNC-53 scaffold protein interacts with ABI-1 to promote longitudinal axon growth in C . elegans [32] . In unc-53 loss of function mutants , longitudinal , but not circumferential axon growth is disrupted . Conversely , MIG-10 is thought to mediate circumferential , but not longitudinal axon growth [16] , [17] . These observations suggest that the mechanisms that control actin polymerization to drive longitudinal outgrowth are distinct from those that control actin polymerization to drive circumferential guidance . We propose that for the process of circumferential axon guidance , ABI-1 links to the MIG-10 scaffold . By contrast , for the process of longitudinal axon extension , ABI-1 links to the UNC-53 scaffold . Therefore , each of these distinct upstream regulatory processes can both utilize the same actin regulatory complex .
AGC1: slt-1 ( eh15 ) ; abi-1 ( tm494 ) ; cueEx1 , AGC2: slt-1 ( eh15 ) ; abi-1 ( tm494 ) cueEx2 , AGC3: abi-1 ( tm494 ) ; cueIs3 , AGC4: cueIs3 , AGC5: cueIs3; wve-1 ( ok3308 ) /hT2 [bli4 ( e937 ) ; let- ? ( q782 ) ; qIs48] , AGC6: slt-1 ( eh15 ) ; abi-1 ( tm494 ) cueEx4 , AGC8: wve-1 ( ok3308 ) /hT2 [bli4 ( e937 ) ; let- ? ( q782 ) ; qIs48]; kyIs262 , AGC9: mig-10 ( ct41 ) /hT2 [bli4 ( e937 ) ; let- ? ( q782 ) ; qIs48]; kyIs262 , AGC10: mig-10 ( ct41 ) /hT2 [bli4 ( e937 ) ; let- ? ( q782 ) ; qIs48]; wve-1 ( ok3308 ) /hT2; kyIs262 , AGC12: abi-1 ( tm494 ) ; slt-1 ( eh15 ) ; zdIs5 , AGC13: abi-1 ( tm494 ) ; unc-6 ( ev400 ) ; zdis5 , AGC14: abi-1 ( tm494 ) ; zdIs5 , AGC15: cueIs3; max-2 ( nv162 ) , AGC16: wve-1 ( ne350 ) /hT2; mig-10 ( ct41 ) /hT2 , AGC18: unc-6 ( ev400 ) ; slt-1 ( eh15 ) ; abi-1 ( tm494 ) ; zdIs5 , AGC19: unc-6 ( ev400 ) ; slt-1 ( eh15 ) ; abi-1 ( tm494 ) ; urEx305 . AGC20: cueIs7; abi-1 ( tm494 ) ; zdIs5 , AGC21: cueIs7; zdIs5 . The wve-1 ( ne350 ) allele was a gift from Martha Soto [23] , [43] . The wve-1 ( ok3308 ) allele was obtained from the CGC and out-crossed 3 times . We found that the ok3308 mutation behaved as a zygotic sterile . The abi-1 ( tm494 ) hypomorphic allele was provided by Shohei Mitani . cueEx1 and cueEx2 [mec-4::abi-1; odr-1::dsred] were created by injecting pAGC2 at 25 ng/ul and odr-1::dsred at 50 ng/ul . urEx305 [mec-4::mig-10a; flp-20::gfp] was created as described previously [17] . cueIs3 [mec-4::mig-10a; flp-20::gfp] was created by integrating urEx305 with a gamma radiation source . cueEx4 [mec-7::abi-1; odr-1::dsred] was created by injecting pAGC3 at 25 ng/ul and odr-1::dsred at 50 ng/ul . kyIs262 [unc-86::myrgfp] was kindly provided by Cori Bargmann . zdIs5 [mec-4::gfp] was obtained from the CGC . evEx344 [mec-7::unc-40; unc-129::gfp] was a gift from Joseph Culotti . cueIs7 was created by integrating evEx344 . pAGC2 contains the mec-4::abi-1 and was created by amplifying the abi-1 cDNA using the following primers fwd: gcagcagctagccaccatgagtgttaatgatcttcaag and rev: gcagcaggtacctcatactggaactacgtagtttc . The PCR product was cut with NheI and KpnI and ligated into the Nhe1 and Kpn1 sites of pIM207 [17] , [44] . pAGC3 contains mec-7::abi-1 and was created by using Nhe1 and Kpn1 to subclone the abi-1 cDNA from pAGC2 into pIM211 [17] . pAGC4 contains GST::abi-1-SH3 and was created by using the following primers to amplify DNA encoding the SH3 domain of ABI-1: fwd:gccacaagcgatccagtctctttgatacgagtgct and rev: gccacaaggaattctcatactggaactacgtagtt . The PCR product was cut with BamHI and EcoRI and cloned into the same sites of pGEX-2T . PAV197 contains an shRNA construct for Abi1 and PAV16 contains a scrambled control shRNA . Both of these constructs were kindly provided by Giorgio Scita [21] . GST binding assays were performed as described previously [45] . Briefly , a GST fusion with the SH3 domain of ABI-1 ( GST::ABI-1-SH3 ) was prepared by transforming BL21 ( DE3 ) cells with pAGC4 . Cell lysates containing MIG-10::GFP were prepared by transfecting HEK293 cells with a plasmid encoding MIG-10::GFP and lysing cells 24 hours later . The GST::ABI-1-SH3 fusion protein was purified and coupled to glutathione-sepharose . Cell lysates were added to the glutathione-GST::ABI-1-SH3 complex and incubated for 2 hours . The glutathione-sepharose-protein complexes were washed three times with 0 . 1% Triton X-100 , boiled in loading buffer and run on and SDS PAGE gel . Bound proteins were detected with an antibody against GFP . HEK293 cells were grown in DMEM with 10% FBS . Fugene 6 was used to transfect or co-transfect cells with the appropriate plasmids . For co-transfection experiments 2 µg of DNA encoding the appropriate shRNA was mixed 1 µg of DNA encoding GFP or MIG-10::GFP . Cells were transfected in plastic dishes and allowed to grow for 72 hours after transfection . Next , cells were removed from the plastic dishes and were replated onto glass coverslips coated with polylysine . After 24 hours , the cells were fixed with paraformaldehyde and mounted for observation and analysis . For analysis of axon guidance phenotypes , animals were mounted on a 5% agarose pad and observed with a 40× objective . For AVM and PVM ventral guidance , an axon was scored as defective if it failed to reach the ventral nerve cord . For HSN ventral guidance , an axon was scored as defective if it traveled laterally for a distance equivalent to 2 cell bodies or more prior to migrating ventrally . For analysis of multipolarity in the ALM , AVM and PVM neurons , a cell was scored as defective if it had a posterior axon that was longer than 2 cell bodies .
|
To form neural circuits , axons must navigate through the developing nervous system to reach their correct targets . Axon navigation is led by the growth cone , a structure at the tip of the growing axon that responds to extracellular guidance cues . Many of these guidance cues and their receptors have been identified . However , much less is known about the internal signaling events that give rise to the structural changes required for growth cone steering . A key component of the internal response is MIG-10 , a protein that becomes asymmetrically localized in response to the extracellular cues . MIG-10 is thought to serve as a scaffold that can spatially control outgrowth-promoting proteins within the growth cone . However , we do not know the identity of the outgrowth-promoting proteins that associate with MIG-10 . Here we report that MIG-10 associates physically with the actin regulatory protein ABI-1 . We present genetic evidence indicating that ABI-1 functions downstream of MIG-10 to mediate its outgrowth-promoting activity . Additional genetic evidence indicates that these proteins function in both attractive and repulsive guidance signaling pathways . We also present evidence suggesting that the connection between MIG-10 and ABI-1 represents a phylogenetically conserved mechanism for the control of cellular outgrowth .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"developmental",
"neuroscience",
"axon",
"guidance",
"biology",
"morphogenesis",
"neuroscience"
] |
2012
|
MIG-10 Functions with ABI-1 to Mediate the UNC-6 and SLT-1 Axon Guidance Signaling Pathways
|
Mutations in LACERATA ( LCR ) , FIDDLEHEAD ( FDH ) , and BODYGUARD ( BDG ) cause a complex developmental syndrome that is consistent with an important role for these Arabidopsis genes in cuticle biogenesis . The genesis of their pleiotropic phenotypes is , however , poorly understood . We provide evidence that neither distorted depositions of cutin , nor deficiencies in the chemical composition of cuticular lipids , account for these features , instead suggesting that the mutants alleviate the functional disorder of the cuticle by reinforcing their defenses . To better understand how plants adapt to these mutations , we performed a genome-wide gene expression analysis . We found that apparent compensatory transcriptional responses in these mutants involve the induction of wax , cutin , cell wall , and defense genes . To gain greater insight into the mechanism by which cuticular mutations trigger this response in the plants , we performed an overlap meta-analysis , which is termed MASTA ( MicroArray overlap Search Tool and Analysis ) , of differentially expressed genes . This suggested that different cell integrity pathways are recruited in cesA cellulose synthase and cuticular mutants . Using MASTA for an in silico suppressor/enhancer screen , we identified SERRATE ( SE ) , which encodes a protein of RNA–processing multi-protein complexes , as a likely enhancer . In confirmation of this notion , the se lcr and se bdg double mutants eradicate severe leaf deformations as well as the organ fusions that are typical of lcr and bdg and other cuticular mutants . Also , lcr does not confer resistance to Botrytis cinerea in a se mutant background . We propose that there is a role for SERRATE-mediated RNA signaling in the cuticle integrity pathway .
The ability to maintain the barrier properties of the epidermis , which covers the aerial surface of higher plants , is largely due to their outermost cell walls which are impregnated and covered with specialized lipids . The fine structure and composition of this complex layer , called the cuticle , has been the subject of numerous studies [1] , [2] . The innermost periclinal layer of the cuticle is , in fact , a cutinized portion of the epidermal cell wall , in which cell wall polysaccharides are perhaps cross-linked to phenolics and aliphatic components of the cutin . The presence of phenolics , which may contribute to the barrier function , is evident in this layer through fluorescence microscopy and chemical analysis [3] . Under transmission electron microscopy , however , this cutinized layer of the cell wall is often heterogeneous in appearance and penetrated by tufts of fibrillar material , or is sometimes barely visible . This is in contrast to the opaque stripe of the continuous cuticle proper which , at a higher resolution , often appears to be finely lamellated; it is composed of polyester cutin and non-hydrolyzable polymer cutan and is , essentially , free of cell wall polysaccharides . In the early stages , in particular , it may have a pectinaceous under-layer . Wax forms the outermost structural layer , although a certain amount of it ( intracuticular wax ) permeates the interior of the cuticle . Individual plant species display some variations of this standard pattern . Waxes and cutin represent two groups of cuticular lipids that are supposed to be primarily responsible for the barrier function of the plant epidermis . The structural , biochemical , biophysical and molecular genetic aspects of cuticular lipids and their role in the defense against pathogens have been reviewed elsewhere [2] , [4]–[10] . Plant growth and development demand that the cuticle changes continuously and also maintains the balance between rigidity and flexibility . During the early stages of epidermal development , cells are covered with osmiophilic amorphous procuticle , but the lamellar structure of the cuticle proper and the reticulate fibrillar pattern of the cutinized portion of the cell wall may become distinguishable as the cuticle forms [2] . Chemical and structural changes in the cuticle may be beneficial , both in terms of the adaptation to fluctuating environmental conditions and in response to various stresses . However , this view of the cuticle as a dynamic structure has yet to be supported by extensive data . Recent molecular , genetic studies of cuticular mutants have lead to the identification and characterization of a number of genes involved in various aspects of cuticle formation ( reviewed in [5] , [6] , [9] , [11] ) . A group of Arabidopsis mutants , including fiddlehead ( fdh ) , lacerata ( lcr ) and bodyguard ( bdg ) , which are thought to be defective in the biosynthesis of cuticular polyesters , reveals secondary phenotypes which include drastic changes in cell differentiation , plant architecture , organ morphology , pathogen resistance and other elements . This suggests that there exists a pathway that is not only essential for cuticle formation , but may , directly or indirectly , control various cellular processes as well [9] . One of these is the adhesive interaction between the epidermal cells of different organs [12] , [13] . No mechanism has yet been demonstrated which might account for the association between the series of phenotypes in the cuticular mutants . Studies of one of these mutants , bdg , which exhibits defects that are characteristic of the loss of cuticle structure , paradoxically revealed that this cuticular mutation accumulates significantly more cutin monomers in the residual cell-wall bound lipids , and more wax [14] . It was , therefore , suggested that plants are capable of repairing cuticular perturbations and re-establishing cuticle homeostasis [14] . Fdh and lcr , like bdg , belong to a class of cuticular mutants that are characterized by secondary phenotypes which include misshapen cells and organs and epidermal fusions [14]–[16] . Their cuticular phenotypes were not , however , examined in detail . Herein , we report that both lcr and fdh display an increase in cutin and wax constituents and provide insight into the structural aspects of their cuticle . Using a microarray-based transcriptome analysis , we demonstrate the transcriptional upregulation of wax , cutin , and cell wall and defense genes in lcr , bdg and fdh . We propose that this is a compensatory adaptive response , representing a part of a cell-wall integrity maintenance mechanism . To compare the responses induced by mutations in cuticular and cellulose synthase genes , we used the meta-analytical method MASTA ( MicroArray overlap Search Tool and Analysis ) , which has recently been developed in our lab . When utilizing it for in silico suppressor/enhancer screens , we identified SERRATE and confirmed that it is required for organ fusions in cuticular mutants . This raises the interesting possibility that there may be a connection between the cuticle formation and morphogenesis .
As do other cuticular mutations of this kind , lcr has a pleiotropic effect on plant development , which affects leaf morphology , cell morphogenesis and differentiation , shoot branching , and senescence [16] . At the rosette stage , lcr plants are easily distinguishable from wild types , but not from bdg and fdh ( Figure 1A ) , by severe deformations of leaves and leaf fusions . When compared to wild type , the staining of rosette leaves with the water-soluble dye toluidine blue ( TB ) resulted in fdh , lcr , and bdg having heterogeneous , patchy patterns ( Figure 1B ) , showing the defects of the cuticle . However , with regard to the intensity of staining , neither the fused or unfused rosette leaves of fdh , lcr and bdg were distinguishable from each other . As direct estimation of the cuticle permeability is not feasible in Arabidopsis , to extend these results , we performed an assay which measures chlorophyll leaching into alcohol [13] . As expected , the leaves of all three mutants lost chlorophyll faster than wild type when immersed in 80% ethanol ( Figure 1C ) , thus corroborating the results of the TB staining . However , whereas lcr and bdg appeared to be very similar to each other , fdh released chlorophyll much more quickly ( Figure 1C ) : after 20 min of incubation , fdh lost about 60% of total chlorophyll , while lcr and bdg only lost about 20% . To study whether there is both a correlation between cuticle permeability to chlorophyll and engagement in ectopic organ fusions , in this assay we examined a sample taken from lcr rosette leaves which were not joined in a fusion . Figure 1C shows that these leaves lose the pigment faster than the wild type control does , but this was still slower than the representative lcr sample ( comprising leaves joined in a fusion and leaves not joined in a fusion ) . This suggests that both features of the polymorphic lcr phenotype are linked . To investigate whether the expression of LCR is restricted to the epidermis , we fused the putative 5′ regulatory regions of the LCR gene with the green fluorescent protein ( GFP ) reporter gene . The expression of LCR:GFP was then studied in transgenic Arabidopsis plants by confocal scanning laser microscopy , and was found to be limited to the epidermal cells of leaves , stems , sepals , petals , style , stigma and ovules ( Figure 1D–1G ) . The expression of LCR in organ primordia was reminiscent of that of FDH and BDG , which have previously been studied in detail by using GFP fusions [14] , [17] . Because LCR belongs to the CYP86A P450 gene subfamily , which includes closely related and highly conserved gene sequences in Arabidopsis , we were not able to design an LCR specific probe which would be long enough . We attempted to concatenate LCR specific sequence motifs , but the resulting probes also failed to yield a consistent in situ hybridization signal ( data not shown ) . However , our results with the LCR:GFP plants support the microarray hybridization data which had suggested that LCR might be the epidermis specific gene in the stem [18] . Collectively , these results back-up the contention that lcr is a typical cuticular gene . To determine whether the lcr mutation distorts or disrupts the cuticle , we examined the epidermis of aerial organs in lcr by transmission electron microscopy ( TEM ) . In leaves and petioles of wild type plants , cutin deposition in the epidermal cell wall forms a regular membranous structure ( called the cuticle proper ) on the outer side ( Figure 2A , 2C , 2G , 2L ) . When viewed under TEM , this electron-dense layer was not only discontinuous and deformed in lcr , but was also characterized by the irregular deposition of multi-layered , electron-dense , sharply outlined material , as well as the presence of empty spaces within the deeper layers of the cell wall ( Figure 2D , 2E , 2H–2K , 2M–2O ) . The presence of empty cavities ( Figure 2H and 2N ) and the over-deposition of an electron-dense material close to the cell wall surface ( Figure 2D , 2H , 2K ) , indicate infiltration , or bursting , of the cell wall materials through the defective cuticle proper in this mutant ( Figure 2O ) , thereby leading to cutin juxtaposition in the inner layers ( Figure 2H , 2M , 2N ) . Supernumerary layers of cutin-like materials may lead to the conclusion that lcr does not suffer from a lack of cutin , but rather from the structural dysfunction of its cuticle , even though a cutin overlay was very thin in many instances ( Figure 2D , 2K , 2N ) . In some cases , electron-opaque material seemed to crystallize inside the lcr cell wall , giving its cuticle a composite appearance ( Figure 2E , 2K , 2O ) . Although many features make the lcr cuticle resemble that in bdg [14] and the CUTE plants [19] , this has not yet been observed in any other mutant , and appears to be characteristic of lcr , which exhibited an extraordinarily irregular cuticle . In various fusion zones in lcr , the cell walls often seemed to be merged , with little or no trace of the intrinsic cuticular membrane , although some inclusions of electron-dense material could be found in areas where cell walls have not been completely fused ( Figure S1A , S1B ) . When exposed to mechanical tension , fused organs might separate , only remaining connected by a fine thread ( Figure S1C ) . The outer cell wall of epidermal cells in lcr was , generally , not as regular as in wild type plants , with severe deformations and some darker stripes giving it a plastic appearance ( Figure 2H–2J ) . This leads to speculation that lcr has a cell wall phenotype . From the results of the chlorophyll leaching measurements , it might have been expected that fdh would display a highly disorganized cuticle , resembling that in lcr . However , Lolle and co-workers reported that the cuticle could always be detected in the fusion zones separating different organs , and three lipid stains failed to detect any differences between fdh-1 and wild-type tissues [12] , [13] . The fdh-1 allele has been isolated in the Landsberg erecta ( Ler ) genetic background . Since the lcr transposon insertion allele used in this study was found in the Columbia ( Col ) ecotype , we sought to examine the cuticle in an fdh allele with the same genetic background . We made the decision to characterize fdh-3940S1 [20] , which has also been used in the chlorophyll leaching assays described above . The extensive investigation under TEM did not reveal any visible ultrastructural changes in the fdh-3940S1 cuticle ( hereafter referred to as fdh ) in different organs when compared to wild-type . This makes fdh different to the lcr , bdg and hth mutants and the CUTE plants [19] . A typical , continuous electron-dense layer was found to be deposited in the epidermal cell wall in fdh leaves ( Figure 2B , 2F ) and in the fusion zones ( Figure S1 ) , showing that two knock-out alleles of fdh behave similarly in the Ler and Col genetic backgrounds . While the cuticle proper in fdh showed no major ultrastructural changes that were detectable by TEM , some images gave the impression that its surface may have a rather more diffused appearance , lacking sharp margins . These results imply that structural defects in the cuticle proper , which are detectable under TEM , may not account for the molecular sieving properties of the cuticle that were estimated by molecular leaching assays . This also suggests that the organ fusions observed in fdh , lcr , bdg and some other cuticular mutants are not a direct consequence of major structural changes in their cuticles , thereby calling into question the conventional view that the bare cell walls of epidermal cells interact to produce a fusion . One of the cuticle's primary roles is to act as a protective barrier against pathogen attack [10] . It has , however , been reported that some Arabidopsis cuticular mutants , such as lacs2 , lcr , bdg and transgenic CUTE plants , are , in fact , more resistant to the major necrotroph Botrytis cinerea than the wild type [21]–[23] . The reasons for this paradoxical resistance remained unclear . The easier diffusion of a plant-produced toxin through the mutant cuticles was , however , considered to be one of the possible mechanisms of this resistance [21] , suggesting that the highly permeable cuticle would greatly increase the resistance of fdh to this pathogen . To test this concept , we compared the wild type and three mutants by the detached-leaf assay which appeared to have the best consistency [24] . As controls , we used the B . cinerea-resistant lacs2 and the hypersusceptible phytoalexin deficient3 ( pad3 ) which is impaired in the accumulation of the antifungal metabolite camalexin [25] . The level of susceptibility after droplet inoculation was measured as the percentage of lesions larger than the original inoculation site ( percentage of outgrowing lesions ) and as the average lesion area ( Figure 3 ) . Three days post-inoculation ( dpi ) with B . cinerea ( strain 2100 in 1/4 PDB ) , lacs2 and bdg showed lower susceptibility ( P<0 . 0001 ) , with 16% and 18% of outgrowing lesions , respectively , as compared to 72% in the wild type , and 37% and 42% in lcr and fdh , respectively ( Figure 3B ) . Consistent with the described hypersusceptibility to B . cinerea , 96% of outgrowing lesions were identified in pad3 ( Figure 3B ) . The higher resistance phenotype of cuticular mutants was revealed by both the lower proportion of outgrowing lesions and lesion area , the latter being on average very similar in all cuticular mutants and always significantly smaller when compared to the wild type ( Figure 3C ) . One interesting observation is that , under our experimental conditions , some leaves of bdg and lacs2 remained free of disease symptoms at 3 dpi ( ∼27% and ∼56% respectively ) , whereas all inoculated leaves of lcr and fdh showed signs of fungal infection . Comparable results were obtained with the B . cinerea strain B05 . 10 ( data not shown ) , corroborating the view that FDH deficiency afforded similar protection against B . cinerea as that observed in other cuticular mutants . However fdh does not seem to be more resistant than bdg and lcr , implying that the resistance phenotype does not correlate well with the cuticular permeability when measured by the chlorophyll leaching rate and the TB stainability . Although lcr and fdh were recognized as classical cuticular mutants revealing enhanced epidermal permeability ( Figure 1B and 1C ) [13] , [16] , [21] , [26] , the chemical compositions of their cuticular lipids had not been characterized in detail . LCR , which functions as a fatty acid ω-hydroxylase when expressed in yeast , was proposed as being active in the biosynthesis of the ω-hydroxy and α , ω-dicarboxy fatty acids that are major cutin monomers in Arabidopsis [16] , [27] . The loss of the LCR function is presumed to reduce the accumulation of the respective cutin monomers . Based on the amino acid sequence similarity , it has been suggested that FDH encodes a β-ketoacyl-CoA synthase ( KCS ) which is involved in microsomal fatty acid elongation [20] , [28] . In FDH-deficient plants , cutin monomers could , therefore , potentially either decrease or comprise shorter chain monocarboxylic and dicarboxylic fatty acids . To determine the effects of fdh and lcr mutations on the chemical composition of cuticular polyesters , we analyzed residual-bound lipids in leaves ( Figure 4A ) . This approach gives a good approximation of the monomer composition of pure cutin , which is difficult to isolate in sufficient amounts in Arabidopsis [29] . This analysis was conducted twice on both mutants , using plant material grown under similar greenhouse conditions in different seasons ( Figure 4A and Figure S2 ) . Both experiments yielded similar results , with lcr and fdh accumulating higher levels of the C18:2 α , ω-diacid which is a major cutin component in Arabidopsis [29] , [30] . The increase , when compared to wild type , was approximately three times and twice in lcr and fdh , respectively ( Figure 4A ) . Remarkably , the levels of C18:2 ω-hydroxy acid , which is a precursor to C18:2 α , ω-diacid , also increased two-fold in both mutants . Both two and one and half times differences were also found for the C18:3 ω-hydroxy acid in lcr and fdh , respectively . It is also worth noting that both accumulated approximately less than twice the C18:2 acid in residual-bound lipids . The partially ( by ∼25–40% ) reduced content of the C16:0 acyl chains in the three classes of fatty monomers ( acids , ω-hydroxy acids , α , ω-diacids ) , might be evidence of an enhanced C16 elongation in the mutants . Examination of the cutin composition analysis data ( Figure 4A and Figure S2 ) revealed that no shift towards shorter carbon chains could be detected in fdh . Moreover , no consistent decrease in the levels of ω-hydroxy fatty acids and α , ω-diacids could be detected in lcr . Given the increase in the levels of C18 ω-hydroxy fatty acids and α , ω-diacids , it could be proposed that the major changes in cuticular lipids observed in lcr and fdh are not due to the deficiency caused by the respective mutations . Instead , it could be evidence of an induced response to these mutations , which leads to the incorporation of more cutin-like material in the outer epidermal cell wall of the mutants . This response could play a compensatory role which contributes to the survival of the mutant plants . It is particularly noteworthy in this context that lcr and fdh appear to possess remarkably similar cutin monomer compositions . There is a strong line of evidence linking the over-accumulation of epi- and intra-cuticular wax to the cuticular deficiency in bdg [14] . We , therefore , analyzed leaf wax composition in lcr and fdh ( Figure 4B ) , revealing that the total amount of wax had been increased two-fold and three-fold in lcr and fdh , respectively; on average , wild-type leaves had a wax load of 0 . 72±0 . 07 µg/cm2 compared to 1 . 56±0 . 25 µg/cm2 and 2 . 66±0 . 25 µg/cm2 in lcr and fdh , again respectively . The major constituents of wax , which are alkanes in Arabidopsis , were increased from 2 . 3 times ( C33 in lcr ) to 9 times ( C27 in fdh ) . The levels of free fatty acids and alcohols had also increased , but to a lesser extent , with up to a 3 . 3-fold increase in the C32 fatty acid ( Figure 4B ) . The marked difference between the two mutants was the extent to which wax aldehydes were produced . Whereas these had not changed in lcr when compared to wild type , the fdh leaf wax appeared to contain much greater amounts of all aldehyde species than the wild type did , with C28 exhibiting the biggest ( nearly a 830-fold ) increase . We also examined the leaf epidermis in lcr , fdh , bdg and wild type by cryo-scanning electron microscopy ( SEM ) which preserves wax morphology . Under SEM , the leaf surface in the wild type appeared to be smooth , without appreciable sculpturing . However , in all three mutants examined , a considerable number of ripples and plate-like wax crystals gave their surfaces a more ruffled appearance ( Figure S3 ) . We concluded from these results that lcr and fdh respond to the loss of respective gene functions by the over-accumulation of wax in leaves . This response is quite similar with respect to alkanes , but only fdh appears to activate a pathway for aldehyde biosynthesis . Our SEM results suggest that stem wax is also affected in these mutants ( Figure S4 ) , but a detailed analysis would be beyond the scope of this paper . The cuticular phenotypes of lcr and fdh described above , and the analysis of their cuticular lipids , suggest that these mutations may induce a kind of a response which includes the genes associated with cutin and wax biosynthesis . If this response is controlled at the level of transcription or mRNA stability , coordinated changes in the abundance of transcripts which encode corresponding genes should be observed in the cuticular mutants . To test this possibility , we studied the gene expression changes by using microarray hybridization with the Arabidopsis ATH1 Genome Array ( Affymetrix ) . We then compared gene expression in young rosette leaves from fdh , lcr and bdg mutants to that in wild type . As described in Materials and Methods , RNA-derived probes were prepared from three biologically independent samples for each mutant . Taking into account the low replicate numbers of the microarray data , we have chosen to detect differentially expressed genes ( DEGs ) using the Rank Product ( RP ) method , as suggested by Breitling and co-workers [31] . This produces a good performance , in particular for replicate numbers below ten [32] . We recently revealed that RP outperforms Cyber-T , Local Pooled Error ( LPE ) , two-sample Bayes T , Empirical Bayes , SAM , fold change and the ordinary t-test in terms of the validity of the DEG lists [33] . The significance of the detection is assessed in RP by a non-parametric permutation test which evaluates the percentage of false positive predictions ( pfp ) or the false discovery rate ( FDR ) . In this study , we regarded genes with a pfp of less than 5% ( 0 . 05 ) to be significantly differentially expressed because , for them , the probability of being consistently selected by the RP method is greater than 95% . This filtering resulted in a list of 440 DEGs in fdh when compared to wild type , followed by lcr ( 260 DEGs ) and bdg ( 126 DEGs ) . The DEGs are listed in Tables S1 , S2 . Figure 5A summarizes the findings , and shows Venn diagrams which represent the number of genes that were changed and up or downregulated in the three cuticular mutants . The microarray analysis suggests that the majority of DEGs were upregulated: 88% for fdh , 91% for lcr , and 93% for bdg ( Figure 5A ) . It also reveals large overlaps between misregulated genes in the different mutants . This supports the notion that these mutants exhibited similar transcriptional changes , as suggested by similarities in their organ fusion phenotypes and the composition of their cuticular lipids . The expression of only 13 genes ( 10% ) was specifically changed in bdg , compared to 50 specific genes ( 19% ) in lcr and 240 specific genes ( 54% ) in fdh ( Figure 5A ) . Table S3 includes 89 genes which were found in the overlap between the genes that were differentially expressed in all of these mutants . Remarkably , these 89 common genes represent 71% of all of the DEGs in bdg . A simplified version of this table is shown in Table 1 . To substantiate this computational analysis , we re-calculated the microarray data as one experiment which consisted of two groups: the cuticular mutant group with nine replicates and the wild type group with three . This experimental design should minimize the inter-dependence between mutant versus wild type group differences . One would also expect statistical power to increase as the number of replicates goes from three to nine for the mutant group . By using the same parameter settings in the Rank Product method , and the same pfp cut-off value , we obtained a list of 744 upregulated DEGs . Compared to the 87 commonly upregulated genes which were identified with the first approach , many more candidates were detected by Rank Product this time , suggesting that the actual number of genes discriminating between the three cuticular mutants and wild type is higher . However , 74 ( 85% ) of these 87 genes were found in the overlap with the top-87 gene list from the second approach . A comparison of gene ranking also reveals that there is a significant consensus between the lists obtained by the two approaches . To further corroborate our results , we sought to first demonstrate by semi-quantitative RT–PCR that the selected genes are , in fact , up or downregulated as predicted by the microarrays . We have chosen 12 candidate genes from among those which are commonly upregulated in the three mutants ( Table S3 ) , as well as two genes which were not included in this list . The first was CER1 , which is thought to be directly involved in wax biosynthesis , although its enzymic function remains unknown; the other was a RAP2 . 6-like gene ( RAP2 . 6L ) which encodes an AP2/EREBP domain protein ( Text S1 ) . RAP2 . 6L was one of the genes which appeared to be upregulated , but did not meet the criteria because one of its pfp values was slightly above the 0 . 05 cut-off ( 0 . 052 ) . From the microarray hybridization analysis , we estimated that the selected genes were upregulated in the range of 3 to 172-fold in the mutants ( Table S3 ) . The results of semi-quantitative RT–PCR assays , shown in Figure 5B , indicated that all selected genes were consistently upregulated in the mutants . Interestingly , one can also observe a correlation between the genes that are strongly upregulated in both assays ( namely the microarray and the semi-quantitative RT–PCR ) with the LTPs , PPT , RAP2 . 6 and DAISY genes which display the most distinct expression between mutants and wild type . However , we did not aim to quantitatively evaluate gene expression measurements , or compare respective fold changes which would have required the use of real-time , quantitative RT–PCR . One of the genes which was found to be highly up-regulated in the leaves of all three mutants ( in lcr 5 . 8-fold , in fdh 11 . 7-fold and in bdg 9 . 0-fold , as revealed by microarrays ) was DAISY ( Table 1 ) . It was shown to encode a KCS which is involved in the biosynthesis of aliphatic suberin in roots . The roots of the daisy mutant accumulate significantly less C22 and C24 very-long-chain fatty acid derivatives in suberin , suggesting that it functions as a docosanoic acid synthase [34] . The RT–PCR analysis and promoter-GUS fusions revealed that it was also expressed in various aerial organs of the plant , although its expression levels in leaves were very low [34] . While DAISY was almost undetectable in unwounded rosette leaves , its expression was rapidly induced by wounding , and correlated with suberin deposition around punctures [34] . These features render DAISY a good target for confirmation by in situ hybridization . In particular , we wondered whether it would be specifically induced in the leaf epidermis of the cuticular mutants . The in situ hybridization results from wild type leaves , as shown in Figure 6A , demonstrate that the expression of DAISY is restricted to the xylem and phloem in vascular bundles of older rosette leaves; the hybridization signal was also observed the transmitting tract and ovules ( Figure 6B ) . In the lcr and bdg mutants , the signal was also detected in leaf primordia and young developing leaves ( Figure 6C–6 F ) , and the intensity of labeling in these organs was similar to that in the vasculature ( Figure 6E and 6F ) . In the mutants , DAISY was ectopically expressed in all cell types in leaves , including the epidermis , and careful examination did not reveal any cell specificity . We concluded that the enlargement of the domain of DAISY expression obtained by in situ hybridization is in agreement with our microarray data and the results of semi-quantitative RT–PCR , all of which evidence the induction of DAISY in young leaves of lcr and bdg . We also concluded that the data from the microarray hybridization are reliable and suitable for a comparative analysis . To find out which biological processes are most affected in cuticular mutants , we first used the Classification SuperViewer [35] which analyses Gene Ontology ( GO ) annotations ( ATH_GO_GOSLIM . 20070512 ) in order to identify overrepresented GO terms when compared to the entire Arabidopsis genome . The most prominent functional group of the upregulated genes ( Table S3 ) was represented by “cell wall” related proteins , with a 15 . 8±4 . 7-fold enrichment ( mean±standard deviation for 100 bootstraps ) when compared to the whole genome . These were followed by the “extracellular” ( 7 . 6±3 . 3 ) , “response to abiotic or biotic stimulus” ( 3 . 5±1 . 0 ) , “plasma membrane” ( 2 . 2±1 . 7 ) , “response to stress” ( 2 . 1±0 . 9 ) , “other enzyme activity” ( 2 . 1±0 . 5 ) , “developmental processes” ( 1 . 7±0 . 6 ) , “hydrolase activity” ( 1 . 7±0 . 5 ) , “transport” ( 1 . 7±0 . 6 ) , and “other membranes” ( 1 . 7±0 . 3 ) groups . Other listed terms included “electron transport or energy pathways” , “transcription factor activity” , “signal transduction” and “transcription” , but these were less conspicuous . These results strongly indicate that the cell wall undergoes extensive changes in response to cuticular mutations . Given that the cuticle is an essential part of the cell wall , this should not come as a surprise . However , it is interesting that the genes associated with responses to abiotic or biotic stimuli were only third in this list , suggesting that the response to cuticular dysfunction leads to a specific compensatory response whereby the normal homeostasis of the cell wall ( and cuticle ) is altered , and the viability of the mutants is increased . Our survey also showed that genes potentially associated with the cell wall , the cuticle and defense responses are upregulated in the cuticular mutants ( Text S1 ) . Although the list of commonly misregulated genes presented in Table S3 may not be complete and may contain about 5% of falsely discovered genes , this is a further indication that bdg , lcr and fdh respond similarly to the dysfunction by remodeling their cuticles and cell walls and activating defenses . The above findings , when taken together , suggest that the transcriptional activation of target genes is an adaptive response to cuticular mutations such as bdg , lcr and fdh . Moreover , the fact that the three mutants display a peculiar phenotype , which is comprised of the overproduction of wax , organ fusions , irregular leaf shapes and defects in cell differentiation , suggests that the underlying signaling pathways may be distinct from , but overlap with , those that are activated in response to conventional biotic and abiotic factors . To identify related pathways , we sought to determine which transcriptional responses were the most similar to those observed in bdg , lcr and fdh . We also thought that contrasting these results to a similar analysis for cell wall deficient mutants would help further in the definition of the mechanism by which cuticular mutations induce plant defenses . To this end , a comparative analysis of differentially regulated genes among the three mutants should be extended to include several hundred of the microarray datasets that are available for Arabidopsis . This is challenging to implement because the generally low consistency of differentially expressed gene ( DEG ) lists achieved with the use of t-type tests has been reported by several groups , including those participating in the MicroArray Quality Control ( MAQC ) project [36] . It would be beyond the scope of this paper to go further into the computational details but , using the human MAQC project [36] and the Arabidopsis datasets available from the Gene Expression Omnibus ( GEO ) , we have conducted a comparative survey of the acceptability of several statistical approaches . This revealed that the DEG lists selected by the Rank Product method [31] outperform , in terms of consistency , those generated by seven other methods [33] . Building upon this finding , we developed guidelines for a large-scale comparative analysis of DEG lists , and re-analyzed over 600 contrasts ( e . g . mutant vs . wild type or treatment vs . control comparisons ) with the Rank Product method , using the raw probe intensity data from the Affymetrix CEL files that we obtained from several databases and authors . We termed this approach MASTA ( MicroArray overlap Search Tool and Analysis ) , after the phrase FASTA that is used for sequence comparison . Mutations in CELLULOSE SYNTHASE4 ( CESA4 ) /IRREGULAR XYLEM5 ( IRX5 ) result in modifications in the composition and structure of the secondary cell wall and lead to the specific activation of defense pathways , suggesting that a cell wall integrity system which is similar to that of yeast may exist in plants [37] . This makes the side-by-side meta-comparison with cuticular mutants worthwhile . We , therefore , probed the MASTA database with the Rank Product generated lists of DEGs from both this study and the cesA4/irx5 microarray experiment [37] . For ease of computation , each probe comprised equal numbers ( in this case 200 ) of the top up and downregulated DEGs , ranked by gene pfp ( FDR ) score . The original MASTA output files are too long to be included in their entirety , and only selected subsets ( 126 out of 1208 overlaps ) are , therefore , shown as examples in Figure 7 for lcr and cesA4/irx5 . For each probe , a MASTA search performs pairwise testing for overlaps between up and downregulated genes in the probe and target DEG lists , thus totaling four combinations for each comparison . Following the customary terminology used in genetic analysis , in MASTA a coupling-phase overlap refers to an overlap between up and upregulated genes , or between down and downregulated genes , whereas repulsion-phase overlap refers to an overlap between up and downregulated genes , or between down and upregulated genes in the probe and target DEG lists , respectively . The MASTA revealed that the DEGs in cuticular mutants were in the coupling phase with those in the CUTE plants [19] , with 53 genes ( P<2 . 0×10−64 ) in the overlap between the upregulated genes ( Figure 7 ) . The cesA4/irx5 mutation induced a number of genes which are shared with cuticular mutants , but gene overlaps varied between 29 and 36 ( P<5 . 3×10−27 and P<6 . 2×10−37 , respectively ) ( Figure 7 ) . It is apparent from the comparative analysis that lcr transcriptional responses are quite similar to those induced by wounding . The overlap between upregulated genes was conspicuously like the early wounding response ( 15 min after wounding ) , but it became stronger at subsequent points in time . Although wounding-inducible genes showed statistically significant overlaps with DEGs in cesA4/irx5 , the salt treatment , osmotic stress and drought appeared to misregulate most of the similar set of genes in cesA4/irx5 . Large gene overlaps were noticeable for both upregulated and downregulated genes ( Figure 7 ) , with salt stress being the most similar to cesA4/irx5 . Remarkable differences between lcr and cesA4/irx5 were seen in the overlaps with transcriptional responses to growth regulators: steroids , methyl jasmonate ( MeJ ) , naphthaleneacetic acid ( NAA ) and abscisic acid ( ABA ) . While hardly any overlaps above the threshold line were detected with the genes downregulated in lcr , the strong repulsion-phase overlaps were displayed with genes downregulated in cesA4/irx5 . This suggests that a number of steroid-inducible genes ( from 50 to 60 ) are suppressed in the cesA4/irx5 plant , thereby evidencing the fact that steroid hormones play an essential role in the cesA4/irx5 phenotype . Adding to the difference between cuticular and cell wall integrity signaling pathways , was the presence of strong coupling phase overlaps between cesA4/irx5 DEGs and those induced by MeJ , NAA and ABA ( Figure 7 ) . The involvement of ABA and jasmonic acid ( JA ) , predicted by MASTA , was in accordance with the results of Hernandez-Blanco [37] and co-workers , who used the Genevestigator Meta-Analyzer tools ( www . genevestigator . ethz . ch/at/ ) to compare selected upregulated genes which showed fold-change values >2 . Remarkably , although cesA4/irx5 was not tested , irx1 and irx3 were crossed with two ABA-insensitive mutants , abi1 and aba3 , as well as the double homozygous mutants , irx1 abi1 and irx3 aba3 , and appeared to be unviable two to three weeks post-germination . Therefore , based on these findings , we propose that in contrast to cesA4/irx5 , major gene expression changes in lcr are induced independently of MeJ , NAA and ABA , and brassinosteroid signaling . Taken together , these data suggest that the underlying response mechanisms in the cesA4/irx5 and lcr mutants to cell wall and cuticle defects , respectively , are sufficiently different . The suppressor/enhancer screens involving the mutagenesis of the targeted mutant provide a solution to the problem of revealing additional genes in a given pathway . Alternatively , a set of known mutants may be crossed with the mutant of interest to make a series of double mutants , then enabling their phenotypes to be evaluated . The public availability of various mutant microarray datasets provides an opportunity to rapidly assess the potential of a large number of genes as genetic modifiers by using MASTA prior to the crosses . When resulting from probing with the mutant of interest , the coupling and repulsion phase overlaps indicate the presence of putative enhancers and suppressors , respectively . To identify these , we surveyed the most significant gene overlaps with the mutant microarray datasets . A particularly noticeable case in Figure 7 was the coupling-phase overlaps of cesA4/irx5 with pad4 ( 54 upregulated genes , P<3 . 7×10−66 ) and eds1 ( 59 upregulated genes , P<4 . 1×10−75 ) . This suggested that both could be enhancers of cesA4/irx5 in double mutants . EDS1 and PAD4 encode lipase-like proteins which can form a heterodimer and are required for the accumulation of the plant defense signal , salicylic acid [38] . Eds1 , but not pad4 , showed a coupling-phase overlap with lcr ( 38 upregulated genes , P<6 . 3×10−40 ) , suggesting that it could enhance the lcr phenotype in double mutants , whereas pad4 could not . Three cases in Figure 7 , which we considered to be particularly relevant as suppressors , were the recessive mutations in SERRATE ( SE ) , SENSITIVE TO FREEZING2 ( SFR2 ) and SENSITIVE TO FREEZING6 ( SFR6 ) . From 23 to 45 ( P<3 . 1×10−19 to 6 . 3×10−51 ) , the genes that were repressed in the se , sfr2 and sfr6 mutants were induced in lcr , fdh and bdg , exhibiting a repulsion phase overlap . The se-1 mutants exhibit conspicuous leaf serration , and are affected , not only in other aspects of leaf development , but also in embryogenesis , flowering time and seedling responses to the hormone , ABA . The stronger se alleles , namely se-2 and se-3 , severely disrupt both meristem activity and leaf polarity [39] . SERRATE ( At2g27100 ) is a zinc-finger protein which has been shown to participate in both RNA splicing and the processing of pri-miRNA transcripts into mature miRNAs [39]–[43] . The MASTA search suggests that SE has a different role in the cell wall stress pathway , with 19 downregulated genes in a coupling-phase overlap with cesA4/irx5 ( Figure 7 ) . We reasoned that if SE is required for the induction of the responses in cuticular mutants , some aspects of their phenotypes should be the opposite to those of se because MASTA reveals a repulsion phase overlap for their misregulated genes . We had previously noticed that lcr and bdg do , indeed , possess smooth-edged elongated leaves [14] , [16] , and this case has , therefore , been selected for further testing . From the above analysis with MASTA , one can anticipate the suppression of either phenotype in double mutants , although given the fact that SE acts in the context of a multiprotein complex which is involved in RNA processing , se is more likely to be epistatic to the cuticular genes . To corroborate this prediction , we obtained double mutants with the weak se-1 allele [41] . The Figure 8A shows that both se lcr and se bdg feature serrated leaves and , essentially , look like se plants [41]; ( se fdh is not yet available because the genes are tightly linked on the long arm of chromosome 2 ) . Remarkably , under normal growth conditions , the double mutants failed to develop the ectopic organ fusions , leaf deformations and plant architecture that were characteristic of the single cuticular mutants in this class ( Figure 8A ) [9] , [14] , [16] . The TB staining phenotype of lcr could be significantly reverted in a se mutant background ( Figure 8B ) . After infection with B . cinerea , se lcr and se plants tended to have larger lesion areas than wild types ( Figure 8C ) ( the lesion areas are not shown as the lesions often covered the entire leaf surface in se and se lcr ) . The rate of infection was higher in se lcr and se mutants than that in wild type plants ( Figure 8D ) . Remarkably , leaves of se bdg exhibited TB staining pattern and resistance to B . cinerea similar to that in bdg ( Figure 8 ) . These findings suggest that secondary phenotypes in distinct cuticular mutants result from the induction of a response , which requires SE . This is also evidence that MASTA provides a powerful way of identifying suppressors and enhancers in the pathway of interest .
It might be anticipated that cuticular mutants would , in general , have reduced levels of cutin monomers or wax and display conspicuous cuticle defects , however our data suggest a more complex picture when considering the cuticular mutants which display the organ fusion phenotype . We showed that the cuticle is highly disorganized in lcr , with cutin-like depositions and cavities in the inner layers of the cell wall . This closely resembles the cuticle of bdg and the CUTE plants [14] , [19] as well as , to a lesser extent , ace/hth [27] . Yet , we also demonstrated that TEM did not reveal any visual aspects in the fdh cuticle which appear to be different from wild-type . However , these organ-fusion mutants are noticeably similar in the accumulation of higher levels of wax and cutin constituents in the residual bound lipids . This sets them apart from the wild-type-looking plants of cuticular mutants which exhibit a concomitant lowering of the levels of cutin components , such as lacs2 and att1 [21] , [44] , [45] . We also found that the chemical composition of the cutin , as revealed by the analysis of leaf residual bound lipids , does not show a decrease in ω-hydroxy and α , ω-dicarboxy fatty acids in lcr; there was also no reduction in the lengths of the fatty acid chains in fdh . It is , therefore , becoming evident that a simple case scenario does not seem to be plausible for all cuticular mutants , meaning that other mechanisms need to be taken into account in order to understand the precise nature of their phenotypes and the role of cuticle in development and immunity . It may be proposed that some cuticular mutations induce a kind of a damage response that is similar to the activation of the cell wall integrity pathway in response to cell wall disrupting drugs and mutations in fungi and plants [37] , [46] . Consistent with the previously observed induction of stress and defense-related genes in some cesA mutants [37] and the CUTE plants [23] , we found by microarray gene expression profiling in lcr , fdh and bdg that a number of upregulated genes fall into this class . The GO analysis suggests that the functions of these genes are mainly associated with the cell wall . We showed that misregulated genes in cuticular and cell wall mutants substantially overlap , but these genes are significantly more similar to each other than to those in the cesA4/irx5 mutants . This may be indicative of the existence of specific signaling routes which engage transcriptional control mechanisms in the cuticular mutants . The fact that cuticular mutations confer a highly pleiotropic phenotype , including organ deformations and fusions , the extensive branching , delayed senescence and resistance to the fungal necrotroph B . cinerea [9] , [10] that were not observed in cell wall mutants is in accord with this notion . Using biochemical analytical techniques may make it difficult or impossible to separate the overlapping loss-of-function and response phenotypes , thereby emphasizing the role of genetic approaches . To further assess the complexity of the factors responsible for the range of cuticular phenotypes , we used a meta-analytical methodology , MASTA , which has involved a reappraisal and re-analysis of several hundred Arabidopsis microarray datasets from public databases and authors [33] . MASTA is based on the proof-of-concept study showing that , a reference database of expression profiles which correspond to diverse chemical treatments and mutations in yeast can be used to functionally annotate uncharacterized genes and pharmacological perturbations in this substance [47] . Using this bioinformatics tool for the in silico suppressor screen , we identified SE as a gene which is epistatic to lcr and bdg . This prediction has been supported by three lines of evidence . First , the double mutants , se lcr and se bdg , failed to produce featured morphological changes , including organ fusions , in particular . Second , the se lcr mutant lost its characteristic TB staining on the epidermis surface . Third , se lcr lost resistance to the necrotroph , also indicating that se is an essential factor contributing to the complex phenotypes of cuticular mutants . The ectopic organ fusion phenotype results from enhanced cell adhesion in epidermal cells and is interesting , because cell adhesion is a fundamental process underlying development . While adhesion between plant cells is generally established when cells are formed during cytokinesis , cells dynamically regulate adhesion , and may undergo separation or establish fusions de novo in a controlled manner , with pollen tube growth and carpel fusion being the best-known examples of the latter process [48] , [49] . This suggests that there are multiple mechanisms by which plants can regulate adhesion between cells . Organ fusions in the fdh mutant of Arabidopsis offered a genetic proof that the developmental program , normally limited to the gynoeceum , could be induced in the whole plant [12] . In addition to fdh , other Arabidopsis mutants have been reported to exhibit organ fusions and impaired morphogenesis [15] , [16] , [19] , [27] , [50]–[57] . However , the function of cuticle in these processes remained open to question . Given the role that the cuticle plays in the isolation of plant surfaces , it is necessary to study the corresponding genes in the cuticle context , particularly because most of the mutants seem to be the consequence of lesions in the genes which encode lipid-modifying enzymes . Remarkably , cell wall-related genes are prominent in the overlap which is comprised of the commonly upregulated genes in the cuticular mutants and downregulated genes in se ( Table 1 and Table S3 ) , Further experiments would be required to determine which of these genes , or others as yet not recognized , are involved in epidermal adhesion and leaf morphogenesis . Since the latest version of the Arabidopsis genome annotation ( TAIR8 ) contains information about 27235 protein-coding genes , and the ATH1 array represents approximately 23750 genes ( 87% ) [58] , we note that about 13% of the downstream genes of interest may escape identification in both our and Lobbes' microarray experiments [43] . The discovery of epistatic effects could be the first step towards identifying the cell-surface molecules and understanding genetics of the putative interactional mechanisms which underlie the organ fusion phenotype . The mechanism by which the distorted cuticle leads to a strong resistance to B . cinerea is also unclear but may involve an inhibitory action of plant-derived toxins such as camalexin and a higher permeability of cuticle to these compounds or fungal elicitors [59] , [60] . Fast induction of camalexin biosynthesis genes in wounded and infected plants accounts for the strong immunity against B . cinerea [59] . MASTA revealed that DEGs in se ( data not shown ) were in the repulsion phase with those in cuticular mutants and wounded plants , suggesting that wounding response may be compromised by the se mutation in se lcr . However , we did not find that se bdg plants were more susceptible ( and become less stainable ) than bdg , indicating a greater complexity of the antifungal resistance in cuticular mutants . The function of the nuclear-localized SE protein in the regulation of the pleiotropic phenotypes of cuticular mutants also remains to be determined . So far it is known that , like DICER-LIKE1 ( DCL1 ) and HYPONASTIC LEAVES1 ( HYL1 ) , SE is required for miRNA biogenesis but not for sense post-transcriptional gene silencing [42] , [43] . In se , 20 upregulated genes have been identified as known targets of miRNAs and/or transacting siRNAs ( ta-siRNAs ) [43] . However , only downregulated genes in se show an overlap with DEGs in the cuticular mutants , and none of the se downregulated genes present on the ATH1 chip are known miRNA targets [43] . Interactions between the SE , DCL1 and HYL1 proteins are essential for the efficient and precise processing of pri-miRNA [61] , although MASTA does not predict the presence of epistatic interactions with dcl1 and hyl1 ( data not shown ) . The possibility that SE may be associated with an alternative pathway in RNA signaling warrants further investigation . Nevertheless , it is an important finding that se is epistatic to the mutations in the distinct epidermis-specific cuticular genes LCR and BDG , suggesting that the impaired cuticle triggers specific cell signaling pathway . Accordingly , this study offers an intriguing and unexpected insight into how cuticle formation , cell adhesion and morphogenesis in plants may be co-regulated .
All plants used in this study were derived from Arabidopsis thaliana ( L . ) ecotype Columbia ( Col-O ) . The mutant alleles used were: fdh-3940S1 [20] , lcr-3P77 [16] , bdg-2 [14] , se-1 [41] , lacs2-3 [21] and pad3-1 [25] . Arabidopsis plants were grown in a greenhouse or controlled environment chamber at 22 to 23°C and 50 to 60% humidity under a short day photo-period ( 8-h light ) for the first 6–7 weeks , and then under a long day photo-period ( 16-h light ) if not , otherwise , indicated . The putative 1 . 3-kb promoter region of the LCR gene was amplified by PCR with the PLCR-H ( TGAACTCCAAAGCTTTACATGACTACATCG ) and PLCR-Xb ( CTCTAGATCTCCTCATAAACTTGGAGTGA ) primers ( HindIII and XbaI sites are underlined in the primer sequences ) , and cloned as a HindIII/XbaI fragment into the pBgreen binary vector [17] . Wild-type Arabidopsis thaliana Col-0 plants were transformed with the resulting pBLCR:GFP by vacuum-infiltration [62] , and BASTA-resistant transgenic plants were selected . The analysis of GFP expression with a confocal laser scanning microscope ( Leica TCS 4D ) was performed in tissue sections from the pBLCR:GFP plants as described [14] . In situ hybridization was performed using the same antisense digoxigenin-labeled riboprobes that were derived from a cDNA clone of the DAISY gene , and an epidermis-specific control , as in Franke [34] . The hybridization products were revealed by an immunohistochemical reaction after incubation with an alkaline phosphatase-conjugated anti-digoxigenin antibody . Probe preparation , the hybridization procedure , and immunohistochemical detection were conducted as previously described [17] . To study the fine structure of the cuticle , plants were grown for 4–5 weeks under long day conditions . Tissue samples were embedded , and ultra-thin sections ( 50–70 nm ) were contrasted as described in [21] ( fdh ) and [19] ( wild type and lcr ) . The samples were examined with a Phillips CM12 transmission electron microscope or a Philips CM 100 BIOTWIN electron microscope . The details of the procedures can be found in our previous work [19] , [21] . To examine wax coating , rosette leaves from 6–7 week-old plants grown under short day conditions , as well as stem internodes ( 4th and 5th ) from 12 week-old plants ( 8 weeks under short and 4 weeks under long-day conditions ) were prepared for cryo-SEM . The samples were deep-frozen and sputtered with palladium using the K1250X cryogenic preparation system ( Emitech , England ) . Leaf surfaces were examined with a Zeiss SEM SUPRA 40VP microscope . Toluidine blue staining was performed according to Tanaka and co-workers [26] . Chlorophyll leakage from rosette leaves into ethanol was performed according to Lolle and co-workers [13] , with modifications as previously described [14] . B . cinerea strains 2100 and B05 . 10 were grown on the potato dextrose agar ( Difco Laboratories ) at 22°C for seven to nine days . Spores were harvested and washed twice in water , and filtered through Miracloth ( Calbiochem ) . Inoculations were made as previously described [24] . Briefly , rosette leaves of four-week-old soil-grown plants were placed in square Petri dishes containing 0 . 8% agar with petioles in the medium . Four µl of a suspension containing 5×105 conidiospores/mL in 1/4-strength potato dextrose broth ( Difco Laboratories ) were deposited on the detached leaves . A single drop was applied to each leaf between the middle vein and the edge of the leaf , and the leaves were incubated under continous light at 22–24°C . Disease symptoms were scored at 3 days after challenge . High humidity was maintained by sealing the dishes with Parafilm . Macroscopic images were acquired with a digital camera ( Sony DSC-W120 ) and the lesion area was measured in pixels using Image J ( software available at http://rsbweb . nih . gov/ij/ ) and then converted to square millimeters . The inoculation experiments were repeated three times with detached leaves using at least 50 leaves per genotype and once in planta with similar results . Fatty acid composition analyses of residual bound lipids and wax were performed as previously described in detail [14] , [27] . Cutin and wax constituents were separated and identified by GC–MS using a gas chromatograph 6890N equipped with a quadrupole mass selective detector 5973N ( Agilent Technologies , Boeblingen , Germany ) . The composition analysis in the lcr and fdh leaves was performed twice . In each experiment , plants were grown for 10–11 weeks under short day conditions prior to tissue harvest . For wax analysis , plants were grown for seven weeks under the same conditions . Mutants ( bdg , lcr , fdh ) and WT ( Col-0 ) were grown side by side in a growth chamber under short day conditions at 20°C/18°C . Plants were five and a half weeks old when tissue was harvested . Three independent samples , each containing typical young leaves ( from 2 mm to 1 cm long ) from 15 plants , were prepared per plant type . Total RNA was extracted using the RNeasy Plant Mini Kit ( Qiagen ) according to the manufacturer's instructions . RNA concentration and quality were assessed with agarose gel electrophoresis . The samples were sent to the Integrated Functional Genomics ( IFG ) platform of the Westfalian-Wilhelms-University ( Muenster , Germany; http://campus . uni-muenster . de/ifg . html ) for a further quality check , preparation of biotin-labeled cRNA probes , hybridization to GeneChip Arabidopsis ATH1 Genome Arrays and scanning of the slides . All of the above steps were performed according to the manufacturer's instructions . For each plant type , we had three biological and no technical replicates . The pre-processing of raw data and the Affymetrix MAS5 . 0 Quality Control tests were performed using Bioconductor packages ( http://www . bioconductor . org ) and custom written scripts in the R programming environment ( http://www . r-project . org ) . For the quality control tests , we used the SimpleAffy package [63] . The quality-control measures indicated that the 12 microarrays used in this study show no systematic signal distortion , similar scale factors , and adequate background levels , sufficient percentage of genes called “present” and acceptable performance in 3/5 ratio tests ( Figure S5 ) . The microarray data reported in this paper have been deposited in the Gene Expression Omnibus ( GEO ) database , www . ncbi . nlm . nih . gov/geo ( accession no . GSE15105 ) . Statistical analysis was performed using custom written scripts for the Bioconductor RankProd package [64] . To estimate the false discovery rate ( FDR ) , pfp ( false positive predictions ) values have been calculated from 100 permutations . The predicted differentially expressed genes ( DEGs ) have been ordered by increasing pfp value . For this report , a 5% ( 0 . 05 ) pfp cutoff has been applied to the definition of the DEGs in the mutants . The meta-analytic software , MASTA ( MicroArray overlap Search Tool and Analysis ) , was written to run in R ( http://www . r-project . org ) . A database for MASTA was created . To this end , Affymetrix raw data files ( CEL files ) were downloaded from the Gene Expression Omnibus ( GEO , http://www . ncbi . nlm . nih . gov/geo ) , ArrayExpress ( http://www . ebi . ac . uk/microarray-as/ae/ ) , TAIR AtGenExpress ( http://www . arabidopsis . org/index . jsp ) , the Integrated Microarray Database System ( http://ausubellab . mgh . harvard . edu/imds ) , or via the NASC Affywatch subscription service ( http://nasc . nott . ac . uk/ ) . Some CEL files have been obtained from authors' websites or from the authors directly . At the time of this study , the MASTA database comprised DEGs for over 600 contrasts ( mutant vs . wild type or treatment vs . control comparisons ) calculated by using custom RankProd scripts in the R programming environment . Other portions of MASTA included the overlap analysis and plotting routines ( details will be published elsewhere ) . The RankProd-selected DEGs were ordered by increasing pfp value , and the top 200 DEGs from the lists containing up and downregulated genes were used for the overlap analysis in this report . Output PDF files from MASTA were imported to Adobe Illustrator ( Adobe Systems , San Jose , CA ) for assembly . The statistical significance of the overlap between two DEG lists was determined by using the online program available at http://elegans . uky . edu/MA/progs/overlap_stats . html . MASTA will be made available for viewing and downloading at http://bar . utoronto . ca/ ( The Bio-Array Resource for Arabidopsis Functional Genomics ) . In brief , 0 . 5 µg aliquots of the total RNA of each hybridization sample , treated with DNase I , were reverse-transcribed to the first-strand cDNA with a One-Step RT–PCR Kit ( Qiagen ) . These cDNAs were used as templates for PCR under the following conditions: denaturation at 94°C ( 1 min ) ; Nopt cycles of 94°C ( 1 min ) , 58°C ( 45 sec ) and 72°C ( 30 sec , except for At4g30280 where it was 1 min ) ; then 72°C final extension ( 15 min ) . Gene-specific primer pairs are listed in Table S4 . The expression of ACTIN2 ( At3g18780 ) was analyzed as an internal control . The semi-quantitative RT–PCR reactions were optimized for number of cycles Nopt to ensure product intensity within the linear phase of amplification ( close to the the lower limit of the linear range ) for each gene ( Table S4 ) . PCR fragments were quantified ( Typhoon 8600 PhosphorImager , Amersham Biosciences ) following electrophoresis in ethidium bromide-containing agarose gels . The segregating F2 populations ( about 200 plants each ) derived from the lcr×se-1 and bdg×se-1 crosses respectively were tested in a blind two-stage screen for the presence of mutant and wild-type alleles in the LCR , BDG and SERRATE loci by PCR . The following PCR conditions were used: denaturation at 94°C ( 2 min ) ; 36 cycles of 94°C ( 30 sec ) , 58°C ( 30 sec ) and 72°C ( 50 sec ) ; then final extension at 72°C ( 5 min ) . At the first stage , DNA was isolated in 96-well blocks , based on the method described previously [65] . To genotype the LCR locus , amplification products were loaded on a 1 . 5% agarose gel . For the BDG and SE loci , half of the PCR products were first digested with MwoI or BfuCI ( NEB ) respectively , and then the undigested and digested PCR products were loaded onto a high-resolution agarose gel ( 3% GenAgarose Tiny HT , Genaxxon Bioscience ) . At the second stage , candidate double mutants were re-screened by repeating DNA isolation with the DNeasy Plant Mini Kit ( Qiagen ) and PCR analysis . Several plants have been identified for each double mutant type . The allele-specific primers and PCR products are listed in Table S5; the sequence of the lcr-3P77 transposon insertion allele was deposited in the GenBank under accession number FJ767868 .
|
As the skin of a plant , the epidermis mediates a broad set of protective functions which includes defense against abiotic environmental stresses and pathogens . The majority of its barrier capacity is localized to the outermost cell wall , which is covered by a waxy cuticle . Several distinct cuticular mutants in the model plant Arabidopsis produce a remarkable syndrome that is characterized by ectopic cell adhesion and changes in plant morphology . We used these mutants to study the constitution of the cuticle and the activation of the molecular compensatory mechanisms that are important for adaptation . We examined whole-genome responses in these mutants and used an appropriate statistical procedure to reveal the genes which change their expression . We then applied the same approach to the analysis of hundreds of datasets in repositories . The comparison of gene expression profiles identified the gene SERRATE , which encodes a protein of RNA–processing multi-protein complexes , and further analysis revealed that the syndrome is suppressed in double mutants , as predicted . Our finding suggests that the mechanism which operates to control the integrity of the cuticle involves the regulation of small–RNA signaling .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"biology/plant-biotic",
"interactions",
"plant",
"biology/plant-environment",
"interactions",
"plant",
"biology/plant",
"growth",
"and",
"development",
"plant",
"biology/plant",
"biochemistry",
"and",
"physiology",
"plant",
"biology/plant",
"genetics",
"and",
"gene",
"expression"
] |
2009
|
Dissection of the Complex Phenotype in Cuticular Mutants of Arabidopsis Reveals a Role of SERRATE as a Mediator
|
Candida albicans is a human commensal and clinically important fungal pathogen that grows as both yeast and hyphal forms during human , mouse and zebrafish infection . Reactive oxygen species ( ROS ) produced by NADPH oxidases play diverse roles in immunity , including their long-appreciated function as microbicidal oxidants . Here we demonstrate a non-traditional mechanistic role of NADPH oxidase in promoting phagocyte chemotaxis and intracellular containment of fungi to limit filamentous growth . We exploit the transparent zebrafish model to show that failed NADPH oxidase-dependent phagocyte recruitment to C . albicans in the first four hours post-infection permits fungi to germinate extracellularly and kill the host . We combine chemical and genetic tools with high-resolution time-lapse microscopy to implicate both phagocyte oxidase and dual-specific oxidase in recruitment , suggesting that both myeloid and non-myeloid cells promote chemotaxis . We show that early non-invasive imaging provides a robust tool for prognosis , strongly connecting effective early immune response with survival . Finally , we demonstrate a new role of a key regulator of the yeast-to-hyphal switching program in phagocyte-mediated containment , suggesting that there are species-specific methods for modulation of NADPH oxidase-independent immune responses . These novel links between ROS-driven chemotaxis and fungal dimorphism expand our view of a key host defense mechanism and have important implications for pathogenesis .
Candida albicans is a ubiquitous commensal fungus and a clinically important opportunistic pathogen of humans . C . albicans is pleomorphic and grows in both yeast and filamentous forms , permitting growth in different environments , tissue invasion , dissemination and immune evasion [1]–[3] . Dimorphic switching is governed by myriad signals in vitro and is co-regulated with virulence factors; the links between dimorphism and pathogenesis in vivo are further complicated by the complexity of signals and the potential for immune control of differentiation [4]–[6] . Immunodeficiencies centered on either the innate or adaptive immune systems predispose for a number of opportunistic fungal infections with Candida spp . [7] , [8] . Lack of phagocyte NADPH oxidase ( Phox ) components causes chronic granulomatous disease ( CGD ) , a rare immunodeficiency associated with susceptibility to bacterial and fungal pathogens [9] . Over 45 years ago , the specific cellular defect in CGD was determined to be an inability of CGD leukocytes to mount a respiratory burst and kill microbes upon in vitro stimulation , yet this may not explain why CGD patients suffer from symptoms beyond susceptibility to acute infection , such as hyperinflammation and B-cell deficits [9]–[11] . Reactive oxygen species ( ROS ) produced during the respiratory burst are highly toxic to pathogens in vitro [12] , [13] , but it is now appreciated that ROS also impact many signaling pathways and cellular processes [9] , [14] . Notably , both the phagocyte oxidase Phox and the dual-specific NADPH oxidase ( Duox ) have been implicated in promoting chemotaxis to specific stimuli and/or sites of inflammation , although it is not clear if either has a role in phagocyte recruitment to sites of infection [15]–[19] . NADPH oxidases are important for immunity to many pathogens , although their roles in protection against C . albicans are not clear-cut . Most in vitro experiments suggest Phox is important for killing C . albicans , and CGD mice are more susceptible to candidemia in the tail vein injection model [20] . Other work suggests a more nuanced role for Phox , as candidemia is a rare cause of death in CGD patients [7] , [21] , Phox is not absolutely required for control of infection in the mouse [22] , and other phagocyte weapons can contain C . albicans both in vivo and in vitro [23]–[26] . The nematode model of mucosal candidiasis suggests that Duox can also play an important role in protection , although this has yet to be tested in mammals [27] . Current in vitro and in vivo models have not yet integrated these disparate data to explain the in vivo role ( s ) of NADPH oxidases in control of candidiasis . The emerging larval zebrafish model provides a unique and powerful platform to discern how the in vitro activities of pleiotropic molecules such as ROS translate into in vivo roles during infection [28]–[30] . We recently showed that a larval model of disseminated candidiasis shares key aspects of mammalian disease [31] . We performed extended intravital imaging of live zebrafish to show that macrophages can inhibit germination of yeast into hyphae in vivo . Additionally , we found that the phagocyte oxidase is important in limiting filamentous growth in vivo [31] . Here we link these two observations to show that NADPH oxidase-dependent recruitment of phagocytes limits filamentous growth because it ensures that C . albicans is phagocytosed efficiently and is thus prevented from germination . We demonstrate that both Phox and Duox are required for efficient phagocyte recruitment , phagocytosis , limiting filamentous growth , and survival . We find that early immune recruitment is a strong and reliable indicator of eventual infection clearance . We also implicate the EDT1-dependent dimorphic switching pathway in modulating both fungal containment and virulence of extracellular fungi . These data identify a new dimension of NADPH oxidase-mediated immunity that strongly impacts fungal dimorphism in the host setting .
We recently provided the first demonstration of an in vivo role for the phagocyte NADPH oxidase in limiting filamentous growth of C . albicans [31] . Here we sought to determine mechanistically how NADPH oxidase activity limits filamentation and susceptibility to infection . Traditionally , the most important role for NADPH oxidase in immunity has been ascribed to its ability to create reactive oxygen species that directly damage or kill microbes [13] . In fact , there is significant oxidative stress experienced by C . albicans when attacked by neutrophils or macrophages in vitro [32] , [33] . Therefore , we first hypothesized that NADPH oxidase-derived oxidants in the phagosome might damage C . albicans and block intracellular germination in this infection to limit filamentous growth . To determine if fungal cells were under oxidative attack in vivo , we used the OxYellow-T oxidative stress reporter strain . This strain has the oxidative stress-induced CTA1 promoter driving EGFP expression and the constitutive ENO1 promoter driving dTomato expression . Using this strain we find oxidative stress at 24 hpi but not at 4 hpi in control morphants , whereas there is no detectable oxidative stress at either 4 or 24 hpi in phagoctye oxidase morphants ( Fig . S1 ) , and using a similar strain we have previously published that there is no detectable oxidative stress at 6 hpi in this model [31] . We also found no activation of the respiratory burst within phagocytes , as observable upon incubation of live infected fish with H2DCF-DA ( Fig . S2 ) , a cell-penetrating molecule that diffuses well into live zebrafish and fluoresces upon oxidation [31] , [34] . Phagocyte oxidase-produced ROS have been demonstrated to drive localization of the autophagy reporter protein LC3 to the membrane of yeast-containing phagosomes in a process referred to as LC3-associate phagocytosis [35] , [36] . To determine if a similar process occurs in vivo in zebrafish , we examined the localization of a GFP-LC3 fusion protein in phagocytes containing C . albicans . We used a transgenic line of zebrafish for which GFP-LC3 localization has been shown to report on autophagic activity [37] . In contrast to previously reported in vitro findings , we found very few phagosomes with GFP-LC3 localized to the phagosomal membrane in vivo , suggesting that ROS-mediated LC3 localization plays a less important role in this in vivo model than has been demonstrated in vitro ( Fig . S3 ) . Further , two treatments recently shown by Huang et al . [35] to strongly inhibit phagosomal LC3 localization in vitro—blockade of NADPH oxidase activity with the pan-NADPH oxidase inhibitor diphenyleneiodonium ( DPI ) and treatment with the anti-oxidant α-tocopherol—mildly reduced but did not significantly affect GFP-LC3 localization to phagosomes . Taken together , these data are not consistent with the idea that NADPH oxidase acts early to produce respiratory burst-derived oxidants that damage the fungi or traffic it to autophagosomes and thereby block filamentous growth . Nevertheless , to determine if blockade of NADPH oxidase permitted germination of C . albicans within phagocytes in vivo , we examined the fungal morphotypes at 4 hpi with and without the pan-NADPH oxidase inhibitor DPI . Regardless of NADPH oxidase activity , there was a striking difference in morphotype between intracellular yeast and extracellular filamentous growth , with filamentous cells found only outside of phagocytes ( Fig . 1A ) . This difference in intra- vs . extracellular fungal morphology is statistically significant in both vehicle- and DPI-treated fish ( p<0 . 0001 by Fisher's exact test ) . Furthermore , there was no intracellular germination even upon blockade with DPI ( 0/204 for DMSO and 0/95 for DPI ) . Thus , in contrast to our expectations , inhibition of NADPH oxidase did not permit germination and filamentous growth within phagocytes early during infection . Instead , while some extracellular fungi switch morphotype and grow as filaments , phagocytosis can block germination even without NADPH oxidase activity . This suggests that there are other host immune mechanisms besides NADPH oxidase that can control C . albicans growth within phagocytes . Because time is a crucial axis of disease progression , we extended these studies to follow the fate of internalized C . albicans beyond 4 hpi to identify any later roles for NADPH oxidase in limiting germination . To determine if phagocytes continue to control filamentous growth of internalized yeast up to 22 hpi , we took advantage of the Tg ( mpeg1:GAL4/UAS:Kaede ) photoswitchable macrophage line [38] . We photoswitched macrophages in the hindbrain ventricle of representative fish at 4 hpi and followed them by time lapse every 2 hours until 22 hpi ( Fig . 1B ) . Again , we found that internalized C . albicans yeast did not germinate within macrophages or neutrophils for the duration of the time-lapse , even in the presence of DPI . In three independent experiments , a total of 279 fungi were followed in DMSO-treated fish ( 217 inside macrophages and 62 within neutrophils ) while in DPI-treated fish with limited engulfment , a total of 50 internalized fungi were followed ( 26 within macrophages and 24 within neutrophils ) . Control experiments suggest that the lack of germination is not due to photoactivation itself or imaging-induced inhibition of filamentation . Specifically , there is no inhibition of filamentous growth by frequent imaging in both green and red channels over the first six hours of infection and there is no clear defect in pathogenesis or immunity upon photoactivation ( Fig . S4 ) . The lack of intracellular germination in these extended time-lapse experiments suggests that macrophages and neutrophils remain effective in suppressing filamentous growth of internalized fungi , even when NADPH oxidase activity is blocked . Current in vivo models have not permitted extended assessment of Candida-phagocyte interactions at the infection site to characterize the dynamics of phagocyte migration . To examine the role of NADPH oxidase activity in controlling immigration and emigration of macrophages , we analyzed time-lapse experiments performed by photoswitching Kaede-expressing macrophages at the infection site at 4 hpi . We categorized the photoswitched and non-switched macrophage populations at the site of infection in the hindbrain ventricle between 4 hpi and 22 hpi . In control fish , about half of the photoswitched ( red ) macrophages left the hindbrain within 12 hours but were replaced by new ( green ) macrophages from outside of the infection site ( Fig . 1C ) . However , in fish treated continuously with DPI there was migration away from the hindbrain but no replacement with new macrophages , leading to uncontrolled growth of the C . albicans ( Fig . 1D ) . Importantly , control uninfected fish treated continuously with DPI suffered no ill effects . This suggests that the defects in immune infiltration associated with blockade of NADPH oxidase extend past 4 hpi and , if anything , are more severe when DPI treatment is continued to later times post-infection . Furthermore , these experiments document for the first time the dramatic flux of phagocytes to and from the infection site for hours post-infection . In mammals , macrophages are heterogeneous in function and can differentiate upon stimulation to promote diverse host responses , although it has not been possible to examine the dynamic roles of different subtypes in the context of C . albicans infection [39]–[41] . Because our transparent model offers a unique tool to identify differential roles of individual phagocytes during C . albicans infection , we were able to quantify the divergent kinetics of egress from the hindbrain in two macrophage populations . We found that macrophages that internalize yeast tend not to move away from the site of infection within the first 22 hpi , while most macrophages that do not engulf fungi move away from the infection site during this time . Approximately 50% of non-phagocytic macrophages leave the hindbrain by 12 hours after photoswitching , while there is no bulk emigration of phagocytic macrophages within 18 hours after photoswitching ( Fig . 1E ) . This suggests that phagocytosis of C . albicans is associated with reduced movement from the infection site . Given that NADPH oxidase is not required for intracellular containment of C . albicans , we sought another mechanistic explanation for its requirement in limiting filamentous growth . Our long-term timelapse experiments suggested that blockade of NADPH oxidase activity limited immune infiltration to the infection site ( Fig . 1 ) , and the Duox NADPH oxidase has been previously implicated in chemotaxis [19] , [42] . To test if NADPH oxidase is required for early phagocyte chemotaxis to C . albicans , we treated fish with DPI and examined phagocyte dynamics in Tg ( mpx:GFP ) i114 transgenic zebrafish [43] , with EGFP-expressing neutrophils . Over the first four hours of infection , we found a reduction in the level of immune infiltration to the site of infection and an even stronger decrease in the amount of intracellular containment of fungi ( Fig . 2A ) . Time-lapse imaging confirms that lower phagocyte numbers are due to loss of recruitment , rather than failure to retain phagocytes at the infection site . We quantified levels of infiltration and phagocytosis at 4 hours post-infection ( 4 hpi ) as the total number of EGFP-positive cells ( neutrophils ) combined with the number of EGFP-negative cells ( macrophages ) with internalized C . albicans and found that total infiltration was significantly decreased ( Fig . 2C ) . The number of total phagocytes with C . albicans inside was also significantly lower , as measured by the total number of immune cells , regardless of EGFP expression , that had engulfed C . albicans ( Fig . 2C ) . Quantification of the number of internal vs . extracellular fungi showed that overall levels of engulfed fungi were significantly decreased by DPI treatment ( Fig . 2D ) . Inclusion of both EGFP-positive neutrophils and EGFP-negative phagocytes—unambiguously scorable with internalized C . albicans—permitted a more robust measurement of the phagocyte response . Quantification of only the number of neutrophils at the infection site demonstrated a decreased number in DPI-inhibited fish , but the low overall number of EGFP-positive neutrophils at the site of infection led to differences that were not statistically significant ( Fig . S5 ) . In contrast to the effects on leukocyte infiltration , NADPH oxidase blockade did not strongly affect the overall ability of phagocytosis by neutrophils at the site of infection , as the percentage of neutrophils with engulfed fungi was largely unchanged , highly variable and not significantly affected by DPI treatment ( Fig . S6 ) . Taken together , these results indicate that short-term inactivation of NADPH oxidase activity evokes a significant deficiency in chemotaxis to the site of C . albicans infection . Previous work has focused on NADPH oxidase-dependent neutrophil migration to the site of wounding , but macrophages also play an important role in response to C . albicans infection [31] . To test if NADPH oxidase activity is required for macrophage chemotaxis , we took advantage of the new Tg ( mpeg1:GAL4/UAS:Kaede ) line of zebrafish with macrophages expressing the photoswitchable Kaede fluorescent protein [38] . NADPH oxidase inhibition caused a significant decrease in macrophage migration to the infection site over the first 4 hpi , as shown by time-lapse microscopy ( Fig . 2B ) . Quantifying these defects revealed significant reductions in total phagocyte infiltration and in total number of phagocytes with internalized fungi ( Fig . 2E ) , similar to results using the neutrophil transgenic ( Fig . 2C ) . In addition , we confirmed the strong defect in containment of fungi by phagocytosis using this transgenic with marked macrophages ( Fig . 2F ) . Quantification of only the number of macrophages at the infection site demonstrated a decreased number in DPI-inhibited fish , a statistically significant difference ( Fig . S5 ) . The use of both neutrophil- and macrophage-specific transgenic lines allowed us to account for different types of phagocytes ( either neutrophils or macrophages , depending on the transgenic line ) that did not phagocytose fungi , as well as all phagocytes with intracellular fungi . Because the results of these two complementary sets of experiments are comparable , this implicates NADPH oxidase in chemoattraction of both phagocyte types to the site of infection . Our results suggest that NADPH oxidase-dependent leukocyte attraction then promotes phagocytosis primarily through efficient chemotaxis to the infection site rather than enhancement of engulfment at the infection site . Serious tissue damage in the zebrafish larva elicits rapid neutrophil chemoattraction that is largely Duox-dependent [19] , [42] . To test if tissue damage accounts for the NADPH oxidase-dependent attraction of neutrophils to the hindbrain ventricle upon C . albicans infection , we performed mock injections of buffer into vehicle ( DMSO ) or DPI-treated larva and measured neutrophil recruitment . We found that there was a small increase in hindbrain ventricle neutrophils in mock-injected larvae ( from 0 . 75 to 2 neutrophils ) , although this was not significantly affected by DPI ( 2 . 05 vs 1 . 98 ) ( Fig . S7 ) . Thus , in contrast to the NADPH oxidase-dependent phagocyte recruitment to infection , the minor recruitment induced by this injection method is not NADPH oxidase-dependent . Our data demonstrate that NADPH oxidase ( s ) direct the early immune response to fungal infection in the zebrafish hindbrain ventricle , tissue in the central nervous system . To test whether there is NADPH oxidase-dependent phagocyte recruitment and fungal containment in a localized infection in a different tissue , we infected the swimbladder of 4 dpf larvae with C . albicans . We have recently shown that the presence of C . albicans in the larval swimbladder elicits similar immune responses to those seen in an in vitro reconstituted human epithelial infection model [44] . Here , we modified the published protocol by injecting fungi directly into the swimbladder of 4 dpf larvae . We pre-incubated larvae with DMSO ( vehicle ) or DPI , injected 5–20 fungi/fish , maintained treatment for 4 hours , and then scored neutrophil migration to the infection site . We found that injection itself leads to a small but statistically significant increase in neutrophils at the site of infection ( Fig . S8 ) . This injection-associated increase is presumably due to a small amount of damage due to the injection procedure itself . Interestingly , this injection-related increase in neutrophil numbers is partially NADPH oxidase-dependent , as there is a small but significant reduction in neutrophil recruitment upon DPI treatment ( Fig . S6 ) . However , while injection of fungi led to a strong increase in neutrophil migration to the swimbladder , this increase was not DPI inhibitible . Thus , the early innate response to C . albicans infection in the swimbladder tissue at 4 dpf is different from the response in the hindbrain ventricle at 2 dpf , where in the swimbladder the early neutrophil response is more robust and not NADPH oxidase-dependent . These differences may be due to tissue-specific or stage-specific immune responses . We have shown that efficient engulfment of C . albicans in the hindbrain ventricle depends on NADPH oxidase-mediated phagocyte recruitment , and that internalization blocks switching from yeast to filamentous form . We therefore speculated that the increased filamentous growth previously observed upon knockdown of p47phox [31] was due to defective early chemotaxis and containment . We tested the requirement of p47phox for early immune response by measuring immune responses to infection in p47phox morphants , with reduced phagocyte oxidase activity . Using the Tg ( mpx:GFP ) i114 line , we find that p47phox knockdown leads to chemotaxis deficits similar to that of DPI treatment . This is seen both with time-lapse imaging ( Fig . 3A ) and upon quantitation of phagocyte behavioral phenotypes at 4 hpi ( Fig . 3B ) . The similar immune deficits upon pan-NADPH oxidase inhibition and p47phox knockdown suggest that the phagocyte oxidase Phox is an important mediator of phagocyte migration and intracellular containment of C . albicans . Further , this suggests that the previously described increase in filamentous growth seen at 24 hpi in p47phox morphants [31] is a direct consequence of early defects in fungal containment . Our results indicate that the phagocyte oxidase is required for phagocyte chemotaxis , suggesting a requirement for NADPH oxidase activity in leukocytes . Although not previously implicated in chemotaxis to microbes , the epithelial NADPH oxidase Duox is highly expressed in the brain and has a previously defined role in neutrophil chemotaxis to wounds [19] , [42] , [43] , [45] . To test a role for Duox in phagocyte chemotaxis to C . albicans , we knocked down expression of duox , confirmed knockdown by rtPCR as described [42] , and verified that this morpholino eliminated all detectable transcript without causing gross developmental effects ( Fig . 4A ) . To assess global and local neutrophil numbers within prim25 zebrafish , we counted EGFP-positive neutrophils in the head and caudal hematopoetic regions of duox and control morphants in a mock experiment , both 1 hour and 4 hours after injection with phosphate-buffered saline . We found no difference in basal numbers of EGFP-positive neutrophils in either tissue of duox morphants at the stages of development most relevant to our studies ( Fig . 4B and 4C ) . In contrast to our expectations , immune responses to infection in the duox morphants were severely impaired , similar to what we find with chemical inhibition and with knockdown of p47phox . This is seen with time-lapse imaging ( Fig . 4D ) , quantitation of phagocyte behavioral phenotypes at 4 hpi ( Fig . 4E ) , and overall failure of phagocytosis ( Fig . 4F ) . Interestingly , we do find a trend toward decreased phagocytosis on a per-cell basis for duox morphant neutrophils that is more consistent than trends seen for both DPI-treated and p47phox morphant neutrophils ( Fig . S2 ) . This stronger phenotype suggests the possibility that Duox plays a more important role than p47phox in directing the phagocytic process , although the small number of neutrophils at the infection site in these treated fish makes such characterizations necessarily tentative . In sum , the phenocopy of DPI treatment in both p47phox and duox morphants suggests that both the phagocyte-expressed Phox and the non-phagocyte-expressed Duox are required for bringing phagocytes to the site of infection , thus promoting efficient phagocytosis and inhibition of germination . Our observations demonstrate that NADPH oxidases act quickly post-infection to attract phagocytes to C . albicans and limit its filamentous growth by internalization . To understand the importance of these early immune responses , we sought to identify the consequences of poor initial infiltration and phagocytosis . Our previous work established that failure to control filamentous growth at 24 hpi correlates with poor survival to 48 hpi [31] , and here we again exploited non-invasive imaging to ask if weak early phagocytosis is linked to extracellular filamentation and poor prognosis . We characterized infected fish at 4 hpi as “low” or “high” responders , depending on whether they had greater or fewer than five extracellular fungi . Surprisingly , there were fish of each type for each treatment group , including the control groups , indicating that there is some heterogeneity in immune competence among genetically identical individuals infected with the same doses . However , consistent with our time-lapse results , there were more low responders in DPI-treated ( Fig . 5A ) , p47phox morphants ( Fig . 5B ) and duox morphants ( Fig . 5C ) than the comparable controls . Thus , despite screening individual fish immediately post-infection to ensure consistent infectious doses , by 4 hpi the infections could be classified into two major categories dependent on the efficiency of intracellular containment . As expected , the number of high-responder fish was strongly decreased by all treatments that blocked NADPH oxidase activity . To determine if these early phenotypes are prognostic for survival , we assayed the fate of individual fish screened at 4 hpi . Fish were imaged and scored for phagocyte response at 4 hpi , then kept in individual wells of a 24-well plate until 24 hpi to assess their fate . As expected , low responders have a much worse prognosis than high responders , with approximately three-quarters succumbing to infection by 24 hpi ( Fig . 5D–5F ) . Remarkably , though , the prognosis among low responders is comparable between controls and treatment groups , and the same is true for high responders . Due to the role of phagocytosis in limiting germination , low responders have excessive filamentous fungal growth , and nearly all of the fish that die by 24 hpi are riddled with C . albicans filaments ( Fig . 5G ) . The close correspondence of early phagocytosis with infection containment and survival highlights the crucial importance of early NADPH oxidase activity in protecting the host against C . albicans . Considering the similar phenotypes between temporary chemical blockade and long-lasting morpholino knockdown , this suggests that early NADPH oxidase activity plays a more important role than later production of ROS in control of this acute disease . Our demonstration of NADPH oxidase-dependent phagocyte recruitment is in contrast to what has been seen with other pathogens [15] , [18] , suggesting that C . albicans may have a special ability to counter ROS-independent chemotaxis . Because the yeast-to-hyphal switch is an important virulence trait associated with genome-wide transcriptional remodeling [46] , [47] , we hypothesized that it may be required to limit NADPH oxidase-independent chemotaxis . To test this idea , we examined the effects of NADPH oxidase inhibition on infections with the yeast-locked edt1Δ/Δ mutant . We infected Tg ( mpx:GFP ) i114 fish with three different strains of dTomato-expressing C . albicans: wildtype , homozygous edt1Δ/Δ mutant , or heterozygous edt1Δ/EDT1 control . We used the heterozygous edt1Δ/EDT1 mutant to control for potential artifacts due to transformation . We treated infected fish with DPI or vehicle and performed time-lapse experiments to measure early immune response . To our surprise , we found that a high proportion of DPI-treated , edt1Δ/Δ-infected fish elaborate a strong early immune response in which most of the fungi is internalized ( Fig . 6A ) . Infections with the heterozygous edt1Δ/Δ/EDT1 control result in an intermediate phenotype , as is found frequently with mutants in the diploid C . albicans . Quantitation of this response in even the limited number of fish examined by time-lapse microscopy suggests that there is a similar level of overall immune recruitment to the edt1Δ/Δ infection site , independent of NADPH oxidase inhibition ( Fig . 6B ) . Internalization of edt1Δ/Δ is also apparently NADPH oxidase-independent , and a much higher percentage of yeast-locked fungi than wild type fungi are phagocytosed by 4 hpi ( Fig . 6C ) . Percent phagocytosis is intermediate for the heterozygous edt1Δ/Δ/EDT1 strain , suggesting that there may be a partial haploinsufficiency phenotype . Consistent with these high-resolution time-lapse results with a small sample size , we also find a large percentage of high responders in edt1Δ/Δ-infected , DPI-treated fish when large numbers of fish are screened at 4 hpi for their ability to contain the fungi ( Fig . 6D ) . The significant difference in NADPH oxidase-independent phagocyte migration to the yeast-locked mutant in fungal containment at 4 hpi suggests that changes in C . albicans during the dimorphic switch may play an important role in limiting phagocyte chemotaxis . Our data demonstrate that germination of extracellular C . albicans is enhanced by poor early immune response , which is associated with poor prognosis . We therefore reasoned that genetic blockade of the C . albicans dimorphic switch would prevent mortality , even in conditions of poor phagocyte containment . To investigate the contribution of the dimorphic switching program to virulence under these circumstances , we followed the fate of high and low responder fish infected with the yeast-locked edt1Δ/Δ mutant . Although the majority of edt1Δ/Δ-infected fish internalize fungi successfully , the naturally heterogeneous early immune response among individuals allowed testing of our original hypothesis that extracellular germination is responsible for the poor prognosis after weak early chemotaxis . As expected , the outcome of infections in low-responding fish diverges significantly between edt1Δ/Δ- and control-infected fish . In contrast to the situation with control-infected fish , most of the low responders infected with edt1Δ/Δ manage to survive to 24 hpi and beyond ( Fig . 6E ) . As is the case for other infections , there are no NADPH oxidase-dependent differences in mortality within high- and low-responder groups . Thus , even when the early immune response fails to successfully contain the majority of edt1Δ/Δ mutant fungi , their inability to turn on the dimorphic switching pathway prevents pathogenesis . These data point to the importance of the EDT1-dependent dimorphic switching pathway in both limiting early fungal containment and in exploiting a weak early response to grow extracellularly in filamentous form and cause mortality .
The advent of intravital imaging has begun to illuminate new aspects of host-pathogen interaction in the intact host . Here , we exploited a transparent zebrafish model of candidemia to address mechanistic questions relevant to human primary immunodeficiency and immune response dynamics . We describe a new role for NADPH oxidase in recruitment of phagocytes to the site of C . albicans infection , demonstrate that this early recruitment is a key event in control of infection , and provide evidence that the C . albicans dimorphic growth program impacts the ROS-dependence of early fungal containment . The discovery of this unanticipated role of NADPH oxidase in phagocyte recruitment highlights the importance of early immune responses and points to a potentially new role of fungal dimorphism in regulating phagocyte activity . In this study , we used a powerful in vivo model to demonstrate a role for NADPH oxidase-driven phagocyte containment of C . albicans . Classically , ROS produced by the phagocyte oxidase and the dual-specific oxidase have been ascribed functions in direct chemical attack against systemic and epithelial insults [9] , [48] . Although we find no evidence for a role of ROS in directly damaging intracellular C . albicans early during infection in vivo , we ascribe a novel role to these two NADPH oxidases in recruitment of leukocytes to the site of C . albicans infection . In addition to our findings , abundant recent work challenges the narrow view of ROS as solely microbicidal and implicates NADPH oxidase-produced ROS in a range of other functions such as autophagy , neutrophil extracellular traps , tryptophan metabolism , kinase signaling , neutrophil recruitment to the endothelium and epithelium , and inflammasome activation [9] , [19] , [35] , [48]–[56] . In this context , it is notable that we did not find frequent LC3-associated phagocytosis ( LAP ) of fungi , in contrast to what has been observed in vitro [35] . Perhaps most relevant to our findings is recent work suggesting a role for the phagocyte NADPH oxidase or Duox in enabling neutrophil chemotaxis in vivo [16] , [19] , [42] , [52] . Our work highlights how this latter function of NADPH oxidases in chemoattraction can play an important role in immunity to the fungal pathogen C . albicans in this zebrafish model . The implication of both Phox and Duox in driving early immune recruitment was unexpected , despite their shared capacity to produce hydrogen peroxide . The consequences of pan-NADPH oxidase inhibition by DPI are no more dramatic than inhibition of either Phox or Duox , suggesting that inhibition of both enzymes is comparable to inhibition of either one alone . The largely non-redundant phenotypes of p47phox and Duox knockdowns imply that these two enzymes may work together in the same pathway or complex , but it is unlikely that p47phox directly control Duox , as Duox regulation is only known to occur through Ca2+ [57]–[59] . Alternatively , they may play separate , indispensable roles in the same or different tissues . We do not know if the relevant activity of these NADPH oxidases is within leukocytes , epithelial cells , or both cell types . Recent work identified a novel function for Phox within human and mouse neutrophils in chemotaxis toward defined chemoattractants in vivo and in vitro and toward sterile inflammation in vivo [16] . While Phox is expressed most highly in phagocytes , it is expressed widely and has recently been shown to have important activities in other tissues [57] , [60] . On the other hand , Duox in epithelial cells has an important function in signaling infection and attracting leukocytes to damage [48] , [52] . In zebrafish , Duox has been shown to play a role in neutrophil chemotaxis to wounds , but may not be involved in leukocyte chemotaxis to bacterial infections [15] . Tissue-specific promoters in zebrafish should enable determination of the cell types requiring NADPH oxidase components in fungal infection . If phagocyte oxidase is not required to limit intracellular germination , what mechanisms contain internalized C . albicans ? Together with previous work [31] , our data suggest that NADPH oxidases play an early role in phagocyte recruitment but not in prevention of germination . This suggests that early during infection other phagocyte mechanisms ensure containment and prevention of C . albicans filamentous growth within phagocytes . While phagocyte oxidase is required for the PMA-stimulated respiratory burst , our work here shows this role does not seem to impact early attack on C . albicans , as measured by oxidative stress or phagocyte activation . Consistent with the selective role of Phox-mediated oxidative damage in vivo , only a subset of all C . albicans experiences oxidative stress at a given time during systemic infection in the mouse [32] . Instead , phagocytes have been shown to have the capacity to use NADPH oxidase-independent mechanisms to limit C . albicans growth , limit germination , and in some cases kill intracellular fungi . Notably , NOX2 −/− NOS2 −/− mice missing essential components of the phagocyte oxidase and inducible nitric oxide synthase are still able to limit C . albicans virulence in vivo in a gut infection model and macrophages from these mice can kill C . albicans in vitro [22] . Several lines of evidence also suggest neutrophil proteases are important in damaging C . albicans and limiting its virulence [23] , [25] . Furthermore , C . albicans is sensitive to antimicrobial peptides [61] and neutrophil extracellular traps [26] . New technical advances may enable the determination of the dynamics of fungal killing by these mechanisms in mice , which have never been directly assayed through the type of continuous observation shown here in the zebrafish . What are the consequences of a poor initial immune response to C . albicans infection ? In both normal and NADPH oxidase-inhibited fish , failure to contain C . albicans in the first four hours is tantamount to overall failure to eliminate the pathogen . C . albicans is prevented from germination within neutrophils and macrophages for at least the first 24 hpi . In contrast , extracellular C . albicans germinates into tissue , and the fungus' ability to turn on the EDT1 transcription factor allows both dimorphic switching and prevention of NADPH oxidase-independent containment . Both of these activities play important roles in virulence . Switching to hyphal form inhibits phagocytosis [62] and permits tissue invasion [63] , leading to florid filamentous growth that destroys host tissue . Effective immune migration to the infection site is a strong prognostic indicator of survival , regardless of the presence of functional NADPH oxidases . The early role of NADPH oxidases is also emphasized by the effect of a temporary , 4 hour , blockade through DPI incubation . In this case , despite wash-out of the drug afterwards , phagocytosis efficiency at 4 hpi predicted survival . Thus , later activity of NADPH oxidase , which mediates fungal oxidative stress , is still not effective at stopping the infection . The importance of early immune response is clear in the mouse , as well , where delayed immune infiltration into the kidney plays an important role in making this organ much more susceptible to C . albicans proliferation [64] , whereas late-infiltrating neutrophils [65] cause excessive damage , a scenario that is more than just a case of “too-little , too-late” . If NADPH oxidases are not required for phagocyte chemotaxis to bacterial infections in the zebrafish [15] , [18] , why are they required for recruitment to C . albicans ? Response to these other infections likely involves redundant immune mechanisms including ROS and other chemotactic cues for ensuring leukocyte chemotaxis . The finding that fungal containment of the yeast-locked edt1Δ/Δ mutant is not significantly reduced by NADPH oxidase-inhibition suggests that specific pathogen mechanisms may limit the immune response and force a dependence on ROS . Recent in vitro work shows that C . albicans hyphae are harder to phagocytose than yeast , which suggests that fungi that escape phagocytosis and germinate extracellularly early during infection in the zebrafish will be able to resist phagocytosis later [62] . Accordingly , this may impact the phagocytosis of yeast-locked edt1Δ/Δ mutants . The potential link between dimorphic switching and blockade of ROS-independent phagocyte chemotaxis suggests yet another way that the ability to switch to an invasive hyphal form may contribute to C . albicans virulence in vivo , as has been shown in both mucosal and disseminated candidiasis in the mouse [66] , [67] . Additional virulence mechanisms depend on the germination program , because even poor early phagocytosis of edt1Δ/Δ mutant fungi did not always lead to mortality . There are several known mechanisms whereby C . albicans regulates interaction with host leukocytes . Live C . albicans blocks phagocyte ROS production [68] , [69] , and hyphally derived soluble products block PMN activation [70] . There are a number of transcriptional pathways co-regulated with the EDT1-mediated yeast-to-hyphal transition , and these may contribute to the differential virulence of the mutant . The transparent model exploited here may prove useful in dissecting how host and fungal determinants govern leukocyte chemotaxis early during infection . While our work using the hindbrain ventricle model of infection is the first to show this new role of NADPH oxidases in early fungal containment , dependence on this mechanism may be pathogen- , developmental- or tissue-specific . Our findings that neutrophil migration to the swimbladder in 4 dpf old zebrafish is NADPH oxidase-independent , in contrast to the NADPH oxidase-dependence in the 2 dpf old hindbrain ventricle model suggests that different mechanisms may be at play in different tissues and/or at different times of development . Thus , the central nervous system may rely more heavily on ROS to drive phagocyte recruitment than other tissues such as the swimbladder mucosa , which shares common ontogeny and gene expression with the mammalian lung [71] , [72] . The tissue-specificity we observe is mirrored in recent work with a different fungal pathogen , Cryptococcus neoformans , which suggests that immune cell recruitment mechanisms are different in the brain and lung [73] . In addition , the lack of NADPH oxidase-dependence in neutrophil responses to bacterial infection of the otic vesicle at 3 dpf suggests that there may be developmental- and/or microbe-specific roles of these enzymes in early immune response [15] . A negative role for NADPH oxidase in neutrophil chemoattraction has been suggested by several mouse models of sterile inflammation as well as in chronic granulomatous disease [9] , although its potential early role in response to fungi remains to be tested . There are likely to be multiple roles of NADPH oxidases in immunity , but our non-invasive time-lapse experiments highlight a previously unappreciated early role for Phox and Duox in leukocyte activity that recent work suggests may be evolutionarily conserved in mammals [16] . Non-invasive imaging of spatiotemporal immune responses with photoswitched Kaede-expressing macrophages revealed a reduced mobility of macrophages with internalized fungi . This could reflect different types of macrophages at the site of infection , similar to the M1 and M2 macrophages in mammals that express different sets of immune receptors and therefore have differential ability to uptake microbes [40] , [74]–[77] . M1 and M2 macrophages also differentially express cytoskeletal modulators such as cadherins [78] . Alternatively , the relatively low mobility of macrophages with internalized fungi could result from macrophage differentiation upon activation in response to pathogen recognition . Differentiation-induced loss of motility is a well-characterized developmental paradigm [79] well described for border cells in Drosophila melanogaster [80] and neural crest in zebrafish [81] . Consistent with this idea , phagocytes have been shown to undergo significant cellular rewiring upon pathogen recognition in vitro [82] , and activation of receptors alters the cytoskeleton and thereby macrophage motility in vivo and in vitro [83] , [84] . Regulation of phagocyte motility in response to pathogen recognition is still poorly understood in vivo , yet may play important roles in pathogen containment and fine-tuning of the host response . As illustrated in our model ( Fig . 7 ) , these results suggest several new facets of the role of NADPH oxidase in early events of the hindbrain ventricle C . albicans infection . First , ROS produced by Phox and Duox are required for recruitment of phagocytes to the infection site . Second , the continued recruitment of phagocytes to the infection site requires NADPH oxidase activity . Third , attraction of phagocytes to the yeast-locked edt1Δ/Δ mutant does not require ROS and may be mediated by fungal products and/or host chemoattractants that are produced in an ROS-independent fashion . Fourth , effective recruitment of phagocytes leads to efficient phagocytosis , containment of fungi , and survival of the challenge . This dynamic picture challenges the existing paradigm for NADPH oxidase-mediated immune responses and highlights a need for further research into host and fungal regulation of chemotaxis and containment .
Zebrafish were kept in recirculating systems ( Aquatic Habitats , Apopka , FL ) at the University of Maine Zebrafish Facility . All zebrafish care and husbandry were performed as described previously [85] . All zebrafish care protocols and experiments were performed in accordance with NIH guidelines under Institutional Animal Care and Use Committee ( IACUC ) protocol A2009-11-01 . Larvae were grown at a density of 50/dish in 10-cm Petri dishes containing 60 ml of egg water ( deionized water with 60 mg/L Instant Ocean salts [Spectrum Brands , Mentor , OH] ) . The fish strains used were wild-type AB and the transgenic strains Tg ( mpx:GFP ) i114 [43] expressing enhanced green fluorescent protein in neutrophils , Tg ( mpeg1:GAL4/UAS:Kaede ) expressing the photoswitchable fluorescent protein Kaede in macrophages [38] , and Tg ( GFP:Lc3 ) expressing enhanced green fluorescent protein fused to LC3 [37] . The Tg ( mpeg1:GAL4/UAS:Kaede ) line has been extensively characterized and shown to provide macrophage-specific gene expression at this developmental stage [38] . Given the infection site in the hindbrain , it is also important to note that differentiation of microglia does not begin until 70 hpf , nearly 36 hours after the prim25 stage used in this infection model [86] . Egg water was supplemented with 0 . 3 mg/L methylene blue for the first 24 hours to prevent microbial growth . Larvae were cleaned by changing the egg water daily . For Tg ( mpeg1:GAL4/UAS:Kaede ) overnight imaging experiments , zebrafish were raised in water containing 15 µg/mL of 1-phenyl-2-thiourea ( PTU ) ( Sigma-Aldrich , St . Louis , MO ) to prevent pigmentation . C . albicans strains were grown on yeast-peptone-dextrose ( YPD ) agar ( Difco; 20 g/L peptone , 10 g/L yeast extract , 20 g/L glucose , 2% agar ) . For infections , liquid cultures of C . albicans were grown overnight with shaking in YPD broth ( DIFCO; 20 g/L peptone , 10 g/L yeast extract , and 20 g/L glucose ) at 37°C . Overnight cultures were washed three times in calcium-and magnesium-free phosphate-buffered saline ( PBS; Lonza , Walkersville , MD ) , counted on a hemocytometer , and adjusted to a concentration of 1×107 cells/ml . OxYellow-T , CAF2-dTomato , edt1Δ/Δ-dTomato and edt1Δ/EDT1-dTomato strains were created by transforming the original CTA1-GFP [32] , CAF2 [87] , edt1Δ/Δ and edt1Δ/EDT1 [88] , [89] strains with a construct containing the PENO1 promoter , a codon-optimized dTomato gene , the TTEF terminator , and a NATr marker . This dTomato construct is described in more detail in Gratacap et al . [44] . Transformants were selected on 100 µg/ml nourseothricin , and screened both for correct integration at the ENO1 locus by PCR and for bright fluorescence by flow cytometry . Modified antisense oligonucleotides ( MOs ) designed to knock down translation of p47phox ( NCF1: 5′-CGGCGAGATGAAGTGTGTGAGCGAG , [31] ) , or to block splicing of duox ( 5′-AGTGAATTAGAGAAATGCACCTTTT , [42] ) were synthesized by Gene Tools ( Philomath , OR ) . Morpholinos were reconstituted in nuclease-free water , and appropriate dilutions were stored in Danieau buffer ( 58 mM NaCl , 0 . 7 mM KCl , 0 . 4 mM MgSO4 , 0 . 6 mM Ca ( NO3 ) 2 , 5 . 0 mM HEPES , pH 7 . 6 ) . A standard control morpholino from Gene Tools was used at the indicated doses in all experiments . Morpholinos were injected into 1-cell embryos in 5 nl volumes to achieve a final dose of 2 . 5 ng of p47phox , and 100 µM ( 4 . 25 ng ) duox . Stocks were prepared in 0 . 01% phenol red for visualization of injection success . For duox MO experiments , embryos were co-injected with 300 µM ( 11 . 7 ng ) p53 morpholino , 5′-GCGCCATTGCTTTGCAAGAATTG , as previously described [42] . Infections were carried out as described [31] . Zebrafish at the prim25 stage ( approximately 36 hours post fertilization ) were staged according to the method of Kimmel , et al . [90] , manually dechorionated , and anesthetized in Tris-buffered tricaine methane sulfonate ( tricaine; 200 µg/ml ) ( Western Chemicals , Inc . , Frendale , WA ) . For infection , 5 to 10 nL of PBS or C . albicans suspension at 1×107cells/mL in PBS was microinjected through the otic vesicle into the hindbrain ventricle to achieve a dose of approximately 10 yeast/larva . Fish were screened by microscopy immediately post-infection to ensure correct inoculum sizes and injection location . Two hundred mpx:GFP zebrafish were collected and incubated overnight in E3 media plus methylene blue at 33°C in 15-cm Petri dishes . Fish were kept at a density of 120 fish per dish with 150 mL E3 . At 1 dpf , the media was changed to E3+PTU ( 20 µg/mL final concentration ) and the fish were incubated for 3 days at 33°C . At 4 dpf , mpx:GFP fish were screened for overall number and distribution of neutrophils; only fish with a homogenous and representative number of neutrophils , as well as an inflated swimbladder were selected . Two groups of 90 fish were treated with vehicle ( 0 . 8% DMSO ) or DPI ( 100 µM ) for one hour in the dark . All fish were anaesthetized in tricaine ( as previously described ) . Thirty ( 30 ) fish from each group were left un-injected , 30 fish from each group were injected in the swimbladder with 3 nL of PBS and the remaining 30 fish from each group were injected in the swimbladder with 3 nL of 1 . 5×107 CFU/mL of C . albicans CAF2-dTomato C . albicans . The latter group was immediately screened by epifluorescence in 96-well-plate imaging dishes ( Greiner BioOne SensoPlates ) . Fish with an inoculum between 5 and 20 yeast cells in the swimbladder were selected and incubated for 4 hours in E3+vehicle ( DMSO ) or DPI in the dark . Fish were then imaged by confocal microscopy in 0 . 4% low melt agarose+tricaine+vehicle ( DMSO ) or DPI . The number of neutrophils at the site of infection was recorded and 30 z-slices were acquired through the swimbladder ( approximately 100–150 µm depth ) in the green ( 488/510 nm ) , red ( 547/618 nm ) and differential interference contrast ( DIC ) channels from representative fish . Images are composites of maximum projections for the red and green channels ( number of slices indicated in the figure legends ) and of a single middle slice for the DIC channel . An Olympus IX-81 inverted microscope with an FV-1000 laser scanning confocal system was used for confocal imaging ( Olympus ) . Objective lenses with powers of 4×/0 . 16 NA , 10×/0 . 4 NA , 20×/0 . 7 NA , and 40×/0 . 75 NA were used . Larvae were anesthetized in Tris-buffered tricaine methane sulfonate ( 200 µg/ml ) and further immobilized in a mixture of 0 . 5% low-melting point agarose ( Lonza , Walkersville MD ) in egg water including the same amount of tricaine . Images are overlays of fluorescence image panels ( red-green ) or overlays of differential interference contrast ( DIC ) and fluorescence images . dTomato , EGFP , unconverted and photoswitched Kaede were detected by optical filters for excitation/emission at 543 nm/610 nm , 488 nm/510 nm , 460–480 nm/495–540 nm , and 540–580 nm/610 nm , respectively . To photoswitch Kaede locally , larvae were imaged at 40×/0 . 75 NA with only the hindbrain ventricle being exposed to laser light . Larvae were then subjected to 15 minute exposure of violet light by scanning of the 405 nm laser at 5% power using the Fluoview X-Y repeat setting . Images post-photoswitching were captured by 10×/0 . 4 NA , 20×/0 . 7 NA , and 40×/0 . 75 NA to confirm only macrophages in the hindbrain ventricle had been photoswitched from green to red fluorescence . Mortality of infected photoswitched larvae ( 37 . 5% ) compared well to mortality of infected non-photoswitched larvae ( 30 . 3% ) , suggesting that photoswitching does not cause significant damage to the zebrafish ( Fig . S4 ) . For all time courses larvae were kept in low-melting-point agarose with egg water and tricaine in cover glass-bottom 24-well dishes ( MatTek , Ashland MA ) , kept at a constant temperature of 28°C , and imaged over time with an Olympus-FV-1000 laser scanning confocal system . Z-stacks from time course images were manually analyzed and the number of C . albicans and fluorescent phagocytes were quantified . Six groups of thirty AB prim25 fish were anaesthetized with tricaine as previously described . Two groups of thirty fish were left uninjected , two groups of thirty fish were injected in the hindbrain with 3 nL of PBS and two groups of thirty fish were injected in the hindbrain with 3 nL of 1 . 5×107 cfu/mL of CAF2-dTomato C . albicans . One group for each treatment was incubated for 1 hour in a solution of E3+in a final concentration of 500 ng/mL H2DCF-DA ( Molecular Probes , Invitrogen ) and 0 . 2% DMSO , the other group in vehicle control ( E3+0 . 2% DMSO ) . Fish were then anaesthetized with tricaine and imaged in a 96-well-plate imaging dish by confocal microscopy . Particular attention was paid to ensure quantitative imaging in the green channel . Thirty slices were acquired through the hindbrain in the green ( 488/510 nm ) , red ( 547/618 nm ) and DIC channels from representative fish . For chemical inhibition of NADPH oxidase , larvae were manually dechorionated at 36 hpf ( prim25 stage ) and pre-incubated for two hours in 100 µM diphenyleneiodonium ( DPI ) ( Enzo , Farmingdale NY ) containing 0 . 8% DMSO , or 0 . 8% DMSO control . After pre-incubation , larvae were infected with C . albicans as previously described and kept in low-melt agarose with tricaine and DPI or DMSO throughout imaging . For mortality assessment from C . albicans infection , larvae were removed from DPI at 4 hpi and incubated in 24-well plates in egg water at 28°C overnight . For PBS mock injection experiments , prim25 Tg ( mpx::EGFP ) i114 larvae were dechorionated and screened by epifluorescence microscopy for abundant EGFP-expressing cells in the caudal hematopoetic tissue . Larvae were then injected with 5 nl of phosphate-buffered saline ( Lonza , Inc . ) or treated the same but not injected; post-injection , fish were incubated in 0 . 8% DMSO or 0 . 8% DMSO+100 µM DPI and neutrophil numbers were quantified by confocal microscopy at 4 hours post-injection . For α-tocopherol experiments , larvae were immersed in 100 µM α-tocopherol ( Sigma-Aldrich , St . Louis MO ) in 1% DMSO or 1% DMSO ( control ) from two hours pre-infection and throughout imaging . Fish were imaged from 3–6 hpi by confocal microscopy as described above . Data on GFP-Lc3 localization from all experiments were pooled and analyzed by Fisher's exact test .
|
Over 45 years ago chronic granulomatous disease ( CGD ) was ascribed to a failure of neutrophils to mount a respiratory burst , and it is now known to result from primary genetic deficiencies in the phagocyte NADPH oxidase complex . Recent work suggests that reactive oxygen species produced by NADPH oxidases have other important functions as diverse as maturing hormones and promoting protein kinase signal transduction . Candida albicans is an opportunistic pathogen that preys on immunocompromised patients to cause lethal candidemia . We used the transparent zebrafish larva to describe a novel function of both phagocyte oxidase and dual-specific NADPH oxidase in directing phagocyte recruitment to C . albicans infection foci . We demonstrate that NADPH oxidase-dependent attraction of neutrophils and macrophages is instrumental in effective containment of yeast within phagocytes , which prevents the yeast-to-hyphal morphogenetic switch and limits mortality . Remarkably , when the fungal morphogenetic switch is prevented by mutation , NADPH oxidase activity is no longer required for effective fungal containment . Our study suggests that defects in CGD may extend beyond reduced microbial killing by superoxide to include impairment of chemotaxis , and provide a basis for exploring this alternative function in mammals .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"&",
"Methods"
] |
[] |
2013
|
NADPH Oxidase-Driven Phagocyte Recruitment Controls Candida albicans Filamentous Growth and Prevents Mortality
|
When patterns are set during embryogenesis , it is expected that they are straightly established rather than subsequently modified . The patterning of the three mouse molars is , however , far from straight , likely as a result of mouse evolutionary history . The first-formed tooth signaling centers , called MS and R2 , disappear before driving tooth formation and are thought to be vestiges of the premolars found in mouse ancestors . Moreover , the mature signaling center of the first molar ( M1 ) is formed from the fusion of two signaling centers ( R2 and early M1 ) . Here , we report that broad activation of Edar expression precedes its spatial restriction to tooth signaling centers . This reveals a hidden two-step patterning process for tooth signaling centers , which was modeled with a single activator–inhibitor pair subject to reaction–diffusion ( RD ) . The study of Edar expression also unveiled successive phases of signaling center formation , erasing , recovering , and fusion . Our model , in which R2 signaling center is not intrinsically defective but erased by the broad activation preceding M1 signaling center formation , predicted the surprising rescue of R2 in Edar mutant mice , where activation is reduced . The importance of this R2–M1 interaction was confirmed by ex vivo cultures showing that R2 is capable of forming a tooth . Finally , by introducing chemotaxis as a secondary process to RD , we recapitulated in silico different conditions in which R2 and M1 centers fuse or not . In conclusion , pattern formation in the mouse molar field relies on basic mechanisms whose dynamics produce embryonic patterns that are plastic objects rather than fixed end points .
The emergence of ordered patterns in multicellular organisms has been a major field of research in developmental biology , revealing a diversity of pattern formation mechanisms . While some patterns appear simultaneously ( e . g . , Drosophila segments , mouse hair ) , others appear sequentially ( e . g . , feathers on chicken’s back ) , most often as the structure grows distally ( e . g . , short-germ insects’ segments , somites , limbs proximodistal elements , palatal rugae ) . Several types of patterning mechanisms have been proposed . Some rely on a prepattern , like the “positional information , ” model in which a gradient of a signaling molecule is turned into a more complex pattern by interpreting the varying concentration at each position in space [1 , 2] . Others rely on self-organization , resulting in spontaneous pattern formation as seen in reaction–diffusion ( RD ) ( Turing ) mechanisms or upon chemotaxis ( see below and [3–5] ) . Depending on the mechanism , temporal dynamics of pattern formation have been more or less emphasized . Sequential formation requires the consideration of temporal aspects that can be neglected when the pattern forms at a glance [6 , 7] . Spontaneous pattern formation results from the internal dynamics of the system , which naturally places the focus on the temporal dynamics . For example , the work of Salazar-Ciudad and Jernvall has emphasized the role for temporal changes in system conditions during 3D morphogenesis when patterning and growth are coupled: patterning at time t modifies the 3D geometry of the system through growth , and this will influence downstream patterning at time t + 1 [7 , 8] . In contrast , positional information has been mostly associated with static representations , for example , in the French flag model [2 , 3] . In most cases , however , patterning is viewed as a directional temporal process: from a prepattern or a spatial heterogeneity emerges the final pattern , which is then stabilized . It is , however , questionable whether biological systems , which result from a historical , contingent process , proceed in such a directional manner , or if transient patterns can be constructed and deconstructed during embryogenesis until the final pattern is formed . Recently , a careful reexamination of the example of simultaneous pattern formation , namely the formation of Drosophila gap gene expression pattern , revealed that , as maternal inputs decay , gene expression patterns change with important consequences for the final pattern [9] . To our knowledge , other examples are lacking . Here , we studied the question in the model of sequential patterning of mouse molars . The search for the general mechanisms generating patterns in biology has been greatly influenced by the theoretical work of the mathematician Alan Turing [4 , 5 , 10] . The generalization of this work has led to many classes of RD mechanisms , in which two ( or more ) molecules characterized by a different spatial range of action and a given topology of interaction can self-organize a stable pattern but also exhibit behaviors such as oscillations or propagating waves [4] . The most iconic example is the case in which a short-range activator that self-amplifies and activates its own long-range inhibitor can create spots , stripes , or labyrinths . Recently , it has been shown that many biological systems such as hair and feathers [11 , 12] , the rugae of the palate [13] , digits [14] , and somites [15] exhibit features of RD mechanisms . This should not be taken too strictly , however . Geirer and Meinhardt pointed out that any process involving local self-enhancement and lateral inhibition has the potential to drive spontaneous pattern formation [16] . For example , color pattern formation in zebrafish can be explained by RD models but at least partly involves cell interactions rather than the diffusion of biomolecules [17 , 18] . Pattern formation can also arise from purely chemotaxis-mediated self-organization . When cell movement is driven by concentration gradients of chemotactic cues , positive feedbacks between cell density and chemo-attractant production are known to enhance the local concentration of cells and may result in self-sustained aggregation [19–21] . Chemotaxis plays a prominent role in feather formation [22] , and this is likely also the case in most other epithelial appendages such as hair [11] . Mouse molars are a good example of repeated structures that form through sequential pattern formation . Mice have only three molars per quadrant , separated from incisors by a diastema , as canines and premolars have been lost in the evolution of mouse lineage [23] . Molars develop sequentially from the first , most anterior molar ( the first molar , M1 ) to the third , most posterior ( the third molar , M3 ) . They develop from a cylinder-shaped invagination of the dental lamina , the so-called dental epithelium [24–26] , where tooth-specific signaling centers , called primary enamel knots ( PEKs ) , are patterned . PEK formation in the epithelium requires signaling from both the epithelium and the mesenchyme [27] , including a mechanical signal induced by mesenchyme condensation [28] . These signaling centers then drive the formation of individual teeth by promoting “cap” formation , the process by which the underlying condensed mesenchyme gets surrounded by the epithelium . Indirect evidence that activation-inhibition mechanisms determine sequential formation of these signaling centers comes from the similarity of tooth formation to other epithelial appendages [29] , namely hair and palatal rugae , whose patterning is clearly ruled by Turing-type mechanisms [11 , 13] . The most direct evidence is a study by Kavanagh and colleagues [30] , showing that when tissue that will form the second molar ( M2 ) is separated from M1 , M2 forms earlier and becomes larger . The earlier development of M2 is also seen when the developing molar row is stimulated with Bmp4 or ActivinβA , which are activators of tooth formation [30] , or upon interference with Shh , an inhibitor of tooth formation [31] . This provides evidence that the first molar somehow acts as a source of inhibitor for the next forming molar . It is noteworthy that three genes encoding inhibitors of tooth development ( Shh , Sostdc1 , Follistatin ) are found in three large genomic regions ( quantitative trait loci ) that affect the size relationships between molars [32] . The sequential patterning of mouse molars in the lower jaw ( Fig 1A ) involves two transient signaling centers [33] that fail to drive proper cap transition , yet form morphologically distinguishable buds ( called MS for mesial segment , and R2 for rudiment 2 ) [33 , 34] . These buds might be vestiges of lost premolars [33 , 35] . Monitoring these signaling centers via Shh expression revealed that the first of these transient signaling centers called MS initiates sequential patterning and then disappears [33] . Subsequently , the R2 signaling center forms . As it starts vanishing , M1 early signaling forms posteriorly [33] . Soon after , R2 and M1–early signaling centers are encompassed in a giant Shh-expressing signaling center [33 , 36] . Here , we will refer to this large center as the mature M1 signaling center . A similar situation with two abortive buds ( called R1 and R2 ) has been noticed in the upper jaw [26] . Their signaling centers have not yet been characterized , although they are morphologically more apparent than in the lower jaw . Interestingly , mutations in genes affecting various developmental pathways ( FGF , Shh , Wnt , BMP , and Eda pathways ) lead to a supernumerary tooth in front of M1 , as if it were a premolar [37] . Studies that specifically addressed the question found that the R2 signaling center was rescued and enabled to form a tooth [36 , 38–42] . The picture is thus fairly complex , especially because we lack direct evidence for the dynamics of activation-inhibition mechanisms that pattern signaling centers in the dental epithelium and promote tooth formation . The Eda pathway has the potential to shed light on these mechanisms . It has a consistent role in activation-inhibition mechanisms in epithelial appendages as distinct as hair , feathers , and teeth [43 , 44] . This role was best studied during pattern formation of mouse guard hairs . The receptor of the pathway , Edar , is first broadly expressed in the epidermis . Concomitantly with patterning of hair signaling centers , it becomes up-regulated in the placodal signaling center and down-regulated in the neighborhood ( Fig 1B ) . Without Edar signaling , no signaling center forms [12 , 45 , 46] , while excessive signaling increases placode numbers and their packing density [12] . Current models posit that the Eda pathway is activated by Wnt , ActivinβA , and BMP4 pathways [12 , 46 , 47] but also feeds back on these and other pathways through the transcriptional activation of their diffusing ligands and inhibitors , for example , WNT10a/b , DKK4 , CCN2/CTGF , Follistatin , and FGF20 [43 , 44] . More recently , Eda signaling was also shown to promote placodal fate by stimulating the centripetal aggregation of epithelial cells [48 , 49] . In teeth , the Eda pathway is dispensable for primary signaling center ( PEK ) formation , but is required for its correct sizing [49–51] . Similarly , it is necessary for correct patterning of the secondary signaling centers controlling cusp morphogenesis [41 , 50 , 52] . Although Eda and Edar mutants have reduced tooth size and cusp numbers , they sometimes form a small supernumerary tooth [53–55] . In gain-of-function mutations , an anterior supernumerary tooth is also found , and teeth are larger , with more cusps [52–54] . In this paper , we aimed at clarifying the temporal dynamics of signaling center formation in the dental epithelium . We studied the temporal dynamics of Edar gene expression , the receptor of the Eda pathway , during molar pattern formation and showed that it recalls the dynamics observed during hair patterning . Based on these data , we built an RD-type model of molar patterning that enables sequential signaling center formation and helps reveal the exquisitely complex temporal interactions leading to the construction and deconstruction of patterns in the developing molar row . Our model explains a counterintuitive result , the rescue of the abortive R2 bud in the inhibitory context of Edar loss of function . Finally , we show that Edar is necessary for the formation of a fused R2–M1 signaling center in the lower jaw only , possibly through a chemotactic effect . We thus showed that patterning is not direct , although it follows simple mathematical rules .
To get insights into molar row patterning , we examined the regulation of the Edar gene ( Fig 1B ) . Because the early period of molar row patterning is complicated by the presence of vestigial signaling centers , we first focused on the patterning of the second molar . Because Edar is exclusively expressed in the epithelium , we performed in situ hybridization on mandibular epithelium that had been dissociated from the mesenchyme , thus providing a 3D view of Edar expression . At 14 . 5 days post coitum ( dpc ) , Edar expression is restricted to the primary signaling center ( PEK ) of the first molar , and no expression is seen in the second molar field , which looks like a "tail" ( Fig 1C ) . At 15 . 0 dpc , the "tail" has elongated and Edar expression is up-regulated in its most posterior part . By late 15 . 5 dpc , it starts restricting to the M2 PEK , just before M2 cap transition occurs ( at 16 . 0 dpc ) . The restriction was concomitant with Shh expression , starting in the PEK of M2 [33] . This dynamic of an initial broad up-regulation of Edar followed by its restriction to a signaling center is reminiscent of what happens during hair patterning ( compare Fig 1B and 1D ) . It suggests that the decision to form a tooth signaling center in the growing molar field proceeds in two phases . First , the whole dental epithelium is activated . This activation results in broad Edar expression , which is so far the only gene whose expression pattern marks the epithelium competent to form a tooth . Second , activation gets restricted spatially and gives rise to a signaling center . This results in the focused expression of Edar and of many other genes , including Shh , which together are known as PEK genes . In this view , Edar expression is a readout of activation levels in the molar field: where activation is high enough , Edar is expressed . To further formalize these ideas , we built a mathematical model of activation in the dental epithelium that could recapitulate the expression pattern of Edar . Switches between different activation states can be identified by Edar expression and suggest that Turing mechanisms occur in the dental epithelium: states first switch from no activation ( no Edar expression ) to broad activation ( broad Edar expression ) and then from broad activation to spatially restricted activation ( focused Edar expression ) . From a mathematical point of view , these spatial behaviors can be modeled as two regimes of the same RD system ( Fig 2A ) . The first regime , referred to in this paper as “the bistable regime , ” admits two homogenous , stable states of a high and a low activation , respectively . This corresponds to the posterior part of the tissue , with transient up-regulation of Edar . Second , the so-called Turing regime is characterized by stable heterogeneous periodic patterns , which emerge from homogeneous patterns . In a minimal pair of reaction equations , the Turing regime is structurally close to the bistable regime ( see Fig 1 in S1 Text ) . Switching between both regimes can be achieved by changing a single parameter value . Because the restricted expression is only seen in the developmentally advanced anterior part and is preceded by broad activation , we impose that the activator has a positive feedback on tissue maturation , resulting in the switch to the Turing regime . Biologically , this means that upon the broad wave of activation , new gene products have been released that modify the activation-inhibition parameters . Our model follows the one-dimensional anteroposterior axis , with posterior domain growth . It describes the time evolution of concentration of an activator and its inhibitor , which diffuse with different speeds and undergo kinetic reactions ( Fig 2A ) . A detailed description of the model and the parameters used in all simulations are found in the supplementary material ( S1 Text ) . Below , we summarize the main characteristics of the model , which exhibits periodic behavior , as shown in a representative simulation ( Fig 2 , see also S1 Movie ) . The tissue grows at its posterior end , and the newly produced tissue then matures exponentially in time , albeit at a slow rate . Maturation is stimulated in the presence of the activator , with some time delay . In zones where maturation reaches a threshold value , the system irreversibly switches from a bistable to a Turing regime . We postulate that this occurs because tissue maturation has an impact on the regulatory feedbacks of the activator-inhibitor system . We opted for a parsimonious modification: a decrease in the rate of auto-inhibition of the inhibitor in the mature tissue . This change of a single parameter value is sufficient to make the transition ( Fig 2B , snapshots 2 to 3 ) . Before this switch can occur , such a simple system first needs to reach high levels of broad activation in the newly grown part ( Fig 2B , snapshot 2 ) . Based on the literature , and as further developed in the Discussion section , epithelia-mesenchyme interactions may play an important part in this broad activation , but through a mechanism that is largely unknown . Here , we simply assumed the existence of an extrinsic component representing the interaction with the mesenchyme . Below a certain threshold of activation , it will act to increase the concentration of the activator . Above a certain threshold , it will feed back negatively on it . This introduces an oscillatory behavior at the posterior end of the domain . Interestingly , although we do not explicitly put this in the model , the oscillatory behavior of the mesenchyme is spatially correlated to the growth of the domain . The transition from “no activation” to “activation” is promoted by the positive feedback from the mesenchyme but will only happen when the domain has grown enough to escape from the influence of the inhibitor from the Turing peak ( Fig 2B , snapshot 1 to snapshot 2 ) . In summary , our theoretical model based on Edar expression involves activation-inhibition mechanisms in the dental epithelium , coupled with periodic activation of the growing dental epithelium . Next , we focused on the dynamics of Edar expression during the complex chain of patterning events ( schematized in 1A ) that precedes the formation of the M1 signaling center , also known as the PEK ( yellow in Fig 1A and 1C ) . The dynamic was partly similar to that observed for M2 patterning , although MS and R2 signaling centers fail to drive cap formation and to form a tooth ( [33] and Fig 1A ) . Indeed , broad Edar expression restricts to these signaling centers in late 12 . 5 embryos for MS and early 13 . 5 embryos for R2 ( corresponding to Shh-signaling center at 12 . 7 and 13 . 3 dpc in [33] and Fig 1A ) . Following the restriction , Edar expression starts again from the posterior part of the dental epithelium ( white arrowhead ) . However , in contrast to what was seen for M2 ( Fig 1C , 15 . 0 dpc ) , the wave of Edar expression did not stop at a distance from the preceding signaling center . Rather , it invaded the whole dental epithelium , including its anteriormost part , thus erasing the previous Turing pattern made by MS and R2 signaling centers ( Fig 3A and 3B , 12 . 5 dpc , and Fig 3C , 13 . 5 dpc ) . In the case of MS , a new signaling center is formed following the wave . Prochazka and colleagues [33] showed by DiI tracing that R2 is just slightly posterior to MS . In the case of R2 , two signaling centers are seen following the wave: R2 recovers and the early M1 signaling center is newly formed . We call this phenomenon a developmental palimpsest , because a palimpsest is a manuscript page that has been scraped or washed off to be used again for a novel text . Here , a first Turing pattern , assimilated as a first text , is erased by the Edar expression wave that “scrapes the first text , ” and a new pattern is formed , representing the novel text . This behavior can be naturally achieved with our model , because waves are generic features of bistable RD systems . Indeed , they exhibit three genuine types of solutions: two constant homogeneous solutions in each of the two states ( high activation or low activation ) , as well as spatially heterogeneous solutions connecting the two stable states and traveling in time . The occurrence of the traveling wave relies on the asymmetry between the two stable states . Furthermore , the direction of the wave is determined by the relative stability of the two states: from the most to the least stable . A representative simulation of this modified model is shown in Fig 3D ( see also S2 Movie ) . The active state is more stable , so that once activation is primed in the posterior part , a wave activates progressively the inactive area ( Fig 3D , snapshot 2 ) . It is important to notice that , before wave initiation , the immature area is maintained naturally in the stable inactive state under the influence of the Turing initial peak . The wave can , however , propagate into the mature tissue and erase the previously formed Turing pattern ( Fig 3D , snapshots 1 and 2 ) . Then , as a consequence of tissue maturation subsequent to wave activation , two activation peaks are formed by a secondary Turing patterning ( Fig 3D , snapshot 3 ) . This would correspond to recovered R2 and to the newly formed M1 signaling center . For this palimpsest to occur , the wave that initiates in the immature ( bistable ) domain should interact with the stable pattern in the mature Turing domain . Both the activation wave and the Turing peak feature stability . As such , understanding their interaction is far from trivial . Several conditions must be fulfilled in order to observe a palimpsest in the numerical tests , which are reviewed in the Supporting information ( S1 Text–section S3 ) . In particular , we found that auto-inhibition , if increased in the bistable regime , strengthened the wave and favored the palimpsest . We also found that it is sensitive to the temporal dynamics . It requires a suitable synchronization between domain growth , posterior activation , wave speed , and maturation rate . Most of the numerous mutants in which a premolar-like tooth forms , supposedly from R2 , have larger- or simply normally sized molar teeth . The teeth of loss-of-function mutants for the Eda pathway are poorly grown , yet a premolar-like tooth can form . As a rescue of R2 is counterintuitive in this context , we decided to re-examine one of these mutants ( EdardownlessJ [EdardlJ] ) in the light of Edar dynamics and of the present model . First , we looked at the dynamics of Edar expression in the EdardlJ mutant to check if R2 was indeed rescued in this mutant . This mutant encodes for a defective Edar protein because of a single amino acid change , but the gene is still transcribed . In contrast with the mutant epidermis , in which Edar expression stays at uniform low levels and hair fails to form ( 56 ) , we still observe Edar restriction to tooth signaling centers ( Fig 4A ) , consistent with teeth being formed . However , the dynamics of activation-inhibition mechanisms were modified in this mutant . We observed high variability between embryonic tooth rows , including left and right rows of the same embryo . This is in line with the high phenotypic variability seen in adults with mutations that affect the Eda pathway , in which two or three , or in rare cases , four , teeth of variable sizes are formed [53–55] . In the lower jaw ( Fig 4A ) , we did not find obvious differences early in 12 . 5 dpc EdardlJ−/− embryos as compared with EdardlJ+/+ embryos , all of them exhibiting restriction of Edar to the MS signaling center ( S1 Fig ) . In both cases , wild type and mutant , Edar is next expressed in the whole dental epithelium ( 13 . 0 dpc , Fig 4A ) . However , no restriction was observed in EdardlJ−/− 13 . 5 dpc embryos as normally seen in their wild-type counterpart . As a matter of comparison , wild-type EdardlJ+/+ embryos sampled with a body weight between 120 and 140 mg showed restricted Edar expression to R2 , and 140–155 mg embryos showed restriction to R2 plus expression in the posterior , corresponding to the starting wave ( S1 Table ) . In contrast , among the 26 EdardlJ−/− embryos sampled between 120 and 160 mg , all showed homogeneous expression in the dental epithelium . In fact , homogenous Edar expression was still observed in most 14 . 0 dpc EdardlJ−/− embryos at a time when homogenous Edar expression is again observed in wild-type EdardlJ+/+ embryos ( Fig 4A , S1 Table ) . From 14 . 5 dpc , a restriction to a signaling center we named T1 PEK ( for Tooth 1 PEK ) was observed in most EdardlJ−/− embryos , with or without expression in the "tail , " while others still displayed more or less continuous Edar expression . We noticed that this signaling center in the mutant is found more posteriorly in the jaw than is the R2 signaling center ( Fig 4B ) . At 15 . 0 dpc , we see either one signaling center with Edar expression in the tail or two signaling centers , named T1 and T2 . Possibly , the latter case is due to approximately simultaneous patterning of two signaling centers from a dental epithelium that was showing continuous expression in the previous stage . At 15 . 5 , T1 has developed into a tooth germ of variable size , from a very small one to a tooth germ equivalent in size to T2 . To conclude , our results show that R2 patterning is both postponed and displaced posteriorly in the EdardlJ−/− mutant , and the resulting signaling center , T1 , persists to form a tooth germ of variable size . This raises the question of whether R2 is rescued in the EdardlJ−/− context in which teeth are poorly grown . We next used our model to explain this nonintuitive observation . Eda pathway loss of function has been shown to increase inhibition in teeth and other appendages [12 , 50] , which is consistent here with the T1 signaling center being patterned further away from the anterior end of the dental epithelium . Therefore , we analyzed numerically the effect of a stronger inhibition , which is achieved in our mathematical setting by decreasing the rate of auto-inhibition , that is , by decreasing the strength of the negative autoregulatory feedback on the inhibitor . Interestingly , this was sufficient to recover qualitative behaviors consistent with the data: ( 1 ) T1 and T2 are formed later than R2 and M1 ( horizontal red dashed line ) . ( 2 ) T1 is displaced posteriorly ( vertical red dashed line ) . ( 3 ) The wave no longer destabilizes T1 . Rather , a Turing pattern is formed that is not subjected to a palimpsest . These results suggest that a new wave of activation associated with the formation of the next signaling center will naturally destabilize any pre-existing signaling center , if inhibition from this pre-existing center is weak enough . In addition , our model does not impose any differences between signaling centers; more precisely , we do not impose that R2 is unstable or weak in essence . Rather , R2 is actively overwhelmed by the activation wave associated with M1 formation . This is a major output of our modeling effort but it is in contradiction with the current thinking , in which the R2 signaling center is considered to be autonomously unstable or weak , possibly due to its position near the diastema . Therefore , we decided to directly test this hypothesis . If the anterior part of the dental epithelium is intrinsically defective for tooth formation , it should not be able to give rise to a fully developed tooth when removed from the early M1 signaling center . On the contrary , if the anterior part is not intrinsically defective but is normally overwhelmed by the M1-forming wave , as we suggest , a tooth should be able to develop from M2 when shielded from the influence of M1 . To test this , we dissected the anterior part of the molar field , comprising R2 , from the posterior part , comprising the early M1 signaling center and the tail . This experiment was performed in Shh–enhanced green fluorescent protein ( EGFP ) mice , where signaling centers are green fluorescent protein ( GFP ) positive . As expected from our predictions , the anterior part developed into a fully growing tooth . Remarkably , the timing of development was advanced compared with the posterior part by 1 day , in accordance with the R2 signaling center having been patterned earlier than the M1 one ( Fig 5 ) . Taken together , these results are consistent with a model in which the R2 region is fully competent to form a tooth but is actively overwhelmed by the forming M1 , resulting in the developmental palimpsest effect described . We then focused on another feature of the mouse dental row , the incorporation of R2 into M1 , which is hypothesized to play a crucial role for the formation of the anterior part of M1 both during development and evolution [33 , 36] . This corresponds to another curious behavior of Edar expression dynamics: the fusion of R2 and early M1 signaling centers soon after recovering from the palimpsest ( Fig 3C ) . We examined in detail the 13 . 5–14 . 5 dpc period , which corresponds to M1 PEK formation . To do so , in parallel with the dynamics of Edar expression , we monitored Shh and Wnt pathway activity ( the latter with the TOPGAL reporter ) , which are recognized markers of tooth signaling centers ( Fig 6A ) . In late 13 . 5 dpc/early 14 . 0 dpc embryos , Shh expression and TOPGAL X-gal staining reveal that the early M1 signaling center starts to form . Some faint Shh expression is occasionally seen in R2 while X-gal staining persists there , presumably in part due to the long half-life of the B-galactosidase . At this stage , Edar is also focused in R2 and early M1 signaling centers , yet low expression can also be seen around . In slightly older embryos , robust Edar expression is seen in a domain spanning both signaling centers and is aligned with the barely formed cervical loops . This Edar expression is followed by anterior expansion of Shh expression , which finally spans the position of former R2 and early M1 signaling centers . We also observed up-regulation of TOPGAL activity during the same period . Altogether , these results show that R2 and early M1 signaling centers are repatterned as a single large signaling center highlighted by Edar expression . This early event prefigures TOPGAL activity and Shh expression relocalization into a large signaling center . Because Edar expression prefigures the anterior expansion of the Shh expression domain and Edar has been shown to regulate Shh [45 , 56 , 57] , we wanted to test if Edar signaling is necessary for anterior expansion and the formation of a large M1 PEK . To specifically test this , we dissected 13 . 0 dpc lower molar regions , when R2 has already formed , and cultured them for 40 hours with or without an interfering antibody , so that we knocked down Edar signaling in the next period of M1 PEK formation . We then visualized Shh expression on isolated epithelia . In untreated samples , we occasionally observed a large M1 PEK similar to in vivo samples ( Fig 6B , state 0 ) , but most often it was split into a R2 spot and an enlarged early M1 spot bridged by a narrow domain of Shh expression ( Fig 6B , state 1 ) . This can be explained by the fact that the dissection process could change the activation-inhibition balance in favor of inhibition ( as proposed in [58] ) , which , according to the predictions of our model , should favor R2 persistence . In treated samples , we mostly recover a small , very posterior signaling center , which corresponds to early M1 Shh expression ( Fig 6B , state 2 and 3 ) . Moreover , the dental epithelium is morphologically different , always showing a bud , followed by a small cap and a "tail" ( states 2 and 3 ) . Thus , in the absence of Edar signaling , Shh expression is lost in the R2 region , and only a small PEK forms that is equivalent in size and position to the early M1 signaling center , and which drives cap transition there . Taken together , our results show that Edar signaling is essential for the formation of a large PEK that encompasses R2 and early M1 signaling centers . Recent studies have pointed out that chemotaxis may play a role in the formation of tooth and hair placodes [11 , 48 , 49] and that the Eda pathway activated centripetal migration in the placodal epithelium ( 45 ) . We noticed that the TOPGAL stainings tend to contract in the anteroposterior direction as the M1 signaling center matures and the distance between R2 and early M1 signaling center decreases ( compare 14 . 0 and 14 . 5 dpc samples in Fig 6A ) . This suggests that cell movements may take part in the formation of the large signaling center . To evaluate this possibility , we incorporated cell motion through chemotaxis in a simple Turing system producing two peaks , thus starting with the situation when R2 and M1 signaling centers coexist ( Fig 6C ) . We assumed that the chemoattractant pattern corresponds to the activator pattern so that cells move towards regions of higher activator concentration . Cell aggregation mediated by chemotaxis requires a positive feedback loop between cell density and chemoattractant concentration [19–20] . In our setting , without the addition of extra molecular entities , this feedback loop can act either directly on the activator-chemoattractant , as in the classical Keller-Segel system [20] , or via down-regulation of the inhibitor , if a higher cell density negatively affects inhibitor concentration . The latter configuration produced the expected behavior: signaling centers first form , and then they fuse into a single large signaling center ( Fig 6E ) . This is consistent with the intuitive idea that fusion requires sufficiently long-range communication between the two Turing peaks and thus a feedback on the long-range diffusing species , which in our case is the inhibitor . Reducing chemotaxis efficiency prevented fusion , consistent with our experiments , in which we reduce Edar signaling , and presumably , chemotaxis ( Fig 6D ) . Nonetheless , adding chemotaxis to our system does not automatically result in fusion into a single spot . In our simulations , we observed that the transition between the absence of fusion and fusion depends on various parameters . The reason is that the activator-chemoattractant has a direct positive effect on the inhibitor but an indirect negative effect on the same inhibitor by means of cell recruitment . These ambivalent effects make chemotaxis able to compensate the segregation due to Turing patterning in some situations or reinforce it in other situations . For example , we observed that chemotaxis can favor pattern formation when the Turing system fails to produce a pattern alone . This suggests that chemotaxis may be part of the normal formation of tooth signaling centers , even when they stay separated . In line with this , we noticed that there is no fusion in the upper jaw ( Fig 6F ) , where the distance between signaling centers is initially larger by about 30% ( Fig 6F and 6G ) . In the model , this small increase of the domain size increases the Turing wavelength and is sufficient to abolish the fusion ( Fig 6E and 6H ) . With our parameter settings , a 15% increase in domain size could prevent fusion under the same chemotaxis efficiency . In conclusion , the nontrivial interaction between chemotaxis and a Turing system appears to be a plausible mechanism to explain the variability in the dynamics of tooth signaling centers between the lower jaw and the upper jaw of the wild type , or between the wild type and an Edar loss of function .
In this study , we have revealed the highly dynamic expression of Edar in the developing molar row . This dynamic is superficially similar to that seen during hair patterning . This is not surprising because hair and tooth patterning share many common features [29 , 59] , making their comparison very instructive . Below , we compare these two systems in light of our results . In teeth , as in hair , Edar expression becomes restricted to the signaling center as it is patterned . We have noticed , however , two substantial differences: ( 1 ) in skin , the initial basal levels of Edar are up-regulated in the placode and down-regulated in its vicinity . This is thought to be pivotal for placode patterning , in which Eda signaling is necessary to stabilize and refine an otherwise labile Wnt-dependent placode prepattern [12 , 46] . In the molar field , Edar expression in the dental epithelium reaches levels pretty similar to restricted expression in the signaling center , suggesting that Edar is actively stimulated in both cases . This up-regulation may rely on ActivinβA , which stimulates Edar expression in tooth cultures [47] , and on the Wnt pathway , which plays a central role in tooth formation and is involved in Edar basal expression in hair [12 , 46] ) . Down-regulation may rely on the BMP pathway , as is the cases in hair [12 , 46] . ( 2 ) We show that this regulation still occurs in the Edar mutant , a major difference with hair , for which the lack of Edar signaling freezes the initial state of uniform basal levels of Edar expression [12] . Self-activation of the pathway thus plays a more minor role , if any , in teeth . We believe that these differences on Edar regulation may reflect differences in the balance of the different processes participating in hair and tooth formation . For the formation of hair placodes , Turing-like mechanisms establish a noisy prepattern , with local sources of FGF signaling . Mesenchyme condensation towards these sources then refines and reinforces the pattern [11] . In mutant conditions that remove the epithelium prepattern , it appears that the mesenchyme can still produce spaced foci of cell condensates [11] . Thus , the mesenchyme is also capable of autonomous self-organization , but this capacity is not fully exploited because of the prepattern imposed by the epithelium . The formation of tooth signaling centers seems to rely on a different equilibrium between the two tissues . The formation of a PEK is highly dependent on mesenchyme condensation , as seen in bud-arrested tooth germs in which condensation fails [28] . Modeling the gene network of epithelium–mesenchyme interactions in teeth also led to the suggestion that both tissues work in concert , rather than one dominating the other [27] . These intrinsic differences may explain why Edar loss of function abolishes pattern formation in the epithelium-dominated context of hair formation but only results in spatiotemporal modifications in the more balanced context of tooth formation . In this study , we assume that the complex spatiotemporal changes in Edar expression highlight waves of activation in the dental epithelium . Each of these waves results in the patterning of a signaling center , and they are reiterated upon posterior growth of the dental epithelium . We note that this growth zone could be the Sox2-positive region shown in [24] . We built an RD mathematical model to describe this behavior . In this macroscopic model , molecules are treated as a continuum and set on a one-dimensional space to model the anteroposterior dimension of molar row formation . We also chose to consider only two types of Turing in-phase molecules , corresponding to an activator and an inhibitor . This is , of course , a high level of abstraction . Tooth genetics has revealed many molecules from the epithelium or the mesenchyme that could participate in the activation-inhibition mechanisms ( with both in-phase and out-of-phase patterns ) . Studies generally choose to focus on only some of them , for the sake of simplicity . For example , Cho and colleagues [31] have shown that perturbation of Shh function releases inhibition on the next forming tooth and , moreover , up-regulates Wnt and FGF pathways and down-regulates Sostdc1 . They then showed that it is theoretically possible to generate a Turing pattern in a distally growing field from such three species: an activator , an inhibitor , and a mediator allowing the negative feedback between the activator and the inhibitor . From this theoretical result , they suggested that Wnt , Sostdc1 , and Shh could play these respective roles to pattern teeth . O’Connell and colleagues focused their work on cross-regulatory relationships between Wnt and BMP signaling [27] . Here , it was not our purpose to identify these molecules . Instead , our modeling effort aimed at providing a theoretical framework for sequential tooth formation , with a special focus on the spatiotemporal dynamics of the system . Moreover , although our model explicitly aims to describe activation in the epithelium ( Edar dynamics ) , this does not mean that the activator-inhibitor couple in our model should be seen as an abstraction for Turing reactions in the epithelium only . We do not rule out that our model could synthesize the activation-inhibition reactions arising from epithelia-mesenchymal interactions and giving rise to the Turing pattern . For example , mesenchymal inhibitors such as Sostdc1 could take their part in the inhibition . One key ingredient in our model is the transition from a bistable regime to a Turing regime , which is facilitated by the structural similarity between both regimes in our activator-inhibitor system . There are many ways to achieve this transition . We opted for decreasing the rate of auto-inhibition of the inhibitor as tissue maturates . However , we believe that our conclusions hold true for various ways of modifying the activator-inhibitor dynamics , provided that the reaction terms follow the picture depicted in S1 Fig . For instance , increasing the activation rate from the activator onto the inhibitor would act similarly . Dedicated experiments would ideally confirm that tissue maturation has an impact on the regulation feedback rates . We note that the maturation-induced modification of RD parameters is not specific to our model . For example , a study by Prochazkova and colleagues [60] proposed that diffusion rates ( rather than feedback rates ) change upon differentiation of tongue papillae , influencing their final size . A second originality of our model came from modeling the observation that new waves of Edar expression repeatedly initiate posteriorly and can wash out the focused Edar expression in the rudimentary signaling centers . This was achieved by means of a bistable wave , which initiates on the posterior side after a long delay of sustained down-regulation . We found that this wave , when interacting with a Turing pattern already established on the anterior side , can destabilize and even erase it . The fate of the Turing mature pattern depends on the details of the model , through the intensity of the feedback regulations as well as the timescales . This spatiotemporal interaction between a Turing pattern and an RD wave is , as far as we know , original . Our model explicitly assumes that the mesenchyme is responsible for periodic activation priming of the newly grown epithelium . This dependence is consistent with a body of evidence showing that mesenchyme activity is necessary for the induction of PEK formation and sequential tooth formation [27 , 30 , 61] . We also know that mesenchyme activity depends on the Msx1-Bmp4 feedback loop [62–64] , which is itself dependent on a mechanical signal provided by mesenchyme condensation [28] . When this loop is defective , sequential tooth formation can stop at different stages , from no tooth forming , only one , or only two instead of three [63 , 64] . It can also simply stall until adequate levels of Bmp4 signaling are reached , as seen in the Barx1 mutant [65] . The mesenchymal Bmp4 signal is part of a Wnt-Bmp regulatory network whose integration drives signaling center formation [27] . It is also known that the mesenchyme produces ActivinβA , a potent inducer of both tooth formation [30] and Edar expression . In the absence of further knowledge about how the mesenchyme could prime the waves of activation observed in the epithelium , we introduced in our model an extrinsic component representing the interaction with the mesenchyme , and chose a parsimonious way to provide it with an oscillatory behavior . For this , we assumed that the mesenchyme activity is stimulated by the activator and feeds back on it in the newly grown area . Below a certain threshold , it will act to increase the concentration of the activator . Above a certain threshold , it will act to decrease the concentration of the activator . This is the main limitation of our current model , which raises the very interesting question of what the mechanisms enabling periodicity in the molar row are . In particular , experimental approaches will be needed to determine if the mesenchyme oscillates and how these oscillations relate to the Edar waves in the epithelium . Can the mesenchyme oscillate on its own , or are oscillations a property of the epithelium-mesenchyme cross-talk along the anteroposterior axis ? For example , one possibility suggested by our model would be that the posterior mesenchyme self-activates when it has grown enough , because it is then far enough from the inhibitory influence of an epithelial signaling center . This could be investigated by cutting experiments that would abruptly remove this inhibitory influence . Despite this limitation , we note that our current model exhibits a relevant feature , because inhibition from the Turing spot locks the bistable system of the newly grown epithelium in the “no activation” state . This means that , in the absence of a wave of activation triggered by the mesenchyme , sequential addition will stop , in contrast with a standard Turing system in a growing field . This is consistent with mutants in the Bmp4-Msx1 axis , in which sequential addition stops after M1 or M2 formation . Although our modeling approach mainly focuses on qualitative insights , we wanted to assess the robustness of pattern formation and developmental palimpsest in our model . We found a suitable model parametrization and tested its sensitivity with respect to patterning ( see S1 Text for details ) . Although the results are generally robust enough to moderate parameter changes ( 10%–50% ) , it is interesting that the developmental palimpsest can be abolished in many ways , changing auto-inhibition but also temporal dynamics and synchronicity between events . This is consistent with the marked tendency of molar row development towards supplementary molar formation: it can happen in mutants of many different pathways; moreover , it often occurs without major changes in other aspects of tooth development . Our model shares some similarities with models of somitogenesis . First of all , almost all somitogenesis models include a clock driving gene expression oscillations , forming traveling waves moving through the tissue [66] . Even cells isolated from the presomitic mesoderm exhibit oscillations [67] . However , whether such a bona fide molecular oscillator will be found in the tooth system remains an open question . We note that tissue-scale oscillations have been observed in limbs , whose development shares similarities with that of epithelial appendages including teeth [68] . We also envision other possibilities relying on tissue properties rather than cell properties , for example , emerging from the cross-talk between the epithelium and the mesenchyme , as suggested above . Second , in the long-prevailing models of somitogenesis , the clock is combined with a gradient of Fgf/Wnt signaling that maintains the oscillations in the posterior part and determines the position where the traveling wave is frozen into a stationary pattern , which will define somite boundaries [69] . Our model does not comprise such positional information . In a more recent model of somitogenesis , traveling wave and pattern formation are produced by a Turing pair with a nondiffusing activator and a diffusing inhibitor [15] . Pattern formation arises when the traveling wave breaks next to the previously formed stripe ( which acts as a stable source of inhibitor ) , and local interactions in this region promote activator increase to form a new Turing stripe in the vicinity of the previous stripe . This model shares an obvious similarity with our model: a Turing pair exhibits different behaviors along the anteroposterior axis . It can be oscillations with traveling wave and Turing in the Cotterel model , versus bistable with traveling wave and Turing in our model . However , the switch between both behaviors arises as a local emergent property next to previously formed stripes in the Cotterel model , whereas it is explicitly introduced in our model as a result of maturation . Moreover , in our model , the oscillations are provided by an exogenous oscillator at the posterior boundary , fulfilling the function of the mesenchyme . We acknowledge that the Cotterel model might apply to the tooth system , and it will be interesting to test if the palimpsest can be obtained with such a model . Our study also shares superficial similarities with the sequential patterning of feathers . In this system , a priming wave of activation is observed in the epithelium , giving rise to a stripe in the back of the chick embryo , which is then broken into a spot pattern giving rise to individual feathers . Pattern formation in the model by Painter and colleagues relies on chemotaxis rather than RD [22]: moderate cell aggregation drives stripe formation in the primed epithelium through an FGF-dependent positive feedback , and strong local aggregation introduces a BMP-dependent negative feedback that contributes to breaking the stripe into spots . The behaviors of the two systems are similar: the broad Edar expression could be compared to the priming wave or the first stripe , and the formation of the signaling centers could be compared to the breaking of the stripe into spots . These models also converge conceptually . Stripe formation in the feather model and Edar activation wave in the tooth model mainly rely on positive feedback . Spot formation and signaling center formation both rely on the introduction of a sharper negative feedback . We take this as an indication that this sequence of activation might be a general property of epithelial appendages ( feathers , hair , teeth ) that can be captured by very different , nonexhaustive models . We also want to stress that our model is meant to recapitulate local activation/long-range inhibition mechanisms rather than specifically RD mechanisms , and we do not exclude that the biological mechanisms it captures are based on chemotaxis , as in the Painter model . Lastly , we would like to emphasize how the current model successfully recapitulates a number of counterintuitive behaviors of the system and informs us on the possible underlying mechanisms . Previous studies had already revealed several complex behaviors in the growing dental epithelium: ( 1 ) the transient patterns of MS and R2 signaling centers , supposedly vestiges of premolar signaling centers; ( 2 ) the rescue of an abortive tooth germ , R2 , in a large number of genetic conditions; ( 3 ) the transient coexistence of R2 and early M1 signaling center , followed by their fusion in a large signaling center in the lower jaw . The present data and our simple model suggest that these complex behaviors are the fruit of rather simple but highly dynamic interactions in the growing tooth field . As viewed from Edar expression , the pattern constituted by MS and , later , R2 signaling centers is erased to give rise to a second wave of patterning , illustrated by a broad Edar expression in the dental epithelium at respectively 12 . 5–13 . 0 dpc and 13 . 5–14 . 0 dpc . This was recapitulated in the model by enabling the bistable domain to form a traveling wave that can destabilize a previously formed signaling center , if inhibition in the latter is not too strong . Aside from recapitulating Edar expression , the traveling wave has more profound implications . Indeed , it implies a first paradigm shift that vestigial buds are not committed to abort as usually thought , for example , due to their proximity with the diastema , thought to serve as a source of inhibitors [70 , 71] , or through the expression of specific molecules [72] . Rather , or on top of that , functional signaling centers are actively competed by the next round of activation as the dental epithelium grows . Such a balance explains why the anterior part of the molar row is very sensitive to environmental perturbations , such as the dissection associated with tooth culture and to genetic perturbations , as many of them result in supplementary tooth formation there . It also explains why even conditions that produce a more inhibitory context than the wild type can produce a supplementary tooth . Indeed , our model predicts that if inhibition is increased , as it is commonly assumed to be in mutants of the Eda pathway , then the Turing pattern can still form yet with a slightly longer wavelength , but the traveling wave is almost immediately suppressed . This is exactly what we document in EdardlJ mutants for the R2 signaling center: it forms more posteriorly , and we see no traveling wave that would erase it . Rather , it persists to form a tooth bud . Our tooth culture experiment with the anterior part of the molar field comprising the R2 signaling center demonstrates that R2 has the potential to fully form a tooth when not actively competed by the M1 signaling center in the wild-type situation . Consistent with our results , Li and colleagues reported that FGF8 application could rescue tooth germ development in the mouse diastema only when it was separated from the molar and incisor buds [73] . In conclusion , our results extend the prevailing model of Kavanagh and colleagues [30] , in which inhibition between forming teeth is unidirectional ( from M1 to M2 , to M3 ) , by showing that inhibition can be bidirectional and subtly dependent on the temporal dynamics of the system . In the wild type , after broad Edar activation at 13 . 5–14 . 0 , a new pattern of Edar restriction forms that is markedly different between the lower and upper jaws . In the lower jaw , independent R2 signaling center and early M1 signaling center are seen either very transiently when looking at Edar expression or for a longer time when looking at Shh expression or TOPGAL activity . These centers are then rapidly fused into a single elongated signaling center . In the upper jaw , Edar restricts to the R2 and M1 signaling centers and remains as such . Therefore , the palimpsest is observed in both jaws , enabling the co-occurrence of two signaling centers , but only in the lower jaw does some additional mechanism enable their fusion into a large signaling center . Here , we introduced chemotaxis to the model because it has been involved in hair placode formation , both at the level of the epithelium [48 , 49] and the mesenchyme [11 , 22] . This was sufficient to recapitulate a number of interesting features: ( 1 ) chemotaxis changes the RD so that the system first makes two peaks that later fuse into a single , larger peak . This is reminiscent of the large M1 signaling center . ( 2 ) This behavior is sensitive to the distance between the initial peaks , and a 15% increase was sufficient to impede fusion . The measured 30% difference between the R2–M1 distance in the lower jaw , where fusion occurs , and the R2–M1 distance in the upper jaw , where fusion does not occur , may thus be sufficient to explain the difference in fate in the two jaws . ( 3 ) Finally , reducing chemotaxis was sufficient to impede fusion . This may explain why , in our culture system , inhibition of Edar activity impedes fusion , although the distance between R2 and M1 does not seem drastically changed . These roles for chemotaxis would require experimental confirmation in our system: This could be tested , for example , by manipulating cultures of tooth germs with drugs impeding actin-based cell movements , as in [48] . Interestingly , we show that chemotaxis plays an ambivalent role in our model . Depending on the conditions , it acts in favor or against the Turing pattern , or it can be relatively neutral . We propose that this ambivalent role contributes to explaining the versatility of this biological system , with regard to genetic and environmental perturbations . An important lesson from the tooth system is that patterning events may be less straightforward than usually thought , and patterns may be dynamically drawn and erased or refined during embryogenesis . In other words , developmental palimpsests may be a common feature . One reason for this is historical . Systems are the product of evolution , and as pointed out by F . Jacob , evolution proceeds as a tinkerer , not as an engineer [74] . Extant patterning mechanisms are thus modified versions of ancestral mechanisms , not purposely designed from scratch . Our study is consistent with recent studies on the fine-scaled temporal dynamics of gap gene patterns in dipterans ( e . g . , a progressive anterior shift of the gap genes pattern ) . As shown here for the Edar mutant , incorporating these dynamics into models provided a better explanation for mutant phenotypes [9 , 75 , 76] . Moreover , these curious dynamics are also likely the vestige of an ancestral mode of segmentation [75 , 77] . In summary , we believe these two systems illustrate that temporal dynamics of developmental systems needs to be studied and , moreover , to be studied in the light of evolution to fully explain how the system reacts to perturbations . Indeed , embryonic patterns can be highly dynamic , and thus dynamics can be essential to the outcome of the patterning process .
Housing of animals and animal experimentations were conducted under animal care procedures in strict accordance with the guidelines set by the European Community Council Directives ( 2010/63/UE ) as well as with the national guidelines ( ID 39/2009 ) . All mice used in this study were killed by cervical dislocation . Experimental procedures relative to Figs 1 , 3 , 4 and 6 were approved by the Institutional Animal Care and Use Committee CECCAPP , Lyon , France ( # ENS-2009-027 et # ENS-2012-046 ) . Experimental procedures relative to Fig 5 were performed under supervision of the Professional committee for guarantee of good life-conditions of experimental animals at the Institute of Experimental Medicine , the Czech Academy of Sciences , Prague , Czech Republic , and approved by the Expert Committee at the Academy of Sciences of the Czech Republic ( permit number: 81/2017 ) . CD1 mice were purchased from Charles River ( Germany and France ) . Other mice have been bred at the PBES ( Lyon , France ) . TOPGAL mice ( Tg ( Fos-lacZ ) 34Efu ) carrying three LEF1/TCF1 binding sites fused to a minimal c-fos promoter driving lacZ expression were backcrossed against CD1 mice for 10 generations [78] . TOPGAL positive mice were screened by standard lacZ staining performed on the first phalange cut from PN4-PN7 newborns . The Edardl-J mice ( FVB background ) were obtained from Denis Headon . They carry a G to A transition mutation causing a glutamate to lysine substitution in the death domain of the Edar protein ( E379K [79] ) . The strain was maintained by crossing heterozygotes with homozygotes , and wild-type and Edardl-J/dl-J mice used in experiments were derived from this same stock . In order to harvest embryos every 12 hours of development , mice were kept under two different day-night light cycles . Mice were mated overnight and vaginal plugs were detected the next morning , noon being indicated as the embryonic day ( ED ) 0 . 5 . Pregnant mice were killed by cervical dislocation and embryos were harvested and weighted as described earlier [33 , 80] . For cut teeth culture experiments , CD1 females were crossed with males carrying the fusion protein Shh-EGFP and Cre recombinase from the endogenous Shh locus ( B6 . Cg-Shhtm1 ( EGFP/cre ) Cjt/J [81] ) , which enabled determination of Shh expression using fluorescence . The breeding pairs B6 . Cg-Shhtm1 ( EGFP/cre ) Cjt/J were purchased from the Jackson Laboratory ( Maine ) . Mice were genotyped using the Jackson Laboratory’s protocols . Mandibles and maxilla were dissected in Hank’s medium and treated with 10mg/mL Dispase II ( Roche ) at 37°C for 1 hour to 2 hours 20 minutes depending on the embryonic stage . Epithelium was carefully peeled and fixed in 4% PFA . Embryonic mandibles , maxilla , or dissociated epithelia were fixed in 4% PFA solution overnight at 4°C and in situ hybridization was done according to a standard protocol . DIG RNA probes were transcribed in vitro from plasmids described elsewhere: Shh [82] , Edar [47] . TOPGAL embryonic mandibles or dissociated epithelia were fixed in 4% PFA for 15 minutes only and stained with X-gal according to a standard protocol . Samples were documented on a Zeiss LUMAR stereomicroscope with a CCD CoolSNAP camera ( PLATIM , IFR128 , Lyon ) or on a LEICA MFA205 stereomicroscope with a DFC450 camera ( IGFL , Lyon ) . The lower molar region of 13 . 0 embryos were dissected and cultured according to methods described in [30] . Dissection was made carefully reproducible in terms of mesenchyme quantity left around the tooth germ . In order to control for possible developmental stage effects , embryo weight of each tooth germ was recorded , and one tooth germ was used as a control , while its contralateral tooth germ was used for treatment . Following a period of 2 hours of recovery , the medium was changed for a new medium supplemented with 5μg/mL of a function-blocking anti-Eda antibody ( ectoD3 [83] ) . Tooth culture was stopped at 40 hours ( S2 Fig ) and epithelia were dissociated for 15–30 minutes with 10 mg/mL Dispase II ( Roche ) at 37°C . Left and right M1 tooth germs of ShhEGFP+ mouse embryos at 14 . 3 dpc were dissected identically from the embryonic lower jaw . In particular , one paid attention that the amount of the mesenchyme was almost identical and it should influence left and right M1 germs in the same way . The left M1 germ was then cut into anterior and posterior parts ( for details , see Fig 5 ) . Both parts were cultured separately on PET track–etched membrane according to the method described previously in [84] . Contralateral intact M1 dissected tooth germ from the same specimen was used as control . Cultures were photographed daily using an inverted fluorescent microscope Leica AF6000 ( Leica Microsystems GmbH , Germany ) , from the day of dissection up to day 6 of culture . The model and the parameters are described in S1 Text . The code written in matlab can be downloaded from https://github . com/monikatwarogowska/teeth-patterning-simulator . In order to run on of the six simulations described in the article , it is sufficient to run the main file ( teethsolver . m ) .
|
Organs , such as teeth , that form regular patterns are of particular interest to developmental biologists . These patterns are established early in the embryo , and it has generally been thought the organs appear in what is their final position . Recent studies that focus on the dynamics of patterning events challenge this view , suggesting that pattern formation can be more complex than previously thought . For example , mouse molars form from “organizing centers , ” which appear , disappear , or fuse in a complex sequence of events , until the final pattern is stabilized . Based on the dynamics of expression of the Edar gene , we built a mathematical model of how tooth “organizing centers” form . We reveal that a newly formed organizing center can actively impair or erase a previously formed one . We call this phenomenon a developmental “palimpsest , ” from the terminology of old manuscripts that were scraped to be reused again . This indirect developmental process likely reflects the evolutionary history of mice , which lost premolars while maintaining their embryonic organizing centers . More broadly , we believe that overwriting or correcting previously established patterns during development might be more common than anticipated , simply due to the fact that developmental programs are modified by incrementation during evolution .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"developmental",
"biology",
"embryos",
"morphogenesis",
"cellular",
"structures",
"and",
"organelles",
"pattern",
"formation",
"digestive",
"system",
"embryology",
"embryonic",
"pattern",
"formation",
"biological",
"tissue",
"signaling",
"centers",
"head",
"chemotaxis",
"cell",
"signaling",
"structures",
"teeth",
"cell",
"biology",
"anatomy",
"jaw",
"epithelium",
"biology",
"and",
"life",
"sciences",
"molars"
] |
2019
|
Modeling Edar expression reveals the hidden dynamics of tooth signaling center patterning
|
A common property of aging in all animals is that chronologically and genetically identical individuals age at different rates . To unveil mechanisms that influence aging variability , we identified markers of remaining lifespan for Caenorhabditis elegans . In transgenic lines , we expressed fluorescent reporter constructs from promoters of C . elegans genes whose expression change with age . The expression levels of aging markers in individual worms from a young synchronous population correlated with their remaining lifespan . We identified eight aging markers , with the superoxide dismutase gene sod-3 expression being the best single predictor of remaining lifespan . Correlation with remaining lifespan became stronger if expression from two aging markers was monitored simultaneously , accounting for up to 49% of the variation in individual lifespan . Visualizing the physiological age of chronologically-identical individuals allowed us to show that a major source of lifespan variability is different pathogenicity from individual to individual and that the mechanism involves variable activation of the insulin-signaling pathway .
A fundamental property of aging in all animals is stochasticity , which refers to the large and unpredictable variability in the lifespan of individuals in a population [1] . For example , human lifespan expectancy at birth in modern societies is about 78 years , but there is a large variability in the age of death of individuals as 29% die after age 85 and 27% die before age 65 ( coefficient of variation is ∼0 . 2 ) [2] . Understanding the reasons that make some individuals die earlier than others would provide key insights about why some succumb to major killers such as infection , cardiovascular disease , cancer and stroke , whereas others do not . The causes of the variability in the aging process are poorly understood . Aging stochasticity can best be studied in model organisms that are genetically identical and can be grown under controlled environmental conditions , such as the nematode C . elegans , a model organism with a normal lifespan of about two weeks [3] . For instance , an isogenic population of worms of the same age grown under identical conditions shows a great deal of individual variability in lifespan , with a coefficient of variation of ∼0 . 24 [4] . Individual animals appear to age at different rates , such that animals that are the same chronological age may have aged to different extents ( younger or older ) and have different physiological ages [5] . Several studies have characterized behavioral , morphological and molecular changes in old worms , These age-related changes include decline in locomotion [5] , [6] , decrease in the rate of pumping of the pharynx [7] , and increase in an age-related pigment called lipofuscin [6] , [8] , [9] . Herndon et al . used electron microscopy to identify ultra-structural changes in different tissues and observed that the nervous system appeared to undergo much less dramatic changes during aging than the muscular system [5] . They also found a stochastic component for age-related decline in individuals . Muscle sarcomeres become damaged in old worms , and this damage can be visualized using a muscle myosin protein tagged with GFP ( MYO-3::GFP ) [5] . DNA microarray analysis has identified genes that change expression with age [10] , [11] , [12] . In some cases , the age-related changes have been shown to be markers of physiological , rather than just chronological age . Individual worms can age at different rates , such that some individuals may be physiologically older or younger than others even though they are the same chronological age . One test of whether an age-related change is a marker of physiological age is to see if it predicts remaining lifespan of individuals of the same chronological age . For instance , high lipofucsin levels in moderately aged worms correlates with short lifespans . Another example is expression of hsp-16 . 2 [13] . Heat shock protects cells and extends lifespan by inducing a number of heat shock response genes , including hsp-16 . 2 . Following heat shock , levels of expression of hsp-16 . 2 correlate with remaining lifespan of individual worms [13] . However , it is not clear whether differences between individuals occur naturally or are due to the heat shock treatment . We wanted to understand the mechanisms underlying aging stochasticity . To pursue this goal we first identified C . elegans genes whose expression in individual worms was predictive of their remaining lifespan . We used sod-3 , one of the genes whose expression was predictive of remaining lifespan , to investigate mechanisms underlying variable aging in individuals . Our results indicate that pathogenicity from Escherichia coli used as food is a major source of lifespan variability due to variable activation of the insulin-signaling pathway .
We wanted to understand the underlying factors causing aging variability in C . elegans . To achieve this goal , our strategy was to find markers that not only change with age , but that are predictive of remaining lifespan of individuals in a synchronous population . A marker whose expression is connected to an age-related process that ultimately limits lifespan should reveal which individuals will die early or live long . We tested eight fluorescent reporters corresponding to genes that change expression with age ( Figure 1A , Figure S1 , Table 1 ) . For each fluorescent reporter we measured fluorescence of individual worms at the point when they had lived ∼50% of their mean lifespan . We then recorded the remaining lifespan of each individual , and compared the level of fluorescence expression with their lifespan . For a sod-3::mCherry reporter ( fluorescent mCherry protein driven by the superoxide dismutase-3 promoter ) , we found that mCherry abundance showed a Pearson correlation of 0 . 57 with lifespan , thus explaining 32% of the variation in lifespan between individuals ( Figure 1B ) . We separated the worms into two groups according to the abundance of sod-3 mCherry , and found that worms with more sod-3 expression lived on average 22% longer than their siblings with less expression ( Figure 1C ) . Besides the sod-3::mCherry reporter , we were able to correlate fluorescence abundance and remaining lifespan using two other sod-3 reporters: a transcriptional GFP reporter containing multiple integrated copies [14] and a transcriptional reporter expressing histone H2B fused to GFP on an extrachomosomal array [10] . The remaining seven reporters showed a Pearson correlation between fluorescent intensity and remaining lifespan , ranging from 0 . 35 to 0 . 51 ( Table 1 ) . In summary , we found a correlation between gene expression and lifespan for eight genes whose expression changes with age . For the remainder of this paper , we focus our attention on sod-3 , which is the best marker of physiological age of those tested . We examined sod-3 expression at different ages to find the optimal time for using sod-3 expression as an aging marker . At ages earlier than day 9 , we found that sod-3 expression can still correlate with remaining lifespan in individual worms , although not as well as in 9 day old worms ( ∼50% of mean lifespan ) ( Table S1 ) . At ages later than day 9 , sod-3 expression is not as useful as a molecular marker because worms begin to show overt signs of aging such as decreased locomotion , decreased pharyngeal pumping and vacuolar appearance . Next , we analyzed sod-3::mCherry expression in the same worm at different ages in a longitudinal assay ( Figure S2 ) . This experiment revealed the age-related downward slope in expression of sod-3 for each individual worm , in addition to levels of sod-3 expression at day 9 . However , using linear regression , we found that the age-related slope of sod-3 expression did not improve the prediction of remaining lifespan beyond the prediction made using sod-3 expression at 9 days of age alone ( p<0 . 21 ) . In summary , we measured sod-3 expression in a variety of ways and found that expression at day 9 was most informative for predicting remaining lifespan in individual worms . sod-3 activity might directly extend longevity by itself , or it might be a marker for other processes in the worm that affect lifespan . sod-3 null mutants are known to have wild-type lifespans [15] , [16] , [17] . We created transgenic lines containing many copies of wild-type sod-3 , and found no differences in lifespan compared to controls ( Figure S3 ) . These results indicate that sod-3 activity does not extend lifespan per se , but rather that sod-3 expression is a reporter of remaining lifespan . As a control , we selected four genes at random from the genome ( mif-2 , his-72 , C08B11 . 3 , and sur-5 ) , measured their expression level in individual worms when they had lived ∼50% of their mean lifespan , and found that none correlated with remaining lifespan as well as the eight age-regulated genes ( Table 1 ) . Thus , not every gene can serve as a marker for aging . Therefore , the correlation between expression of age-related genes and remaining lifespan is not due to a general decrease in gene expression during aging . Finally , we tested whether amounts of the age-related pigment lipofuscin correlated with remaining lifespan in middle-aged worms . Lipofuscin is composed of a set of fluorescent breakdown products that accumulate in the gut with age ( Figure S1 ) [3] , [8] . Previous studies have shown that lipofuscin levels correlate with remaining lifespan; specifically , in a chronologically old population , there is a small amount of worms that can barely move ( Class C ) , and these worms have high levels of lipofucsin and short remaining lifetimes [9] . We tested whether lipofucsin autofluorescence correlates with remaining lifespan in middle-aged worms ( day 9 ) , before overt signs of aging appear . We found no correlation between lipofucsin levels and remaining lifespan ( Table 1 ) . There are many potential sources that could lead to variable abundance of sod-3 between worms , including sources that are intrinsic ( variable activity of transcription factors at the sod-3 promoter ) or extrinsic ( variable effects from the environment ) . To begin to characterize the types of mechanisms that could be responsible for the correlation between variable lifespan and variable sod-3 expression , we examined whether expression of two different sod-3 reporters fluctuated either together or independently in individual worms from a synchronous population . We created a transgenic line expressing sod-3::GFP and sod-3::mCherry , and compared the fluorescence intensity of GFP and mCherry in individual worms . There was a high degree of correlation ( r = 0 . 87 ) between sod-3 expression from the GFP and mCherry reporters ( Figure 2 ) . This indicates that variability in sod-3 expression is caused by a mechanism that is variable from worm to worm , but has similar effects on both reporter genes within an individual worm . We tested several environmental factors that could vary between worms and be responsible for fluctuation in expression of sod-3 and stochasticity in lifespan . One possibility is that variability in individual lifespan might result from differences in feeding on bacteria provided as food on the culture plate , which could lead to heterogeneity in caloric restriction . However , sod-3::GFP fluorescence from worms grown on plates with an even lawn of E . coli correlated with remaining lifespan as well or better than that of worms grown on plates on which the supply of E . coli was restricted to one small spot ( Figure 3A and 3C ) , indicating that heterogeneity in caloric restriction does not contribute to the link between variability in sod-3 expression and lifespan . Another possibility is that variable pathogenicity from bacterial food could lead to variable sod-3 expression and lifespan . E . coli , the common diet for worms , is mildly pathogenic whereas Bacillus subtilis is not pathogenic [18] . Accordingly , worms fed B . subtilis live longer than worms grown on E . coli [19] , [20] ( Figure S4 ) . We tested whether variable pathogenicity could lead to individual differences in sod-3 expression and lifespan in two ways . First , we examined whether growing worms on B . subtilis rather than E . coli reduced variability in sod-3 expression . We cultured worms on E . coli or on B . subtilis for 8 , 12 , and 14 days and calculated variability in sod-3 abundance ( defined as Standard Deviation/mean ) at each age . We found that the variability in sod-3 expression was lower in worms fed B . subtilis than it was in worms fed E . coli at all ages ( Figure 4 and Table S2 ) . Second , we showed that the correlation between abundance of sod-3::GFP and remaining lifespan was considerably lower for worms maintained on B . subtilis than for worms grown on E . coli when worms had lived 50% of their mean lifespan ( Figure 3A and 3B ) . Besides B . subtilis , we obtained similar results using two other non-pathogenic diets , UV-killed E . coli and Caulobacter crescentus . Specifically , worms fed these diets lived longer than E . coli fed worms ( Figure S4 ) , and the correlation between sod-3 abundance and remaining lifespan was lower for worms fed these non-pathogenic diets than for E . coli fed worms when worms had lived 50% of their mean lifespan ( Figure S5 ) . Thus , these results suggest that variable pathogenicity from individual to individual may cause variation in abundance of sod-3 and lifespan variability . The finding that the amount of sod-3 expression present in a middle-aged worm is correlated with its remaining lifespan indicates that events have occurred that affect its future aging trajectory . If so , feeding a worm either E . coli or B . subtilis should have greatest effect when it is young rather than when it is old . To test this , we fed worms one type of bacteria ( E . coli or B . subtilis ) when they were young and then shifted them to the other type of bacteria at day 8 of adulthood . Young worms fed E . coli had short lifespans , no matter what they ate when they were old . Conversely , young worms fed B . subtilis had long lifespans no matter what they ate when they were old ( Figure 5 and Table S3 ) . This result indicates that pathogenicity or some other factor associated with E . coli initiates changes in young worms that affect their time of death later on . Pathogenicity might affect sod-3 abundance through regulation of the insulin-like signaling pathway because this pathway mediates response to pathogen infection [19] , [21] , [22] . Furthermore , the terminal step in the insulin-like signaling pathway is daf-16 , which encodes a FOXO transcription factor that directly regulates sod-3 expression [23] . Fluctuation in the activity of daf-16 FOXO could be a source of lifespan stochasticity in individual worms . We tested this possibility using four approaches . First , we showed that a daf-16 null mutation eliminated the correlation between sod-3 expression and lifespan ( Figure 6C and Figure S6 ) . The daf-16 null mutation also prevents sod-3 from acting as an aging marker at younger ages ( Figure 6D ) . The lack of correlation between sod-3 expression and lifespan in daf-16 null mutants is not simply because sod-3 is less abundant . sod-3::GFP was also less abundant in mutant worms for elt-3 GATA ( Figure S7 ) , another transcription factor that regulates sod-3 expression [10] , but sod-3 abundance in elt-3 GATA mutants still correlated with remaining lifespan ( Figure 6B ) . Second , we tested whether the specific amount of DAF-16 in an individual worm led to corresponding changes in sod-3 expression . The insulin-like signaling pathway controls activity of DAF-16 FOXO primarily by affecting protein phosphorylation , which controls nuclear versus cytoplasmic localization [24] , [25] . However , it is possible that insulin-signaling may also affect the levels of total DAF-16 protein in the cell , which can be measured using a daf-16::GFP translational reporter . We constructed a strain that expressed a fusion of GFP with DAF-16 [24] as well as mCherry under the control of the sod-3 promoter . We measured expression of GFP and mCherry in individual worms . Accumulation of these reporter proteins in individual worms was highly correlated ( Figure 6E ) . This result is consistent with the model that DAF-16 is a direct regulator of sod-3 expression . Third , since daf-16 is a direct regulator of sod-3 expression [14] and expression of these genes is highly correlated with one another , one might expect that independently measuring expression of these two genes in the same worm would be either partially or wholly redundant . To test this , we evaluated a regression model that included expression of both genes to find out if expression of both genes together was more informative about remaining lifespan than expression of sod-3 alone . We found that measuring expression of both genes together or sod-3 alone showed little difference in predicting remaining lifespan ( p = 0 . 26 ) ( Table 2 ) . This shows redundancy in lifespan information from daf-16 and sod-3 expression . Fourth , we tested whether pathogenicity from E . coli used as food induces expression of daf-16 . We fed worms either E . coli or non-pathogenic B . subtilis and measured the level of expression of DAF-16 during aging . We found that expression of DAF-16::GFP was significantly higher for worms fed E . coli than for worms fed B . subtilis at days two , five , and eight of adulthood ( Figure S8 ) . In summary , results from all four experiments indicate that fluctuations in DAF-16 FOXO activity account for much of the individual variability in sod-3 expression and lifespan in individual worms . sod-3 is expressed throughout the worm , including many cells in the head as well as the intestine . The intestine is a primary site for response to pathogenic infection [8] and also for transcriptional regulation of daf-16 FOXO [14] . Three results indicate that the intestine is a major tissue responsive to variable pathogenicity , which in turn determines individual lifespan . First , we measured mCherry fluorescence produced from the sod-3 promoter separately in the head and the anterior intestine . Expression in the intestine showed a higher correlation with lifespan than did expression in the head , demonstrating that intestinal sod-3 expression is the main contributor to the lifespan correlation ( Figure 7A and 7B ) . Second , E . coli and B . subtilis had different effects on sod-3 expression in the intestine , but similar effects on expression in the head throughout lifespan ( Figure S9B and S9C ) . This result is consistent with the idea that differences in lifespan due to growth on E . coli or B . subtilis are due to effects in the intestine . Third , we observed that there were variable levels of sod-3 expression in the head and intestine in individual worms of the same age ( Figure S9A ) . However , sod-3 expression in the head showed little correlation to expression in the intestine in individual worms ( Figure 7C ) . Expression from two markers could provide more information about remaining lifespan than one marker alone if , for example , expression of the markers were to vary independently from each other in individuals . We calculated the correlation between marker expression and remaining lifespan for sod-3 , each of the seven other marker genes , and for each of the seven double combinations with sod-3 ( Table 2 ) . For six genes ( ugt-9 , unc-54 , myo-3 , pha-4 , C26B9 . 5 and elt-3 ) , we found a higher correlation with remaining lifespan using these markers in combination with sod-3 expression than using one marker alone ( p<0 . 05 linear regression ) . The best combinations were sod-3 expression combined with expression from ugt-9 , C26B9 . 5 or pha-4 , which had correlations with lifespan between 0 . 66 and 0 . 70 , accounting between 43% and 49% of the variability in individual lifespan . As discussed above , the seventh gene is daf-16 , which acts in the same pathway as sod-3 and thus daf-16 expression provides redundant information with sod-3 expression about remaining lifespan in individual worms . As a control , we looked at the correlation between expression of sod-3::GFP and sod-3::mCherry in individual worms . As expected , we found that the combined GFP and mCherry expression levels did not improve the correlation with remaining lifespan compared to either sod-3 marker alone ( Table 2 ) . These results show that a combination of aging markers can significantly improve the correlation with remaining lifespan in individual worms .
This paper shows that a common cause of death for worms grown under normal lab conditions is pathogenicity from ingested food . We propose a model in which there is a variable effect of pathogenicity between individuals in an isogenic and chronologically identical population , resulting in differences in lifespan ( Figure 8 ) . E . coli is mildly pathogenic [18] , [19] , [20] , and individuals may be affected to different extents when eating E . coli as food . The pathogenicity from the ingested bacteria primarily affects the intestine . Ingestion of E . coli activates daf-16 FOXO activity via the insulin-like signaling pathway , which induces a beneficial stress response that is protective and extends lifespan . sod-3 is a downstream target of DAF-16 , and expression of sod-3 indicates levels of DAF-16 activation . High levels of sod-3 expression correspond to high levels of DAF-16 activation and longer lifespan , and vice versa for low levels of sod-3 expression . Mutations in sod-3 do not affect lifespan , indicating that sod-3 activity is not in itself functionally important for lifespan but that sod-3 expression is a marker for physiological age because it reports the level of activity of daf-16 FOXO . This paper provides novel insights about a common cause of death for worms grown under normal laboratory conditions , and highlights the complex and interconnected roles of aging and disease in specifying lifespan . Where does the variability in bacterial pathogenicity between individuals arise ? One possibility is that the variability arises from extrinsic environmental differences . For instance , there might be variability in pathogenicity of the E . coli itself in different regions of the plate , which might affect worms to different extents depending on which region they occupy . However , this appears unlikely because worms can traverse all regions of a plate in a day , and because we see a correlation between sod-3 expression and remaining lifespan even when the plates are specially prepared to have even lawns of E . coli . A more likely possibility is that the variability might be intrinsic to the worm itself . Different worms might express different levels of DAF-16 stemming from intrinsic differences such as noise in the expression machinery . When confronted by mildly pathogenic E . coli , some worms could mount a stronger protective response than others , giving rise to correlated differences in sod-3 expression and variability in lifespan . Evidence that bacterial pathogenicity in the intestine is a major source of stochasticity in physiological aging in individual worms comes from comparing the effects caused by feeding worms live E . coli versus three other food sources ( B . subtilis , UV-killed E . coli , and C . crescentus ) . The most obvious differences between these foods is that live E . coli is mildly pathogenic whereas the others are either non-pathogenic or have less pathogenicity . It seems unlikely that the effects on lifespan seen with UV-killed E . coli , B . subtilis and C . crescentus are due to caloric restriction stemming from difficulty in ingesting these foods . Worms fed these three bacteria appear normal in size rather than thin , develop normally and produce a normal brood size [19] , indicating that these worms are not dietary restricted . What happens when worms are switched from E . coli as a food source ( mildly pathogenic ) to non-pathogenic bacteria such as B . subtilis ? According to the model , worms fed E . coli experience variable levels of pathogenicity whereas this variability is reduced or absent for worms fed B . subtilis , resulting in four measurable differences . First , worms have a longer lifespan on non-pathogenic bacteria ( Figure S4 ) . Second , since sod-3 expression is induced by pathogenicity , expression is lower on non-pathogenic bacteria compared to E . coli ( Figure S9B and S9C ) . Third , the variable pathogenicity arising from growth on E . coli generates more fluctuation in sod-3 expression compared to growth on B . subtilis ( Figure 4 ) . Fourth , variable pathogenicity from growth on E . coli results in a correlation between expression of sod-3 and remaining lifespan of individual worms ( Figure 3 ) . Finally , when worms are grown on B . subtilis , they die at a later time and the cause of death is currently unknown . The cause of death for B . subtilis-fed worms could have more , less or similar variability between individuals compared to death from E . coli-derived pathogenicity . Hence , lifespan of worms fed B . subtilis could show more , less or the same amount of variability as worms fed E . coli , depending on the cause of death . In fact , we found that the variability of lifespan of worms grown on B . subtilis is 20% less than that of E . coli-fed worms ( A . S . B . , unpublished observations ) . Our results identify a number of fluorescent reporters that can be used as markers of physiological age . Previously , the most common way to determine the rates of aging in C . elegans was to measure the lifespan of a population of worms . The aging markers presented in this paper will be useful tools to study aging because one can measure the age of individual worms and because expression analysis of fluorescent markers in individual worms is less time-consuming than lifespan analysis of a population of worms . hsp-16::GFP has been previously used as a marker for physiological age [13] , but this marker requires heat shock treatment in order to predict remaining lifespan and this treatment prolongs lifespan in itself . Additionally , drosomycin , hsp22 and hsp70 are partially predictive of remaining lifespan in Drosophila [26] , [27] . In addition to sod-3 , we identified seven other aging biomarkers that could provide information about mechanisms responsible for variability in aging . For six of the biomarkers ( ugt-9 , pha-4 , myo-3 , unc-54 , C26B9 . 5 , and elt-3 ) , combinations of sod-3 and a second marker provide better prediction of remaining lifespan than just one of the markers alone . This result suggests that the different markers may be responsive to aging pathways that are distinct from those controlling sod-3 expression . For example , age-related decrease in expression of the GATA transcription factor ELT-3 is caused by drift of the upstream regulatory network that controls ELT-3 expression [10] . Thus , simultaneous measurement of sod-3 and elt-3 expression in a worm would show the status of two aging pathways ( pathogenic induction and developmental drift ) in that individual , and therefore provide better information about its physiological age and remaining lifespan . For the remaining five aging biomarkers , DNA microarray experiments indicate that they are not regulated by daf-16 FOXO [10] , [28] . Future studies may reveal the source of variation governing the other aging markers described in this paper , which will illuminate other mechanisms that limit lifespan of worms grown under normal lab conditions .
UV-killed E . coli was prepared as described [8] . C . crescentus was grown as described [29] . Since C . crescentus does not grow on NGM plates , plates were seeded with 300 µl of resuspended C . crescentus , allowed to dry , and then subjected to UV irradiation as described [8] . We used three sod-3 fluorescent reporters . The first is a multi-copy insertion reporter that expresses cytoplasmic GFP from the sod-3 promoter ( referred to as sod-3::GFP ) [14] . The second is a low-copy insertion reporter that expresses histone H2B fused to mCherry from the sod-3 promoter ( referred to as sod-3::mCherry ) [30] . The third is a multi-copy extrachromosomal array reporter that expresses histone H2B fused to GFP from the sod-3 promoter ( referred to as extrachromosomal sod-3::GFP ) [10] . daf-16::GFP is a multi-copy integrated translational reporter that expresses GFP at the C terminus of the DAF-16 protein [24] . myo-3::GFP is an multi-copy integrated reporter that expresses nuclear-targeted GFP–LacZ from the myo-3 promoter [31] . unc-54::mCherry and pha-4::mCherry are low-copy integrated reporters that express histone H1 fused to mCherry from the unc-54 or the pha-4 promoter [30] . ugt-9::GFP and elt-3::GFP are multi-copy extrachromosomal array reporters that expresses histone H2B fused to GFP from the ugt-9 or the elt-3 promoter [10] . Promoter::H1::mCherry constructs for C26B9 . 5 and mif-2 were made using promoters from the Vidal lab promoterome [32] in combination with a modified Gateway cloning system [33] vector ( pD4H1cherry [30] ) . Transgenic lines with integrated copies of the reporter were made by microparticle bombardment [34] . Approximately 80 age-synchronized worms were transferred to 1 mM aldicarb-NGM plates for 2–3 hours to induce paralysis [35] . Worms were then individually transferred to FUDR plates . Each worm was then photographed using 20× lens . Images were analyzed using ImageJ . For any given comparison , all pictures were taken on the same day with the same microscope settings . Each plate containing a single worm was labeled with the corresponding picture number and plates were scored for dead worms daily . Lifespan analyses were conducted at 20°C as previously described [36] . Age refers to days following adulthood , and p values were calculated using the log-rank ( Mantel-Cox ) method . Individuals were excluded from the analysis when their gonad was extruded , or when they desiccated by crawling onto the edge of the housing plate . Worm gut autofluorescence was imaged using a 525 nm bandpass filter . A Zeiss Axioplan microscope equipped with Zeiss AxioVision 4 . 6 software was used for quantitative fluorescence microscopy . Images were captured with 10× lens and analyzed using ImageJ . Gut autofluorescence time courses were done using at least 15 worms per age . All pictures were taken on the same day with the same microscope settings . To test remaining lifespan prediction of the age-related slope in the longitudinal sod-3::mCherry expression experiment , linear regression analysis was used with the following model:Yi is the lifespan of worm i , day 9i is the fluorescence of worm i at day 9 , slopei is the slope of fluorescence by age for worm i and εi is a random error term . The coefficients were estimated by least squares from the data . Expression at day 9 ( term ) was statistically significant ( p<3 . 0×10−7 ) . Age-related slope of sod-3 expression ( term ) was not statistically significant ( p>0 . 21 ) . To test the combined effect of multiple measurements within single worms , multiple regression analysis was used with the following model: is the lifespan of worm i , sod-3 marker and 2nd marker are the sod-3 and additional marker expression levels respectively , is a random error term , and and are regression coefficients . Likelihood ratio F-tests were used to determine the significance of and . For example , to test whether addition of a daf-16::GFP second marker adds significant information about lifespan prediction compared to using sod-3::mCherry alone , we test the hypothesis that . This is done by comparing the following two predictions of lifespan: Y is the original matrix of measured lifespans , is the matrix of predicted lifespans from the full model ( 1 ) , and is the matrix of predicted lifespans from the partial model ( 2 ) that leaves out the daf-16::GFP term . With n independent worms , the likelihood ratio test compares the residual sum of squares of the two models and rejects the hypothesis that if , where is the cumulative distribution of an F-distribution with 1 and n-2 degrees of freedom . Using this test , we determined that addition of the daf-16::GFP expression term did not significantly improve the model compared to using sod-3::mCherry alone ( p-value>0 . 2 , and so we cannot reject the hypothesis that ) . A similar approach showed that adding sod-3::mCherry expression significantly improved a model using daf-16::GFP expression alone ( p-value<0 . 001 , so we can reject the hypothesis that ) . The sod-3 gene and regulatory regions ( 2 . 57 kbp fragment ) were amplified by PCR from wild type N2 worm genomic DNA ( forward primer: 5′ ATT CGC AGA AAA AAG TCG TTG C 3′ , reverse primer: 5′ TTT CAG TGT ACC GAG TGA AGT TC 3′ ) . The sod-3 PCR fragment was cloned into the TOPO pCR 2 . 1 vector ( Invitrogen ) . TOPO pCR 2 . 1 plasmid with the coding sod-3 region was coinjected at different concentrations ( 5 ng/µl and 40 ng/µl ) with pha-1 encoding plasmid into pha-1 mutant worms to generate extrachromosomal array sod-3 overexpressing lines . A control line was generated by injecting pha-1 encoding plasmid alone into pha-1 mutant worms .
|
One of the long-standing mysteries in aging is that some individuals die early whereas others die late . The age at which a specific individual will die is difficult or impossible to predict , and thus a fundamental aspect of aging in all animals is that it is stochastic . Aging stochasticity is particularly interesting in model organisms such as C . elegans because they are genetically inbred , can have the exact same chronological age , and can be grown under standard lab conditions . In this paper , we uncover a major mechanism underlying stochasticity in aging . To do this , we first developed a fluorescent aging marker ( sod-3::GFP ) whose expression declines with age and can be used to measure physiological age . In young animals , the level of expression of this fluorescent marker indicates which animals will live longer and which will die earlier . We used this fluorescent aging marker to show that variable pathogenicity from individual to individual is a major source of lifespan variability and that the mechanism involves variable activation of the insulin-signaling pathway .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"aging",
"physiological",
"processes",
"developmental",
"biology",
"organism",
"development",
"evolutionary",
"biology",
"physiology",
"biology",
"anatomy",
"and",
"physiology"
] |
2011
|
Variable Pathogenicity Determines Individual Lifespan in Caenorhabditis elegans
|
Dominance hierarchies are group-level properties that emerge from the aggression of individuals . Although individuals can gain critical benefits from their position in a hierarchy , we do not understand how real-world hierarchies form . Nor do we understand what signals and decision-rules individuals use to construct and maintain hierarchies in the absence of simple cues such as size or spatial location . A study of conflict in two groups of captive monk parakeets ( Myiopsitta monachus ) found that a transition to large-scale order in aggression occurred in newly-formed groups after one week , with individuals thereafter preferring to direct aggression more frequently against those nearby in rank . We consider two cognitive mechanisms underlying the emergence of this order: inference based on overall levels of aggression , or on subsets of the aggression network . Both mechanisms were predictive of individual decisions to aggress , but observed patterns were better explained by rank inference through subsets of the aggression network . Based on these results , we present a new theory , of a feedback loop between knowledge of rank and consequent behavior . This loop explains the transition to strategic aggression and the formation and persistence of dominance hierarchies in groups capable of both social memory and inference .
Individuals from social species must interact with each other to reproduce , find food , and survive . Higher-level social structures such as hierarchies emerge when interacting individuals need to manage trade-offs in the costs and benefits of social associations [1 , 2] . One of the most important is the dominance hierarchy , where group-wide “global” rankings are derived from local aggressive interactions , and form emergent social properties [3–5] . Aggression has obvious immediate costs , including energy expended and the possibility of injury . Benefits , conversely , can be both immediate and delayed . Individuals may fight to gain immediate access to contested resources , or they may aggress in order to gain rank , which then provides these individuals with delayed rank-dependent benefits . Aggression that results in higher dominance rank often increases an individual’s access to foraging resources and reproductive opportunities ( e . g . , Ref . [6–8] ) . Groups across a broad range of taxa are structured by dominance rank [9] despite large variation in cognitive skills . Dominance hierarchies are found in primates [10 , 11] , social carnivores [12 , 13] , ungulates [14 , 15] , birds [16–19] , fish [20] , and even crustaceans [21 , 22] and insects [23] . These group-level social structures form and stabilize on the basis of perceptions and actions necessarily made at the individual level [24] . Dominance rank is generally achieved through a series of aggressive events , and hierarchy formation takes place without top-down control in a manner that is largely independent of the intrinsic properties of the individuals involved [25–27] . Previous experimental and theoretical work on dominance hierarchies shows that interaction outcomes can shape how individuals behave in subsequent interactions [28] , that individuals may use a jigsaw approach to determine how to interact with a few individuals based on observations of event outcomes [25 , 29] , and that observed win/loss outcomes can be strong determinants of an individual’s choice of targets of future aggression [30–32] . In some systems , “badges of status” , or conventional signals , can enable individuals to estimate the rank of others [33 , 34] . However , these visual signals of rank can differ in their prevalence within species under different conditions ( e . g . [35] and [36] ) , and may only explain part of the dynamics of dominance hierarchy formation ( e . g . [37] ) . Aggression preferences of individuals in some groups can also be driven by spatial patterns , especially in cases where closely-ranked individuals are more likely to be in spatial proximity [38–41] . However , little is known about rank formation in groups where these simpler behavioral rules or cues to rank are absent . Recent evidence suggests that individuals in many species have the cognitive ability to use their observations of social interactions of others to inform their own behaviors , including several primate species [10 , 11 , 42–44] , ravens [45] , hyenas [46] , and fish [47–50] . However , while this evidence shows that individuals observe each other and react to these observations , we do not currently understand how individuals integrate information on the outcomes of their own interactions , with observations of others’ interactions , in order to determine rank . We are thus faced with two distinct questions . How does rank become relevant to individual decisions to aggress ? And , what information or mechanism might individuals use in order to learn emergent social properties like rank ? This paper presents three findings , based on detailed , highly-resolved observations of aggression in a social avian species , that answer these questions by reference to the interaction between local decision-making and global system properties . Our first finding is a strong signal of the influence of rank on decision-making . This is seen in how aggression is allocated: target choice became structured over and above what is necessary to reproduce the rank order alone . This structuring happened in a manner that could not be accounted for by individual characteristics , or by the spatial position of individuals . In both groups , it occurred around a week after initial group formation . After this transition , individuals preferentially directed aggression more frequently towards those nearby in rank and avoided interactions with those far below them in rank . Our second finding is that in this structured society , both levels of aggression and subsets of the full network ( network motifs in the form of aggression chains ) provided cognitively-accessible signals of rank . These pathways are the likely mechanisms through which rank is inferred . Rank is a global property , but in these structured systems can be learned by judicious observation of local interactions . Our third finding is that the motif pathway not only provided a signal of relative rank , but was strongly predictive of actual behavior . Individuals were far less likely to direct aggression against the terminal individual in an observable aggression chain . Taken together , these results help explain the emergence of rank as an interaction between two processes: inference of rank from cognitively-accessible social signals , and decision-making that correlates with these signals . They indicate a critical role played by a knowledge-behavior feedback loop—between inference of group-level properties , and consequent decision-making . Such feedbacks may be a critical pathway for how evolved systems reduce uncertainty by tying together multiple timescales [51]; our findings here have parallels in discoveries of how signal use in primates tracks coarse-grained features of a social network [24 , 52] . These results provide new insight into the problem of choosing targets and establishing a dominance hierarchy in species that lack simple perceptual cues , such as size or spatial location , for an individual’s rank . In these more complex societies , rank order is necessarily a cognitive construct that summarizes the many dyadic-level interactions into an emergent group-level property . Our results derive from studies of two independent groups of monk parakeets ( Myiopsitta monachus ) , a small neotropical parrot native to temperate South America and notable for its highly social colonial and communal nesting behavior [53] as well as its widespread success as an invasive species [54–57] . Monk parakeets exhibit several characteristics of complex sociality [58–60] , and to our knowledge , is the first parrot species in which detailed and quantitative dominance hierarchy analysis has been conducted [58] . Studies of dominance hierarchies in parrot species present an intriguing comparison to those conducted on primates and humans . Parrots share many characteristics with primates , such as large relative brain size and advanced cognition [61 , 62] , extended developmental period [61 , 63] , long lifespans [64–66] , and individualized relationships within complex social groups [58 , 67] . Additional characteristics , such as vocal learning and high fission-fusion dynamics , are uncommon in most primates [68] , but are shared by parrots and humans [69] . Understanding how parrots form and maintain dominance relationships in complex social groups thus has the potential to further our more general understanding of rank in socially and cognitively complex species .
All animal activities conducted during this study were approved by the New Mexico State University Institutional Animal Care and Use Committee ( protocol number 2006-027 ) . Our study is based on observation of directed aggression in groups of monk parakeets housed in captivity at the U . S . Department of Agriculture National Wildlife Research Center in Gainesville , Florida . We formed two independent groups ( N = 21 and 19 ) and observed aggressive events during novel group formation ( additional details in Refs . [58–60] ) . Prior to our study , parakeets were housed in smaller cages ( median group size = 2 ) ; while some were in visual contact , direct physical contact between individuals in different cages was not possible . Random group assignment resulted in the formation of social groups largely comprised of novel dyadic associations: only 3% ( Group One ) and 6% ( Group Two ) of dyads were composed of birds housed together during the 8 months preceding the study [59] . To facilitate individual identification , we marked each bird with a unique facial pattern using colored nontoxic permanent markers ( Sharpie , Inc . ) . Each captive group was released sequentially into a 2025 m2 semi-natural outdoor flight pen and observed over the course of 24 days by 1–4 observers . We used all occurrence sampling [70] to record data on directed agonistic behaviors . As in Ref . [58 , 60] , we restricted our analysis to dyadic aggression where events had clear outcomes . We focused on intentional aggressive behavior , which we define here as events where one individual ( the actor ) approached another ( the target ) and the overt aggressive actions of the actor caused the target to be physically displaced and supplanted from its perch by the actor . This resulted in a win for the actor and a loss for the target of the aggression . Because some actors aggressed against targets in a string of frequent sub-attacks ( e . g . 6 attacks in 10 sec ) , we required a cessation of aggression from the actor toward the target of at least 60 sec in order to define the aggression as a clear win; thus an actor could only achieve a full win once against a target per minute . We do not include more subtle forms of aggressive signaling , nor aggression that occurred during scramble competition , because specificity of the direction of aggression was less clear and overt aggression was often reactive rather than intentional in these cases . We hereafter refer to these intentional , dyadic , aggressive wins as ‘aggressive wins’ or ‘aggression’ . We divided the 24-day study period into four 6-day study quarters to facilitate comparisons across the two replicate social groups . Keeping track of the total number of aggressive events between any two individuals allows us to define the directed aggression network; dij , the number of observed aggressive events directed by i against j . We then use Eigenvector Centrality ( EC , Ref . [71] ) on the directed aggression network as our primary means of determining rank . EC assigns a centrality score , vi , to each individual i , using both the direct and indirect links in aggression networks in a recursive fashion . High centrality then equates to low power . Intuitively , an individual has low power if it is the recipient of many aggressive events from those who themselves have low power . Once we have observed dij , we can then define the normalized aggression , tij , as t i j = d i j + ϵ ∑ k = 1 N ( d i k + ϵ ) , ( 1 ) where ϵ here is a small regularizing term . Then , the centrality scores for the individuals in the system , vi , are those that satisfy v i = ∑ j = 1 N t i j v j ( 2 ) EC is one of the main algorithms for determining consensus beliefs within a network [72] , and dominance ranks based on EC power scores are strongly correlated with the main alternative ranking methods such as I&SI [73] . A benefit of EC over pure ranking methods is that EC allows for the direct quantification of power , rather than just a linear order; this means it can distinguish cases where two individuals are “nearly equivalent” in rank from those where differences in dominance are reliable and large . Both the bootstrap error estimates we report , and the null models we use , preserve these distinctions: they reproduce the underlying power scores , rather than just the rank difference . EC is closely related to David’s Score ( DS; Ref . [74] ) . In contrast to EC , which uses all the interactions in the aggression network to define the status of a particular individual , DS only includes interactions up to two steps away ( my aggressors’ aggressors; my targets’ targets ) . Both DS and EC have found widespread use in the characterization of rank in animal groups [24 , 52 , 60 , 73 , 75–78] . EC and DS are both “depth” methods [75] . They quantify rank by reference to network properties , and weight the interactions between two individuals i and j in ways that depend on interactions each individual has had with others . Conversely , measures like Weighted Simple Consensus ( WSC ) are “breadth” methods [75] , in which rank is estimated based only on the interactions directed towards individual i . For example , WSC estimates rank as the product of the total number of an individual’s aggressors and the total amount of incoming aggression [75] . Ref . [75] found EC comparable to WSC , both of which performed well . In our data , EC correlates strongly with all three measures ( I&SI , DS , WSC ) ; see S1 Table . We are most interested in how relative rank influences how and where individuals direct their aggression . To measure this , we use average rank aggression , R ( Δ ) . R ( Δ ) is the amount of individual-level aggression , per unit time , directed at those Δ ranks away . It is defined as: R ( Δ ) = 1 N Δ T obs ∑ i = 1 N d i Δ ( 3 ) where diΔ is the amount of aggression directed by i at the individual whose rank is Δ steps away; NΔ , the total number of individuals who have a potential target Δ ranks away; and Tobs the total observation time . A rank lower than i , i . e . , “down” the hierarchy , is indicated by Δ greater than zero; a rank higher than i is indicated by Δ less than zero . Average rank aggression is our primary signal of individual decision-making . We are interested in determining whether the observed aggression indicates the influence of rank on decision-making . In order to do this , we construct null models for the range of behaviors we expect to see if individuals interact in a unstructured fashion . We explore a related measure , average preferential rank aggression , in S1 Text . A critical step in our analyses is determining when and how patterns of aggression differ from what might be expected from random noise in otherwise unstructured behavior . To do this , we construct a hierarchy-constrained null model . This null model produces aggression networks that reproduce the observed power scores , without imposing particular rules about who directs aggression against whom . Such a model is possible because in a group of n individuals there are n ( n−1 ) free parameters in an aggression network , but only n−1 numbers are required to specify that network’s EC power scores . EC ( indeed , any ranking or scoring system ) thus amounts to a lossy compression of the original data [79] , summarizing the behavioral patterns relevant to the establishment of a dominance hierarchy . Conversely , for any given dominance hierarchy there are many possible behavioral patterns . In particular , because there are many possible d and t matrices compatible with a particular power distribution v , we can define a null model as random draws from the set of matrices that have , on average , the same v . Our null model for aggression is defined as a random sample from this larger set; we also preserve the total aggression of each individual ( additional details in S2 Text ) . Any particular sample from this null will preserve ( on average ) the EC power scores , but will be otherwise unstructured and contain no correlations that are unnecessary to preserve those scores . We can measure relevant properties , such as R ( Δ ) , on these null networks . Deviations in the real data from these nulls indicate individuals are systematically directing aggression in ways that differ from what is otherwise expected for an aggression network with that dominance structure . We investigated whether parakeets could use size characteristics or spatial positioning , rather than social observations , to infer rank . We measured several morphometric characteristics which reflect body size: wing chord , culmen depth , culmen width , and mass . We used these data to determine whether rank could be predicted by an individual’s size . We also collected data on spatial patterns . As in [58 , 60] , we determined the identity of each individual’s nearest neighbor ( within 10m2 quadrats ) during scan samples to determine whether rank affected spatial association patterns . If birds nearby in rank tend to spend more time physically near each other , spatial proximity could serve as a signal of rank . In particular , we would expect a negative correlation between ( absolute ) relative rank difference between i and j , and the number of times i was observed to be j’s nearest neighbor in space . Estimating rank via social observations and signals can be computationally challenging . As noted by Ref . [75] , breadth measures such as WSC are of particular interest because they correlate with the more sophisticated depth measures , but are more likely to be cognitively-accessible to individuals within the system . For this reason , we also measure WSC on the directed aggression network . The WSC score of an individual is the product of the total amount of aggression ( number of events ) received , multiplied by the number of distinct individuals who directed that aggression ( additional details in S4 Text ) . We order individuals by these dominance scores to determine individual rank . As in the case of EC , a higher WSC equates to lower rank because these individuals receive more aggression from more individuals . Individuals may use breadth-based signals such as WSC , but they may also use measures sensitive to other features of the network . In particular , even though EC rank is a property of the aggression network as a whole , small portions of that network may contain signals of relative rank , and these smaller network subsets may be more easily perceived . To study information of this second kind , we focus on a particular kind of motif—the aggression chain—where we can trace a line of aggression from individual i , to individual j , to individual k , and so forth . We measure the signals contained in such chains using average weighted rank difference , W ( n ) . W ( n ) quantifies the extent to which an aggression chain provides information about relative rank difference for chain length n . W ( 2 ) is defined as W ( 2 ) = ∑ i , j , k ; ∅ d i j d j k Δ ( i , k ) ∑ i , j , k ; ∅ d i j d j k , ( 4 ) where Δ ( i , k ) is the rank difference between i and k , and ∅ indicates that in any instance , the identities of individuals i , j , and k do not overlap . W ( 2 ) is then the average rank difference between any two individuals , weighted by the product of aggression seen along all chains connecting them . W ( 2 ) takes into account not only the existence of the chain , but also its strength . In general , weighted network ties are generally more informative about the social relationships among individuals and are more robust to sampling differences [59 , 80–82] than relying solely on presence-absence information . W ( n ) for n larger than two is defined similarly ( see S3 Text ) ; the number of possible motifs we need to examine to compute W ( n ) grows exponentially with depth , and we stop our analysis at n equal to six . ( Note that while the total number of motifs grows exponentially with n , W ( n ) quantifies the average amount of information in any particular aggression chain , not the total amount of information in all chains . ) When W ( n ) is significantly different from the null , this indicates the presence of information in aggression chains over and above what is expected from systems with the same power scores , but otherwise unstructured aggression . To determine if individuals actually perceived and used these motifs as a signal of rank , we measure the behavioral signatures of transitive inference . Transitive inference occurs when one individual uses knowledge about its own interactions with a target ( j ) and third-party observations of how j interacts with additional targets ( k , l , m , … ) to infer its own likelihood of winning over one of j’s targets , such as k . For the case of three individuals , i and k would have a transitive relationship with individual j if the amount of aggression directed from i to j ( dij ) and the aggression from j to k ( djk ) was related to the amount of aggression i directed to k ( dik ) . We tested for transitive relationships between individuals in the first and last positions anchoring each aggression chain ( Fig 1 ) , up to a chain length of 6 . We quantified fractional transitivity , T ( n ) , for a chain length of two as: T ( 2 ) = ∑ i , j , k ; ∅ d i j d j k ( d i k - d ¯ i ; j ) / d ¯ i ; j ∑ i , j , k ; ∅ d i j d j k ( 5 ) where d ‾ i ; j is the average aggression i directed towards all individuals other than j , and ∅ indicates that our sums exclude cases where the identities of individuals i , j , and k overlap . The natural extension to n-step chains , T ( n ) , is defined in S3 Text . T ( n ) is positive or negative when an individual i increases or decreases , respectively , its aggression against k given that k is at the end of an n-step chain . Very positive T ( n ) means that individuals prefer to aggress against those at the end of chains while very negative T ( n ) means that individuals preferentially avoid such aggression . When T ( n ) is significantly different from null , this indicates the use of information in aggression chains over and above what is expected from systems with the same power scores , but otherwise unstructured aggression . Our initial analysis indicated that differences in aggression patterns in Group One and Group Two were strongly driven by a single individual , NBB , in Group Two . This individual was persistent in her attempts to affiliate with others in Group Two , but was not able to form a strong affiliative relationship within the group [58] . Aggression directed at NBB appeared to be mostly reactive aggression in response to NBB’s persistent and apparently unwanted attempts to affiliate rather than intentional target selection choices by actors choosing to attack NBB . Because our focus with this work was on intentional target selection and strategic aggression , and because the actions of NBB were anomalous compared to all the other individuals in the group , we excluded NBB from the main analyses . However , we present the results for Group Two including NBB in the Supplementary Information ( S1 Fig and S2 Fig ) to show how unusual decision-making by this individual affects the overall patterns created by the other 18 individuals .
We find no evidence that rank could be reliably determined based on simple underlying cues such as size or spatial proximity . None of the morphometric body size measures provided any rank signal in either of the two study groups . Rank was not significantly associated with the physical size of individuals , including weight , wing length , and beak properties ( ∣r2∣ < 0 . 18 , p > 0 . 05 , S3 Fig ) . Nor did we find that spatial proximity provided a signal of rank ( S4 Fig ) . Neighbor identity is not a strong signal of rank in either Group One ( r2 = −0 . 12 ) or Group Two ( r2 = −0 . 02 ) . Even in the final three-quarters , where behavior is most regular , in Group One , proximity provides only a weakly anticorrelated signal ( r2 = −0 . 14 ) ; in Group Two , it provides no signal at all ( r2 = 0 . 04; consistent with null ) . Average rank aggression in both Groups One and Two was consistent with the null model during the first quarter of the study period ( Fig 2a and 2e ) . However , behaviors quickly diverged from null expectations as individuals began to structure their behavior in ways strongly correlated with rank . Aggression patterns in the final three-quarters of the study period all diverged strongly from null expectations ( Fig 2b–2d and 2f–2h ) , with aggression strongest towards individuals nearby in rank . We refer to this as rank-focused aggression . Combining the final three quarters of the data increases this signal ( Fig 3a and 3b ) . We can quantify the emergence of this phenomenon by reference to the ratio of null to observed aggression for nearby ranks ( one to five steps below ) . While neither group showed above-null aggression to nearby ranks in quarter one ( p > 0 . 05 ) , both showed significant deviations and large effect sizes for later quarters . Group One had aggression at 49% above-null ( p < 0 . 05 ) in the second quarter; and a factor of 2 . 7 and 2 . 2 times higher in quarters three and four ( p < 0 . 001 ) . Group Two had aggression at 60% above null ( p < 0 . 05 ) in the second quarter , and 64% and 56% in quarters three and four ( p < 0 . 05 ) . Aggregating over the final three quarters gives an overall signal of rank-focused aggression of a factor of 2 . 1 times above null ( Group One , p < 0 . 001 ) and 51% above null ( Group Two; p < 0 . 01 ) . We find equivalent results for a related measure , average preferential rank aggression; see S5 Fig . This is our first main result . The onset of strong deviations from the nulls points to the emergence of structured behavior at the individual level . The structure of this behavior is dictated , at least in part , by relative rank . Determination of rank via depth-based measures is computationally intensive . In our data , depth-based EC correlated strongly with breadth-based WSC ( r2 ≈ 0 . 73 in both groups ) and thus knowledge of WSC can provide at least partial knowledge of EC . Because of WSC’s reliance on levels of aggression alone , rather than network structure , we refer to this rank-inference mechanism as the “magnitude pathway” . We look for signals of the use of this pathway by considering evidence that aggression is structured as a function of relative WSC rank . We find evidence for reduced aggression at individuals widely separated in WSC rank ( large positive Δ ) compared to the null . This indicates that , in addition to providing knowledge of rank , WSC-derived rank is also predictive of some features of individual aggression . However , we do not see strong evidence for increased aggression to those nearby in WSC rank—none in Group One , and only weak evidence in Group Two ( S6 Fig ) . The fact that WSC was predictive of aggression indicates that the magnitude pathway may play an important a role in structuring aggression . However , the absence of a signal of rank-focused aggression implies that there were patterns of aggression invisible to the breadth-based WSC measure . If individuals used WSC signals to help direct their aggression to those nearby in rank , they must have been supplementing them with other sources of information . The absence of rank-focused aggression in the WSC case is an example of how rank defined by reference to WSC does not capture all of the structure relevant to individual aggression . This provides an implicit justification for the decision , made earlier , to use EC as the primary measure of rank in the system . We also evaluated evidence for the use of depth-based measures of rank , where individuals are evaluated based not only on the aggression they receive , but on the characteristics of their aggressors . In particular , weighted rank difference , W ( n ) , allows us to investigate whether social information was encoded within smaller subsets of the total aggression network in chains of aggression ( Fig 4 ) . We refer to this rank-inference method as the “motif pathway” . In the null ( dashed line in Fig 4 ) , we find that chain length encodes little or no information about relative rank . Individuals tend to be lower-ranked than their aggressors , but seeing an individual at the end of a long chain provides little or no additional information about its rank . By contrast , the observed data ( solid line in Fig 4 ) encoded significant information in aggression chains . In both Groups One and Two , local motifs contained a substantial amount of global information . An observation of an individual at the end of a longer chain ( length ≥ 3 ) provided more information about relative rank than an observation of an individual at the end of a short chain ( length 1 or 2 ) . This encoding potentially allowed individuals to distinguish between individuals nearby ( Δ ∼ 5 ) and distant ( Δ > 10 ) in rank—a discrimination impossible in the nulls . This is our second main result: network motifs , in addition to breadth-based magnitude measures , can provide signals of relative rank . Fractional transitivity , T ( n ) , allows us to investigate whether these aggression chains are predictive of actual behavior . Our analysis found a strong difference between the null model and the observed data ( Fig 5 ) . In the null ( dashed line ) , chains predict increased aggression: if i aggresses against j , and j against k , this leads i , on average , to direct increased aggression against k . This is independent of chain length—both long and short chains predict similar levels of increased aggression . In contrast , behavior in the observed data ( Fig 5 , solid line ) showed the opposite effect . While short chains predicted a ( small amount ) of increased aggression ( T ( 2 ) greater than zero ) , long chains were associated with reduced aggression ( T ( n ) less than zero for n greater than two ) . This remarkable inversion is what is expected when individuals use the information content of motifs to predict relative rank , and then both ( 1 ) preferentially avoid conflict with much lower-ranked individuals and ( 2 ) focus aggression on rank neighbors . This is our third main result: network motifs predict behavior . Having demonstrated the existence and predictive power of signals , we can investigate the time-frame over which the signals themselves emerge by looking at W ( n ) quarter-by-quarter . In Group One we can already find evidence for the existence of these signals , over and above null expectation , at the one-step and two-step level , in the first quarter . The signals were present , but R ( Δ ) results suggest that they are not yet used by participants . By contrast , the signal is almost entirely absent in the first quarter for Group Two . After the first week ( i . e . , for the later quarters ) the signal became significantly stronger , covered a wider range of ranks , and we saw signals at the three-step , and often at the four-step , level in both groups ( S7 Fig and S8 Fig ) . We can also examine T ( n ) quarter-by-quarter to look for the dynamical emergence of behavioral preference . The first quarter showed some evidence for the use of motif information to structure behavior , with shifts in preferences of on average 32 percentage points relative to the null in both groups . Effects became larger at later times , with behavioral preferences shifting by on average 82 percentage points relative to the null in the final three quarters . Preference inversion appears only in these later quarters . The emergence of this behavioral pattern parallels the emergence , over time , of the large scale order seen in Fig 2 , and of cognitively-accessible signals seen in Fig 4 ( S9 Fig and S10 Fig ) .
How individuals come to know their social worlds , and how that knowledge feeds back to influence social properties , is a crucial part of group dynamics . This paper has tracked the emergence of strategically directed aggression , the signals that could enable it , and how these signals predict decisions to aggress . Previous work on transitive inference by non-human animals has often focused on experimental manipulation of trained subjects [83 , 85] . These experiments are generally removed from social situations . They can provide critical evidence that individuals have the necessary cognitive skills , but usually cannot show a direct link between those abilities and an understanding of the social landscape of an actual group and its effect in real-world situations . Our work provides new evidence for the importance of transitive inference in the real-world problem of directing aggression . Our findings on the information contained in chains of aggression and the strategic use of this information allow us to construct a mechanistic account of both what signals of rank in observed behavior might be available to individuals ( the knowledge pathway ) and how these signals influence decision-making ( the behavior pathway ) . It allows us to explain the dynamical transition as the onset of a complex interaction between knowledge of rank and consequent behavior . In the closing of this knowledge-behavior feedback loop are the seeds of complex society .
|
An individual’s success depends critically on socially-constructed properties such as rank . A detailed study of two independent captive parakeet groups reveals how these properties come into being . We show that individuals can use localized patterns in the aggression network to learn the relative ranks of individuals , and that these signals of rank strongly correlate with individual decisions to aggress . Over time , feedback between knowledge and behavior leads to the emergence of strategic aggression: individuals focus their aggression on those nearby in rank .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Social Feedback and the Emergence of Rank in Animal Society
|
Phagocytosis is required for proliferation and pathogenesis of Entamoeba histolytica and erythrophagocytosis is considered to be a marker of invasive amoebiasis . Ca2+ has been found to play a central role in the process of phagocytosis . However , the molecular mechanisms and the signalling mediated by Ca2+ still remain largely unknown . Here we show that Calmodulin-like calcium binding protein EhCaBP3 of E . histolytica is directly involved in disease pathomechanism by its capacity to participate in cytoskeleton dynamics and scission machinery during erythrophagocytosis . Using imaging techniques EhCaBP3 was found in phagocytic cups and newly formed phagosomes along with actin and myosin IB . In vitro studies confirmed that EhCaBP3 directly binds actin , and affected both its polymerization and bundling activity . Moreover , it also binds myosin 1B in the presence of Ca2+ . In cells where EhCaBP3 expression was down regulated by antisense RNA , the level of RBC uptake was reduced , myosin IB was found to be absent at the site of pseudopod cup closure and the time taken for phagocytosis increased , suggesting that EhCaBP3 along with myosin 1B mediate the closure of phagocytic cups . Experiments with EhCaBP3 mutant defective in Ca2+ -binding showed that Ca2+ binding is required for phagosome formation . Liposome binding assay revealed that EhCaBP3 recruitment and enrichment to membrane is independent of any cellular protein as it binds directly to phosphatidylserine . Taken together , our results suggest a novel pathway mediating phagocytosis in E . histolytica , and an unusual mechanism of modulation of cytoskeleton dynamics by two calcium binding proteins , EhCaBP1 and EhCaBP3 with mostly non-overlapping functions .
A variety of cell types , such as macrophages and neutrophils and many unicellular eukaryotes have the ability to engulf particles of size greater than 0 . 5 µm through a process called phagocytosis . In the former this process has evolved as one of the critical elements of host defence , while in the latter it serves as a mode of nutrition . Entamoeba histolytica , a parasite that colonizes the human gut and causes dysentery , is endemic in many developing countries and causes a high level of morbidity and mortality [1] , [2] . Phagocytosis is considered to be important in E . histolytica pathogenesis , as a phagocytosis-deficient mutant showed reduced virulence [3] . In another study , the virulence potential of E . histolytica isolates could be directly correlated with their ability to phagocytose red blood cells ( RBCs ) [4] . Phagocytosis is initiated when a particle binds to a cell surface receptor , leading to local reorganization of actin cytoskeleton and providing the necessary force needed for the formation of phagocytic cups and phagosomes [5]–[7] . The rim of filamentous ( F ) actin ( periphagosomal F-actin ) , surrounds early phagosomes and then progressively depolymerizes as the phagosome matures [5] , [8] , [9] . It is believed that this disassembly of the F-actin rim is necessary for phagosome maturation , as it may act as a barrier for phagosome-vesicle fusion [8]–[11] . Therefore , spatial and temporal regulation of actin dynamics is the key to controlling phagocytosis . This is achieved through a number of actin binding proteins ( ABPs ) [12] . ABPs are involved in regulating actin cytoskeleton dynamics at multiple levels; for example , promotion of nucleation and polymerization of F-actin by Arp2/3 complex and profilin [13] , [14] and depolymerization of F-actin by ADF/cofilin and gelsolin [15] . Ca2+ is a prominent regulator that can exert multiple effects on structure and dynamics of actin cytoskeleton . Ca2+ transients during phagocytosis initiate these processes in many systems [16]–[18] including E . histolytica [19] . Cytoskeletal remodelling by Ca2+ may occur through Ca2+ binding proteins ( CaBPs ) that can sense alteration in Ca2+ concentration and undergo conformational change [20]–[22] . In Dictyostelium discoideum , a 34 kDa protein is involved in actin bundling in a calcium-regulated manner [23] and a 40 KDa protein restricts the length of actin filaments in the presence of Ca2+ [24] , [25] . Ca2+ is also involved in other processes related to cytoskeleton remodeling , for example Ca2+-Calmodulin regulates actin polymerization via Fesselin [26] and a low molecular weight protein CBP1 in D . discoideum has been shown to regulate the reorganization of actin cytoskeleton during cell aggregation [27] . The role of actin in endocytic/phagocytic processes has been studied in different systems and cell types using a number of different inhibitors or pharmacological compounds [28] . Some of the results of these studies suggest that clathrin-coated vesicle formation may not require actin dynamics [29] . However , its role in post vesicle processing cannot be ruled out . In a different approach , over expression of Y282F/Y298F-FcgR , a signaling- dead mutant receptor in COS-7 cells is unable to signal to the actin cytoskeleton , but specifically binds IgG ligand , had no effect on phagocytosis [29]–[31] . In some of these cases it is thought that phagocytosis takes place via passive zipper mechanism in which ligand-receptor binding remains specific and strong but reversible due to the absence of actin polymerization . Passive engulfment is generally slower and produces much more variable phagocytic cups [32] . The genome analysis of E . histolytica has revealed 27 CaBPs with multiple EF-hand calcium binding domains [33] . Of these , EhCaBP1 has been studied in much more detail and it is now clear that EhCaBP1 is a central molecule involved in initiation of erythrophagocytosis along with EhC2PK , a C2 domain containing protein kinase [34] . EhC2PK accumulates at the site of RBC attachment in a Ca2+- dependent step and recruits EhCaBP1 , which in turn brings actin filaments resulting in initiation of phagocytosis [34] . Ca2+ has been shown to participate in the initiation process at two levels . Firstly , it is necessary for membrane localization of EhC2PK and secondly , Ca2+-EhCaBP1 is required for phagocytic cups to progress towards phagosomes [34] . Therefore , Ca2+ has an important role in regulating erythrophagocytosis in E . histolytica . A calmodulin-like calcium binding protein EhCaBP3 has been identified and partially characterized in E . histolytica [35] . Three dimensional structure , using nuclear magnetic resonance ( NMR ) spectroscopy , suggests that EhCaBP3 has a well folded N-terminal domain and an unstructured C-terminal counterpart , somewhat similar to calmodulin and EhCaBP1 . Interestingly , EhCaBP3 was found in all three major cellular compartments; nucleus , cytoplasm and membrane [35] . In this report we show that EhCaBP3 is involved in the process of phagocytosis at both initiation and phagosome formation stages . In vitro experiments suggest that EhCaBP3 binds actin , and affects its polymerization and bundling . Therefore it is likely that EhCaBP3 regulates phagocytosis by participating in actin dynamics . Our studies also show that EhCaBP3 and EhCaBP1 have different roles though both are recruited early during phagocytosis . We conclude that E . histolytica displays unique mechanism of regulating phagocytosis using a number of novel calcium binding proteins not observed in any other system .
Ca2+ is required for phagocytosis in E . histolytica as chelation of cytoplasmic Ca2+ blocks phagocytosis [19] . Therefore , it is expected that CaBPs may be participating in phagocytosis as Ca2+ sensors . We have earlier shown the involvement of one of the calcium sensing CaBPs of E . histolytica , EhCaBP1 in erythrophagocytosis [19] , [21] . EhCaBP3 was identified as a calmodulin-like calcium binding protein of E . histolytica as its structure showed similarity with calmodulin [35] . Since multiple CaBPs are likely to be involved in different steps of phagocytosis , the subcellular localization of EhCaBP3 was checked during RBC uptake by immunostaining with specific anti-EhCaBP3 antibody . The results are shown in Figure 1 . Fluorescence signals clearly showed that EhCaBP3 was present in phagocytic cups , as has been shown for EhCaBP1 [19] . Actin was also observed to line the cups and the complete superimposition of both EhCaBP3 and actin suggested that both proteins are colocalized at the phagocytic cups ( Figure 1A ) . EhCaBP3 was also found on early phagosomes along with actin . Superimposition of both molecules suggested that both EhCaBP3 and actin are also co-localized at the newly formed phagosomes . Our earlier studies had shown that EhCaBP1 was found only at cups and not on phagosomes . Therefore relative localization of EhCaBP1 and EhCaBP3 were studied in actively phagocytosing cells in order to see functional differences between the two CaBPs , using antibodies against EhCaBP1 ( red ) and EhCaBP3 ( green ) . Since we wanted to see both phagocytic cups and phagosomes , amoebic cells were incubated with RBCs for different times . As expected EhCaBP1 was observed only in the phagocytic cups whereas EhCaBP3 was found in both phagocytic cups as well as in early phagosomes ( Figure 1B ) . The results suggest that EhCaBP3 is likely to be involved in erythrophagocytosis and it may be functionally different from EhCaBP1 . In order to check if EhCaBP3 may also participate in phagocytosis of other particles , EhCaBP3 was immunostained during phagocytosis of CHO cells and the results are shown in Figure S1 . Fluorescent signals were found in the cups that are in the process of phagocytosing CHO cells . However , it was not clear whether any significant signal was present around the phagosomes , as observed with RBCs ( compare Figure S1 and Figure 1 ) . Phagosome with low intensity staining could be discerned in some cases and these are marked with asterisk . Many CHO cells formed tunnel like structure during phagocytosis and EhCaBP3 was localized at the tip ( marked by an arrow ) . These tunnel-like structures have also been observed before [36] . The results suggest that EhCaBP3 may also be involved in phagocytosis of CHO cells . However , the extent of participation and the exact roles may be different from that of RBCs . We have further characterized the role of EhCaBP3 in phagocytosis using RBC uptake as our model . Dynamics of EhCaBP3 recruitment and release during erythrophagocytosis was studied by expressing EhCaBP3 in E . histolytica cells as a GFP fusion protein on a plasmid vector maintained in the presence of G418 ( Figure 1C ) . While there was no change in the expression of endogenous EhCaBP3 ( 17 kDa ) , the expression of GFP-EhCaBP3 ( 43 kDa ) increased with increasing concentration of G418 as seen by western blotting using anti-GFP antibodies which do not stain endogenous EhCaBP3 ( Figure 1D ) . There was no change in the levels of endogenous EhCaBP3 visualized by anti-EhCaBP3 antibody under the same conditions . Since it is likely that GFP tagged proteins may not behave like native proteins we checked the localization of GFP-EhCaBP3 during erythrophagocytosis using anti-GFP antibodies . Confocal microscopy revealed that GFP-tagged EhCaBP3 ( but not GFP alone ) enriched at phagocytic cups and early phagosomes along with actin ( Figure 1E , 1F , 1G ) , suggesting that GFP-EhCaBP3 behaves in a similar way as endogenous EhCaBP3 . The results reported so far show that EhCaBP3 is required both at the initiation and end stages of phagocytosis . It appears to redistribute during the whole process . In order to observe this dynamic behaviour of EhCaBP3 , time-lapse fluorescence microscopy was used with cells expressing GFP-tagged molecules in the presence of RBCs . The results clearly showed that EhCaBP3 first accumulated rapidly at the site of RBC attachment before moving towards the tip of the cups ( Figure 2 ) . EhCaBP3 was present at the time of scission and remained even after complete phagosomes were formed and detached from the membrane . The whole process took about 3 min after addition of RBCs ( supplementary movie S1 ) . The results shown earlier clearly indicate colocalization of EhCaBP3 with F-actin in the context of phagocytosis . This may be brought about by binding of EhCaBP3 to F-actin directly or indirectly through a third molecule . In order to check these possibilities a direct binding assay of EhCaBP3 to F-actin was carried out by co-sedimentation . Polymerized actin was incubated with recombinant purified EhCaBP3 or other indicated proteins , and the complex was centrifuged and analysed by SDS-PAGE . Actin alone was found in the pellet fraction suggesting that the preparation contained mainly polymerized or F-actin . In the absence of actin , the pellet did not contain EhCaBP3 ( Figure 3A , lane10 ) . However , when actin was present EhCaBP3 was found in the pellet fraction ( Figure 3A , lane 8 ) . EhCaBP1 was also present along with polymerized actin in the pellet , as expected , being an actin-binding protein ( Figure 3A , lane 4 ) . In contrast , EhCaBP2 , a close homolog of EhCaBP1 with a different function did not co-sediment with F-actin [21] ( Figure 3A , lane 6 ) . When actin was incubated with both EhCaBP1 and EhCaBP3 , interestingly a complex containing both CaBPs and actin was detected in the pellet ( Figure 3A , lane12 ) . This could be due to a ternary complex ( Actin , EhCaBP1 and EhCaBP3 ) or two separate binary complexes ( Actin and EhCaBP3; Actin and EhCaBP1 ) , which cannot be distinguished at present . Our results suggest that EhCaBP3 can bind F-actin directly . To test the binding of EhCaBP3 to G-actin , a solid-phase assay was performed in the presence or the absence of Ca2+ . It was observed that EhCaBP3 bound G-actin in the presence of Ca2+ . However , the binding was inhibited by 75% when EGTA was added ( Figure 3B ) . This data suggests that EhCaBP3-G-actin interaction requires Ca2+ . We then checked if binding of EhCaBP3 affects properties of actin , as EhCaBP1 was shown to alter the bundling of actin but not its polymerization [21] . EhCaBP1 was used as a negative control as it does not have any effect on actin polymerization [21] . First we tested if EhCaBP3 has an effect on actin polymerization by using pyrene-labelled G-actin . The rate of actin polymerization increased on adding increasing amount of EhCaBP3 reaching a saturation at about 10 µM . At this concentration both the rate as well as the value at saturation was higher by 50% compared to the control . No change in the rate of polymerization was observed in the presence of EhCaBP1 as expected ( Figure 3C ) . To test whether EhCaBP3 influences bundling property of actin , the assay was performed in the presence and the absence of Ca2+ . Majority ( 91% ) of the actin was found in the supernatant fraction when actin alone , or in the presence of BSA were incubated without EhCaBP3 , suggesting that there was no significant amount of actin in the form of bundles ( Figure 3D ) . However , incubation of actin with EhCaBP3 led to bundling of actin as the majority of actin was in the pellet fraction . The result with EhCaBP3 was similar to that with known actin bundling agents , such as EhCaBP1 and alpha actinin [21] . In both cases actin was recovered from the pellet fraction after incubation . Our data also shows that actin bundling property of EhCaBP3 is independent of Ca2+ as actin was seen in the pellet in the presence and the absence of Ca2+ . Our results suggest that EhCaBP3 is an actin remodelling protein and that EhCaBP1 and EhCaBP3 have different functional effects on actin . Myosin IB is thought to be one of the proteins that interact with actin and is involved during some of the cellular processes in E . histolytica , such as phagocytosis [37] . The relationship between EhCaBP3 and myosin IB was investigated in the context of erythrophagocyosis using GFP-EhCaBP3 expressing cells and anti-myosin IB antibodies . E . histolytica cells were incubated with RBCs for different time points so as to capture different stages of phagocytosis . In all stages , that is , from cups to newly formed phagosomes , GFP-EhCaBP3 and myosin IB were found to co-localize ( Figure 4 ) . The presence of both myosin IB and EhCaBP3 at the tip just before phagosome closure ( denoted by star ) suggests that EhCaBP3 along with myosin IB may be involved in the process of phagosome closure . Localization of EhCaBP3 with myosin 1B suggests that these proteins might interact with one another . To confirm this , co-immunoprecipitation was carried out using immobilized anti-EhCaBP3 antibody and total cell lysate of E . histolytica trophozoites . The result is shown in Figure 5 . While anti-EhCaBP3 antibody precipitated myosin 1B along with EhCaBP3 in the presence of Ca2+ ( Figure 5 ) , no myosin 1B was observed when EGTA was added , suggesting that Ca2+ is essential for their interaction . We then investigated the importance of Ca2+ binding in the functioning of EhCaBP3 . It was achieved by generating a mutant of EhCaBP3 which could not bind Ca2+ ( EhCaBP3mEF ) . This was done by D→A and E→A mutagenesis respectively of the first D residue of all EF-hand motifs and last E residue of EF-I and EF-III ( Figure 6A ) . The recombinant mutant protein did not bind Ca2+ as shown by ruthenium red staining ( Figure S2 ) . EhCaBP3mEF was checked for its ability to bind both F and G actin ( Figure 6B and 6C ) . Polymerized actin co-sedimentation assay revealed that both wild type and mutant EhCaBP3 bound F-actin ( Figure 6B ) . Binding to G-actin was carried out using a plate binding assay . EhCaBP3mEF did not bind G-actin unlike the wild type protein ( Figure 6C ) suggesting that binding of EhCaBP3 to G-actin requires involvement of Ca2+ whereas binding to F-actin does not . EhCaBP3mEF was also checked for its ability to get recruited in phagocytic cups and phagosomes . This was done by expressing a GFP-tagged mutant protein in E . histolytica cells ( Figure 6D ) and monitoring GFP as described in “materials and methods” . In order to mark the phagosomes properly , a plasma membrane marker ( EhTMKB1–9 ) was used [34] . Immunofluorescence images revealed that while the mutant protein was observed in the cups ( Figure 6E; upper panel ) , none of the phagosomes contained GFP-EhCaBP3mEF ( Figures 6E and 6F; lower panel ) unlike wild type protein ( Figure 6F; upper panel ) . Further , actin was present in both cups and phagosomes in cells expressing the mutant protein ( Figure 6E ) . However , myosin 1B enrichment and recruitment to those phagocytic cups was hampered where EhCaBP3mEF was present ( Figure 6G ) , suggesting that Ca2+ is essential for recruitment of myosin 1B to phagocytic cups via EhCaBP3 . This is supported by co-immunoprecipitation result as binding of EhCaBP3 to myosin 1B was inhibited in the presence of EGTA . Interestingly cells over expressing the mutant protein displayed only 20% reduction in phagocytic cups , while the reduction in phagosomes was 65% compared with cells over expressing the wild type EhCaBP3 ( Figure 6H ) , suggesting a dominant negative effect of expression of the mutant protein . Since wild type EhCaBP3 continues to be expressed from the endogenous gene it is likely that these molecules help continuation of phagocytosis at a slower rate , even in the presence of EhCaBP3mEF . The results presented so far suggest that EhCaBP3 is associated with phagocytic machinery . In order to show whether it was also required for phagocytosis to occur , the level of EhCaBP3 was reduced by expressing specific antisense RNA . We have been able to down regulate expression of a number of genes using tetracycline-induced whole gene antisense RNA and this system was also employed to study the role of EhCaBP3 in phagocytosis [38] . The vector used and details of different constructs is shown in Figure 7A . On tetracycline addition the level of EhCaBP3 was significantly ( 55% ) reduced in cells carrying antisense construct ( EhCaBP3AS ) as compared to the cells carrying only the vector ( Figure 7B ) . This effect was specific as the amount of EhCaBP1 did not change . When EhCaBP3 gene was over expressed using the cloned gene in the sense orientation ( EhCaBP3S ) the amount of EhCaBP3 increased by 30% in the presence of 10 µg/ml of tetracycline ( Figure 7C ) . E . histolytica cells carrying the sense and antisense constructs were then checked for erythrophagocytosis using a spectrophotometric assay . There was a 70% reduction in cells expressing EhCaBP3 antisense RNA ( that is , in the presence of tetracycline ) as compared with cells carrying only the vector in the presence of tetracycline , and cells carrying EhCaBP3 antisense construct in the absence of tetracycline . Over expression of EhCaBP3 , that is addition of tetracycline to cells carrying a sense construct displayed an increase ( 40% ) in erythrophagocytosis as compared to cells without tetracycline or vector containing cells in the presence of tetracycline ( Figure 7D ) . The results of immunostaining of these cells are shown in Figure S3 . The data showed that in cells expressing anti-sense RNA the cup formation was greatly reduced in the presence of tetracycline , while cup formation took place normally in cells expressing EhCaBP3 in the sense orientation with or without tetracycline . Reduction in phagocytosis on down regulation of EhCaBP3 expression may be due to either a reduction in initiation , progression or scission of phagosome formation . It is also possible that all steps may be affected . In order to identify the site ( s ) affected , cells expressing EhCaBP3 antisense RNA were incubated with RBCs for indicated time and analysed by immunostaining . The results are shown in Figure 7E . In cells expressing EhCaBP3 in sense orientation many phagocytic cups were observed at about 3 min of incubation with RBC . However , the process of cup formation was delayed in antisense expressing cells . A few cups were visible only at about 8 min of incubation ( Figure 7E ) . We also noticed that there was a defect in the closure of the cups to form phagosomes when EhCaBP3AS cells were incubated with RBC for 20 min ( data not shown ) . The statistical analysis of the above data showed that cups appear in EhCaBP3AS cells at about 7 min after addition of RBC and there was a 58% reduction in the number of cups formed ( Figure 7F ) . Interestingly cells over expressing EhCaBP3 consistently showed increased number of cups . It is also clear from Figure 7E that the amount of phalloidin staining in the cups is substantially less in EhCaBP3AS cells as compared to control cells suggesting that F-actin recruitment may also be affected . Quantitation of phalloidin staining in the cups showed 41% reduction in the intensity of F-actin in the phagocytic cups as compared to cells carrying only vector ( Figure S4 ) . This suggests that EhCaBP3 participates both in the initiation as well as closing stages during phagosome formation and that actin dynamics plays a critical role in EhCaBP3 function . We have observed colocalization of myosin IB with EhCaBP3 in phagocytic cups and phagosomes ( Figure 4 ) . Therefore the distribution of myosin IB in EhCaBP3AS was studied in order to further validate interaction of these two proteins during phagocytosis . The results showed the absence of myosin IB at the phagocytic cups even after 20 min of incubation with RBCs in cells expressing anti-sense RNA of EhCaBP3 , suggesting that EhCaBP3 is required for recruitment of myosin IB ( Figure 8A ) . We have also visualized distribution of EhCaBP3 , myosin IB and actin in over expressing EhCaBP3S cells . EhCaBP3 and myosin IB were found to accumulate at the site of cup closure whereas actin was mainly present just at the neck ( Figures 8B1 and B2; lower panel ) . There was no colocalization of EhCaBP3 and actin at the tip ( Figure 8B1; lower panel ) , unlike myosin IB and EhCaBP3 . Overall this data suggests that EhCaBP3 and myosin 1B are involved in phagocytosis and both these proteins may be needed for scission of vesicles . To check whether EhCaBP3 level has any effect on recruitment of EhC2PK and EhCaBP1 in phagocytic cups; these proteins were immunostained in EhCaBP3 anti-sense cells ( Figure 9 ) . Reduced levels of EhCaBP1 and EhC2PK were observed in the phagocytic cups suggesting that EhCaBP3 may be involved in creating a macromolecular complex along with actin , EhC2PK and EhCaBP1 . We have shown earlier that EhC2PK binds liposomes in the presence of Ca2+ and recruits EhCaBP1 in a calcium dependent manner [34] . To test whether EhC2PK also recruits EhCaBP3 at the plasma membrane , we have used liposome sedimentation assay as described before [34] . The results are shown in Figure S5 . The presence of a specific immunostained band in the pellet ( which contains liposomes ) is an indicator of interaction . Unlike EhCaBP1 , EhCaBP3 bound liposomes directly without the involvement of EhC2PK in the presence of Ca2+ . Actin was found in the pellet only when EhCaBP3 was present ( Figure S5A ) . The interaction required the presence of Ca+2 as EGTA reduced the intensity of bands in western immunostaining . EhCaBP1 alone was not able to bind liposomes and EhC2PK-bound liposomes , as expected ( Figure S5C , B ) . This suggests that EhCaBP3 alone can bind lipids , and consequently membranes , unlike EhCaBP1 . These results are consistent with our previous finding that EhCaBP3 is also localized at the membrane in E . histolytica [35] .
Regulated actin dynamics is required at different stages of phagocytosis and is achieved through participation of a number of molecules , many of which are actin binding proteins [12] . In mammalian system Arp2/3 complex , aphiphysin 2 , coronin , cofilin , WASP and Scar ( also called WAVE ) are some of the molecules known to participate in regulating actin dynamics by manipulating different steps , such as nucleation , polymerization , bundling and depolymerization , including fragmentation of filaments [39]–[44] . Many processes involving actin dynamics , such as cell polarity , psuedopod formation and endocytosis in higher organisms have been studied in detail and the molecular mechanisms mediating different steps of actin dynamics have been worked out [45] , [46] . However , the mechanism of initiation of phagocytosis is understood only in a few systems , of which the best studied , is opsonisation involving Fc receptors [47] . Our laboratory has shown that a C2 domain-containing protein kinase EhC2PK along with a calcium binding protein EhCaBP1 is involved in initiating a signal transduction pathway that eventually results in phagocytosis of RBCs in the protist parasite E . histolytica [34] . EhCaBP1 helps in recruiting actin at the site of phagocytosis by bridging with EhC2PK , a Ca2+-dependent membrane binding protein . This is one of the first examples of direct involvement of a calcium binding protein in actin dynamics and initiation of an endocytic process . In this report we show that E . histolytica erythrophagocytosis requires participation of yet another calcium binding protein EhCaBP3 . Our results suggest that unlike EhCaBP1 which acts only at the initiation stage of phagocytosis , EhCaBP3 is likely to participate in both initiation and phagosome closure stages . It also appears from our results that the proposed mechanism may not be applicable for RBC phagocytosis alone , but also applicable in phagocytosis of CHO cells , though the detailed mechanisms may be somewhat different . A number of our observations support the conclusion that EhCaBP3 is involved in phagocytosis . Firstly , EhCaBP3 was observed in phagocytic cups and phagosomes by fluorescence imaging of both fixed and live cells . Secondly , the rate and extent of phagocytosis was greatly reduced in cells where EhCaBP3 expression was down regulated by antisense RNA , and finally , over expression of a Ca2+ binding- defective mutant of EhCaBP3 reduced the rate of phagosome formation showing a dominant negative phenotype . Though the involvement of EhCaBP3 in the initiation of phagocytosis appears to be similar to that of EhCaBP1 , there are important differences in their chemical and biological properties . For example , Ca2+ binding affinities of the two molecules are different . The overall Ca2+ binding affinity of EhCaBP1 was more than 700 fold that of EhCaBP3 , and their dissociation constants ( Kd ) were 1 . 3 nM and 1 . 85 µM respectively [19] , [35] . This would indicate that the two proteins function optimally at different Ca2+ concentrations . However , we are not in a position to correlate actual local transient Ca2+ concentrations during phagocytosis with the function and Ca2+ binding properties of the two proteins due to lack of data about Ca2+ concentrations in E . histolytica . However , we can speculate that attachment of RBC to the surface of E . histolytica generates a local Ca2+ spike . EhCaBP3 and EhCaBP1 are likely to get activated at different stages of a spike when Ca2+ concentration can vary about 2 orders of magnitude resulting in sequential activation of the two EhCaBPs . However , we do not have at present any evidence in support of this . Further the two proteins are functionally different since EhCaBP1 is absent in newly formed phagosomes , while EhCaBP3 is present . This is an indication that EhCaBP3 may be participating in the process of phagosome closure . The presence of EhCaBP3 along with myosin IB at the tip of membranes before closure to form phagosomes strongly suggests that EhCaBP3 along with myosin IB may be involved in psuedopod extension , phagosome closure and finally release of the vesicle into the cytoplasm . Association of myosin 1B and EhCaBP3 has also been validated by a pull down assay . This appears to be similar to a mammalian long tail class 1 myosin that also localizes to phagosomes at late stages and participates in phagosome closure [48] . Further , transient localization of class 1 myosin to phagocytic cups has also been observed in Acanthamoeba , and in yeast myosin 1 facilitates different events of endocytosis , such as membrane fusion and vesicle scission [49] , [50] . Myosins are also known to manipulate dynamics of actin filaments [51] . However , the interplay between myosins and actin in filament dynamics in relation to phagocytosis , psuedopod formation and motility in E . histolytica is not yet understood . Both EhCaBP1 and EhCaBP3 can bind G- and F-actin directly . However , the effect of this binding translates into different biochemical changes . EhCaBP1 alters bundling properties of actin filaments without changing polymerization [21] . On the other hand , as shown here , EhCaBP3 enhances polymerization in addition to enhancing bundling formation . Together these two calcium binding proteins modulate dynamic properties of actin cytoskeleton , a unique feature not seen in any other system . Though the effects of EhCaBP1 and EhCaBP3 on actin polymerization and bundling were studied in vitro , we believe that the same properties are likely to be seen in vivo . Our assumptions are based on colocalization of actin filaments with these two proteins during phagocytosis and the observation that there is a reduction in phagocytosis when either the expression is reduced or mutant proteins are present . The reduction in phagocytosis is likely to be due to a defect in actin filament formations and this has been seen in reduced amount of F-actin in the phagocytic cups of EhCaBP3AS cells . It is also likely that the process of initiation is achieved through multiple steps and that EhCaBP3 interacts with as yet unknown molecules ( other than actin ) that participate in these steps . The molecular details of sequential changes in the state of actin , and the possible recruitment of other proteins by the CaBPs need to be worked out . It is not clear how EhCaBP1 moves out of the phagocytic cups before phagosome closure while EhCaBP3 does not . We suspect that other proteins , such as myosin IB may be involved as both myosin IB and EhCaBP3 were seen at the tips before closure . Since EhCaBP3 is a small Ca2+ binding protein and only contains Ca2+- binding EF-hand motifs , it is expected that its function must be executed through binding of Ca2+ . However , we observed that EhCaBP3mEF , a mutant of EhCaBP3 that could not bind Ca2+ , is present in the phagocytic cups and over expression of this mutant protein led to only a small reduction in phagocytic cup formation . This is not surprising as the mutant protein is capable of binding F-actin and can cause bundling of actin similar to the wild type protein . The reasons for reduction in cup formation , though small , as compared to phagosome formation , on overexpression of the mutant protein in spite of being recruited in the cups , may be due to involvement of other proteins . We need to characterize the initiation complex and identify all the players before we can answer this question . However , it is clear that Ca2+ binding of EhCaBP3 is necessary for phagosome formation as only Ca2+-bound form of EhCaBP3 interacts with myosin 1B , and the latter's recruitment in phagocytic cups requires the wild type protein . Therefore , it appears that Ca2+ has multiple facilitators in the form of different CaBPs , and a large number of different species ( Ca2+ bound and free forms ) participate at different steps in the process of phagocytosis . We are beginning to understand some of the steps as outlined here . The mechanism of recruitment of EhCaBP3 during the process of initiation of phagocytosis is not clear . Since it does not bind EhC2PK it may require participation of yet other unknown molecule ( s ) . Alternately , molecules that are present in the membrane or may be recruited to the membrane due to changes in local Ca2+ concentration could form initiation complexes along with EhC2PK , EhCaBP1 and actin , along with other participants . Support for this comes from our observations that EhCaBP3 can bind liposomes in the presence of Ca2+ and can also form a complex with liposomes and actin . Interestingly the requirement of a complex formation involving EhCaBP1 , EhC2PK and EhCaBP3 for initiation of phagocytosis is evident from the reduced recruitment of both EhCaBP1 and EhC2PK in EhCaBP3 down regulated cells . However , it is not clear if EhCaBP3 present in phagocytic cups migrates from other parts of the membrane or from a pool of membrane-bound EhCaBP3 , or from the cytoplasmic pool . Further studies are needed to work out the detailed mechanisms including the pathway involved in formation of a multimeric complex of these proteins . EhCaBP3 is likely to participate in multiple processes other than phagocytosis and actin mobilization . It is also present in the nucleus and the function of nuclear EhCaBP3 is not clear . Our studies show that the process of RBC phagocytosis in the human parasite E . histolytica follows a unique mechanism involving a number of molecules that have been identified only in this organism . Deciphering this pathway will be highly useful in understanding evolution of phagocytic mechanisms in eukaryotic cells , as E . histolytica is an early branching eukaryote . Moreover , phagocytosis is essential for the growth and survival of this parasite and blocking this process leads to inhibition of cellular proliferation . Therefore , unique molecules involved in the pathway could be potential targets for developing newer drugs .
E . histolytica stain HM1: IMSS and the transformants were maintained and grown in TYI-S-33 medium as described before [52] . Neomycin or Hygromycin ( Sigma ) were added at 10 µg ml−1 for maintaining transgenic cell lines as indicated . Transfection was performed by electroporation . Mid-log phase cells were harvested and washed first by PBS and then cytomix buffer ( 10 mM K2HPO4/KH2PO4 ( pH 7 . 6 ) , 120 mM KCl , 0 . 15 mM CaCl2 , 25 mM HEPES ( pH 7 . 4 ) , 2 mM EGTA , 5 mM MgCl2 ) . The washed cells were then re-suspended in 0 . 8 ml of cytomix buffer containing 4 mM adenosine triphosphate , 10 mM glutathione and 200 µg of plasmid DNA . The suspension was then subjected to two consecutive pulses of 3 , 000 V cm−1 ( 1 . 2 kV ) at 25 µF ( Bio-Rad , electroporator ) . The transfectants were initially allowed to grow without any selection for 48 h . Selection was carried out by adding G418 or hygromycin B ( 10 µg ml−1 ) depending on the plasmid used . EhCaBP3 gene was cloned in the BamH1 site of pEh-Neo-GFP vector . The vector has been previously constructed ( Gullien , unpublished ) by cloning the GFP mut3 allele of GFP [53] in the unique BamH1 site of the pExEhNeo plasmid [54] . Calcium binding defective mutant was also cloned in the pEh-Neo-GFP vector at the C-terminus of GFP . The CAT gene of the shuttle vector pEhHYG-tetR-O-CAT [55] was excised using KpnI and BamHI and EhCaBP3 gene was inserted in its place in either the sense or the antisense orientation . The sequences of oligonucleotides used for making the above stated constructs are described in the Supplementary Table S1 . Standard molecular techniques were used for making all these constructs . Co-sedimentation assay was carried out following published conditions [21] . Briefly , 5 µM of rabbit muscle G-actin ( Sigma ) was polymerized in a polymerization buffer containing 100 mM KCl and 2 mM MgCl2 at room temperature for 1 h . After polymerization , actin was mixed with 1 mM ATP and appropriate target protein ( 5 µ M ) in a total volume of 150 µl of G-buffer ( 10 mM Tris-Cl , pH 7 . 5 , 2 mM CaCl2 , 2 . 5 mM β-Mercaptoethanol , 0 . 5 M KCl , 10 mM MgCl2 ) and incubated for 2 h at room temperature . The samples were centrifuged at 100 , 000 g for 45 min at 4°C . The supernatant ( one-fourth of the total ) and pellet fractions ( total ) were analysed by 14% SDS-PAGE followed by Coomassie blue staining . In addition to WT ( EhCaBP3 ) , mutant ( EhCaBP3mEF ) , EhCaBP1 and EhCaBP2 were also used as positive and negative controls respectively . Solid phase G-actin binding assay was carried out as described before [21] . Briefly , different wells of a 96-well plate were coated with 5 µM G-actin in PBS overnight at 4°C and were blocked with 3% BSA in PBS for an additional 24 h . After washing with PBS-T ( PBS containing ( 0 . 05% Tween-20 ) , EhCaBP3 and CaBP3mEF were added to the wells in duplicates at concentrations ranging from 0 . 5 µM to 10 µM . Bound protein was detected with anti-EhCaBP3 antibody followed by HRPO-linked anti-rabbit IgG using the colorimetric substrate TMB ( Sigma ) . The absorbance was monitored at 405 nm with a microplate reader ( Bio-Rad , USA ) after stopping the reaction with 2 N H2SO4 . The reaction was carried out in the presence of 5 mM CaCl2 or 2 mM EGTA as indicated . Polymerization assay was done as per the protocol supplied by the manufacturer ( www . cytoskeleton . com ) . Polymerization of actin was monitored by an increase in fluorescence of pyrene-labeled actin ( cytoskeleton , USA ) with excitation at 366 nm and emission at 407 nm . The assays were carried out at 20°C in a Safas Flx spectrofluorimeter . A 100 µl sample containing 3 µM pyrene-labelled G-actin , was saturated with increasing concentration of EhCaBP3 ( 3 µM , 5 µM , 10 µM and 15 µM ) . EhCaBP1 ( 5 µM ) was used as a control and the reactions were carried out in polymerization buffer ( 5 mM Tris-HCl , pH 7 . 5 , 1 mM dithiothreitol ( DTT ) , 0 . 2 mM ATP , 0 . 1 mM CaCl2 , 0 . 01% NaN3 , 0 . 1 M KCl and 1 mM MgCl2 ) . The assays were carried as described before [21] and details are given in Text S1 . CNBr-activated Sepharose-4B ( 1 g , Pharmacia ) was conjugated with anti-EhCaBP3 antibody following a protocol supplied by the manufacturer . Briefly , crude immunoglobulins were collected from the immunized serum using 40% ammonium sulphate and subsequently dialysed in coupling buffer ( bicarbonate buffer ) . Usually , 10 mg protein was added per gram of resin . The resin was mixed gently for 18 h at 4°C . After coupling the coupled resin was processed as per the manual provided by the manufacturer . The conjugated Sepharose beads were incubated with E . histolytica lysate for 6 h at 4°C . The beads were then washed with wash buffer ( 10 mM Tris-Cl ( pH 7 . 5 ) , 150 mM NaCl , 1 mM imidazole , 1 mM magnesium acetate , 2 mM β-ME and protease inhibitor cocktail ) three times . Ca2+ and EGTA were maintained throughout the process as required . After incubation the beads were washed sequentially with 60 mM Tris-Cl ( pH 6 . 8 ) , 100 mM NaCl and with 60 mM Tris-Cl ( pH 6 . 8 ) . The pellet was suspended in 2× SDS polyacrylamide gel electrophoresis ( PAGE ) buffer and boiled for 5 min followed by centrifugation for 5 min . The proteins were then analyzed by western blotting . E . histolytica cells were labelled for immunofluorescent imaging following methods described before [21] . Cells were first washed with PBS and incomplete TYI-S-33 medium , and then resuspended in the same medium before transferring onto acetone-cleaned coverslips placed in a Petri dish . The cells were allowed to adhere for 10 min at 37°C and then were fixed with 3 . 7% paraformaldehyde ( PFA ) for 30 min at 37°C after removing the culture medium . The fixed cells were then permeabilized with 0 . 1%Triton X-100/PBS for 5 min . Additional treatment using chilled methanol ( −20°C ) for 3 min was needed for staining myosin IB . Fixed cells were then washed with PBS and quenched with 50 mM NH4Cl for 30 min at 37°C , followed by blocking with 1%BSA-PBS for 1 h . The cells were then incubated with primary antibody for 1 h at 37°C , followed by washing with PBS and 1%BSA-PBS before incubation with secondary antibody for 45 min at 37°C . When F-actin was labelled with phalloidin , the methanol step was omitted . Antibody dilutions used were: EhCaBP3 at 1∶50 ( purified antibody ) , EhCaBP1 at 1∶200 , phalloidin ( Sigma; 1 mg/ml ) at 1∶250 , GFP monoclonal ( Molecular Probes , Cat no . A11120 ) at 1∶250 , myosin IB at 1∶30 [56] , anti-rabbit or mice Alexa 488 ( Molecular Probes , Catalogue No . A-11008 or A-11001 ) at 1∶200 , anti-rabbit or mice Alexa 555 ( Molecular Probes , Cat . No . A-21428 or A-21422 ) at 1∶300 . The preparations were further washed with PBS and mounted on a glass slide using DABCO [1 , 4-diazbicyclo ( 2 , 2 , 2 ) octane ( Sigma ) 10 mg/ml in 80% glycerol] . The edges of the coverslips were sealed with nail-paint to avoid drying . Confocal images were visualized by using an Olympus Fluoview FV1000 laser scanning microscope . CHO cells were stained for 30 min with 20 mM Cell tracker orange dye ( Molecular probes , Eugene , OR ) in F12 medium containing 10% FCS . After staining , CHO cells were washed three times with fresh BI-S-33 medium , and approximately 4×105 CHO cells were incubated with 2×105 cells of amoeba expressing GFP-CaBP3 for indicated time points at 37°C in 500 µl of TYI-33 medium . The cells expressing GFP-EhCaBP3 were plated onto a 35 mm Mat Tek glass bottom culture dish ( MatTek Corporation ) at 37°C . The medium was then removed after the cells got settled at the bottom and the glass chamber was filled with pre-warmed PBS . The dish was kept on a platform with a temperature controller to maintain temperature at 37°C . High-resolution fluorescent time-lapse imaging ( Nikon A1R ) of a moving and phagocytosing amoeba was performed . The images were captured at 8 s interval and 60× objective was used . The raw images were processed using NIS element 3 . 20 or Image J software available freely on the web ( http://rsb . info . nih . gov/ij/ ) . Samples were separated on a 14% SDS–PAGE and the gel was then transferred on to a polyvinylidine fluoride ( PVDF ) membrane by semi-dry method and processed using standard protocols . The antigens were detected with polyclonal anti-GFP ( 1∶5000 , Molecular probes; Cat . No . A6455 ) or anti-EhCaBP1 or EhCaBP3 raised in mice and rabbits ( 1∶5000 , raised in our laboratory ) , followed by secondary anti-rabbit and anti-mice immunoglobulins conjugated to HRPO at 1∶10 , 000 dilution ( Sigma , Cat No's A6667 or A2554 ) . ECL reagents were used for visualization ( Millipore ) . The concentration of proteins in a sample was estimated by bicinchoninic acid ( BCA ) assay using BSA as a standard . The assay was carried as described before [57] and details are given in Text S1 . The liposomes were prepared as described by Avanti Polar Lipid , Inc . http://avantilipids . com . The proteins were incubated with liposomes in binding buffer ( Tris-Cl ( pH 7 . 5 ) 10 mM , β-ME 0 . 25 mM , NaCl 50 mM ) . CaCl2 and EGTA were used at 2 mM and 5 mM respectively at 37°C for 2 h with intermittent tapping . The liposomes were centrifuged at 18 , 000 g for 30 min , followed by washing with binding buffer to remove the nonspecific-binding proteins . Liposomes were than dissolved in SDS buffer and separated on SDS–PAGE . Specific proteins were detected by western blotting . For actin-binding assay the liposomes were incubated in polymerization buffer ( Tris-Cl ( pH 7 . 5 ) 10 mM , MgCl2 2 mM , KCl 50 mM , ATP 2 . 5 mM , β-ME 2 . 5 mM ) with EhCaBP3 , and actin . Both mice and rabbits used for generation of antibodies were approved by the Institutional Animal Ethics Committee ( IAEC ) , Jawaharlal Nehru University ( IAEC Code No . : 18/2010 ) . All animal experimentations were performed according to the National Regulatory Guidelines issued by CPSEA ( Committee for the Purpose of Supervision of Experiments on Animals ) , Ministry of Environment and Forest , Govt . of India .
|
Entamoeba histolytica is one of the major causes of morbidity and mortality in developing countries . Phagocytosis plays an important role in both survival and virulence and has been used as a virulence marker . Inhibition of phagocytosis leads to a defect in cellular proliferation . Therefore , the molecules that participate in phagocytosis are good targets for developing new drugs . However , the molecular mechanism of the process is still largely unknown . Here , we demonstrate that Calmodulin-like calcium binding protein EhCaBP3 is involved in erythrophagocytosis . We show this by a number of different approaches including immunostaining of actin , myosin1B , EhCaBP1 and EhCaBP3 during uptake of RBC; over expression and down regulation of EhCaBP3 , and over expression of calcium defective mutant of EhCaBP3 . Our analysis suggests that EhCaBP3 can regulate actin dynamics . Along with actin and myosin 1B it can participate in both initiation and formation of phagosomes . The Ca2+-bound form of this protein is required only for progression from cups into early phagosomes but not for initiation . Our results demonstrate the complex role of Ca2+ binding proteins , EhCaBP1 and EhCaBP3 in regulation of phagocytosis in the protist parasite E . histolytica and the novel mechanisms of manipulating actin dynamics at multiple levels .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"parastic",
"protozoans",
"signal",
"transduction",
"molecular",
"cell",
"biology",
"entamoeba",
"histolytica",
"parasitology",
"protozoology",
"biology",
"microbiology",
"host-pathogen",
"interaction"
] |
2012
|
The Calmodulin-like Calcium Binding Protein EhCaBP3 of Entamoeba histolytica Regulates Phagocytosis and Is Involved in Actin Dynamics
|
The vertebrate gut harbors a vast community of bacterial mutualists , the composition of which is modulated by the host immune system . Many gastrointestinal ( GI ) diseases are expected to be associated with disruptions of host-bacterial interactions , but relatively few comprehensive studies have been reported . We have used the rhesus macaque model to investigate forces shaping GI bacterial communities . We used DNA bar coding and pyrosequencing to characterize 141 , 000 sequences of 16S rRNA genes obtained from 100 uncultured GI bacterial samples , allowing quantitative analysis of community composition in health and disease . Microbial communities of macaques were distinct from those of mice and humans in both abundance and types of taxa present . The macaque communities differed among samples from intestinal mucosa , colonic contents , and stool , paralleling studies of humans . Communities also differed among animals , over time within individual animals , and between males and females . To investigate changes associated with disease , samples of colonic contents taken at necropsy were compared between healthy animals and animals with colitis and undergoing antibiotic therapy . Communities from diseased and healthy animals also differed significantly in composition . This work provides comprehensive data and improved methods for studying the role of commensal microbiota in macaque models of GI diseases and provides a model for the large-scale screening of the human gut microbiome .
The human intestine is home to some 100 trillion microorganisms of at least 400 species . The density of bacterial cells in the colon has been estimated at 1011 to 1012 per ml , which makes it one of the most densely populated microbial habitats known [1 , 2] . The number of unique genes in the microbial pool is estimated to outnumber the genes in the human nuclear genome by two orders of magnitude [1 , 2] , and these genes contribute many essential metabolic functions to the host . The great majority of gut bacterial species have not been cultured outside the human host and are known only by fragments of their DNA sequences . A few pioneering reports have begun to survey the intestinal microbiota of humans and mice using DNA sequencing of uncultured communities [1 , 3 , 4] or using microarray-based methods [5 , 6] . It is widely expected that human disease states will be linked to characteristic transitions in the intestinal microbiota , and connections have been proposed between GI bacterial communities and obesity [7 , 8] and Crohn's disease [9 , 10] , but studies in this area are just beginning . Here we report characterization of GI microbial communities in rhesus macaques and their alteration accompanying colitis associated with SIV infection or in animals with chronic enterocolitis . The mammalian GI tract is a major locus of immune tissues responsible for blocking invasion by pathogens , and more recently , these tissues have been implicated in normal homeostasis of the gut microbiota as well . For example , B-cells of the gut associated lymphoid tissues ( GALT ) synthesize IgA , which is secreted in large amounts into the lumen of the gut , and mice genetically incapable of normal IgA synthesis have abnormally large proportions of anaerobes in the small intestine [11 , 12] . Secreted antibacterial peptides have also been implicated in regulating the composition of the gut microbiota [13 , 14] . Effects of host genotype are also documented by the finding that genetically obese mice have detectably different gut microbiota compared to wild-type controls [8] . HIV infection causes rapid and massive destruction of GALT [15–20] , and HIV infection is also frequently associated with gastrointestinal disorders , many of which are of unexplained etiology [21] . Destruction of GALT and gastrointestinal disorders are also a well-characterized consequence of simian immunodeficiency virus ( SIV ) infection in macaques [15 , 16 , 22–24] . A role for the GI microbiota in AIDS disease progression has recently been suggested–bacterial antigens are proposed to pass through the damaged GI mucosa and promote immune activation , which in turn promotes viral replication and disease progression [20] . Chronic enterocolitis is fairly common in rhesus macaques even in the absence of SIV infection or other known infectious or parasitic agents . Analysis of the clinical course and histopathology of idiopathic chronic enterocolitis shows many parallels with human inflammatory bowel disease ( IBD ) , and indeed the macaque disease has been studied as a model for the human disorder [24–26] . Evidence of proinflammatory dysfunction of the IL6-JAK-STAT3-SOCS3 pathway has been reported [24] . A role for the gut microbiota in human IBD has long been suspected , and several studies have profiled uncultured GI bacteria from healthy and diseased patients ( e . g . [9 , 27] ) . Such studies have not yet yielded a clear-cut picture of the relationship of GI microbiota to pathogenesis , though a reduction in microbial diversity has been proposed . Studies of the macaque disease have suggested that several GI microbes may be slightly more common in macaques with an IBD-like disease , but the macaque GI communities have not been comprehensively analyzed [20 , 25] . Here we characterize the macaque GI microbial communities and compare community composition in health and GI disease . To profile the bacterial taxa present , we purified bacterial DNA from samples of intestinal contents , amplified segments of the 16S rRNA gene , determined the sequences using massively parallel pyrosequencing , then used these data to identify and quantify the types of bacteria present [28] . The approach used here was based on extensive reconstruction studies [29] , which showed that known clustering of microbial communities could be recaptured using 16S rRNA gene sequences of lengths generated by pyrosequencing using technology commercialized by 454 Life Sciences [30] . These preliminary bioinformatic studies also disclosed that some short segments of the 16S rRNA gene sequence were especially useful for phylogenetic reconstruction , allowing optimized primers to be chosen for the study reported here . We found that the macaque microbiota was distinct from other vertebrates studied previously . Even in healthy animals the taxa present in the gut microbiota differed between individuals and changed substantially within individuals over time . Unexpectedly , communities from males and females also differed . Distinctive GI microbial communities were also obtained in samples of colonic contents taken at necropsy from animals with GI disease . Most of these animals were also treated with antibiotics to ameliorate their symptoms , so our analysis models human cases of colitis accompanied by antibiotic therapy . These data indicate that colitis and its treatment are associated with transitions in GI microbiota in the macaque , providing a model that may be useful in understanding the human GI microbiota in health and disease .
We surveyed a range of sample types and disease states for possible effects on the macaque GI microbiota . We analyzed a total of 100 samples , including healthy animals , SIV infected animals , and animals with chronic enterocolitis . For the colitic samples , some of the animals were SIV infected and had colitis as a result of simian AIDS , while others were colitic but not SIV infected . Sample types included colonic contents collected at necropsy , bacterial communities adhered to biopsy specimens from the upper and lower GI mucosa ( jejunum and colon respectively ) , and stool ( Table 1; detailed data for each animal is in Table S1 ) . DNA was isolated from all 100 samples and amplified by PCR using primers BSR357-A and BSF8-B , which anneal to conserved regions of bacterial 16S rRNA gene . All sequence reads extended from the BSR357-A primer . The median read length was 264 nt ( Figure 1A ) . These primers were chosen based on a series of simulations carried out to investigate the optimal region of the 16S rRNA gene to query using the short sequence reads expected from pyrosequencing . Use of a moderately conserved region yielded relatively stable phylogenetic placements , though at the expense of reduced ability to discriminate low-level taxa . Biased amplification of 16S rRNA gene sequences from mixed bacterial populations can lead to distortions in abundance estimates , but these are typically in the range of only a few fold [6 , 31–33] . To facilitate comparison among samples , only a single region of the 16S rRNA gene was amplified , and uni-directional reads were used for the analysis , so that any biases introduced during amplification are common among all samples . The primers for the 16S rRNA gene sequences were each marked with a unique DNA bar code by including distinctive 4 base sequences in the primers between the 16S rRNA gene complementary region and the binding sites for the pyrosequencing primers ( Figure 1B ) . This allowed the PCR products from many samples to be sequenced using the 454 Life Sciences [30] technology , then indexed afterwards . After removal of low quality sequences , a total of 140 , 356 useable sequence reads were obtained . All bar codes were well populated , with an average of 1404 sequences per community tested . The error rate for the bar coding procedure could be estimated by cataloging all those sequence reads with bar codes that were not among those used for labeling . The analysis indicated that only 0 . 01% of sequences were likely to be miscataloged due to errors parsing the bar codes . One DNA sample was sequenced twice to assess reproducibility . To determine the bacterial taxa present , the 16S rRNA gene sequences were aligned using NAST and GREENGENES and then inserted into pre-established phylogenetic trees of full length 16S rRNA gene sequences [34 , 35] using ARB . Over all the sequences analyzed in this study , 99 . 94% sequences aligned with previously determined 16S rRNA gene sequences ( 80 total sequences failed to align ) . The bacterial taxa in each sample were then tabulated . Comparison of the two independent sequencing experiments showed excellent reproducibility of the phylogenetic assignments ( Figure 1C ) . Ninety-four near-full length macaque bacterial 16S rRNA gene sequences from two communities were also determined by conventional Sanger sequencing to provide a check on the pyrosequencing data ( Figure 1D; Table S2 ) . As can be seen from Figure 1D , the major types and relative numbers of taxa were closely similar in the Sanger and pyrosequencing analysis for each sample , indicating that the pyrosequencing data yielded an accurate reflection of the species detected by conventional sequencing , though the minor taxa detected by pyrosequencing could not be detected in the Sanger reads due to the lower number of the latter . We first investigated the bacterial diversity present in our 16S rRNA gene sequence data . Sequence reads were aligned using NAST and compiled in OTUPicker . When sequences were condensed under conditions demanding 99% identity , about 20 , 000 different operational taxonomic units ( OTUs; groups defined by pairwise sequence identity ) were found ( Figure 2A ) . When OTUs were defined using a threshold of 97% identity or greater , a criteria that in previous studies was judged to match roughly the species level [36 , 37] , about 5 , 000 OTUs were identified . Errors introduced during pyrosequencing may influence this value , but effects are expected to be small ( discussed further in the Methods section ) . In an effort to determine whether all the OTUs present in the data set had been recovered in the pyrosequencing study , a rarefaction analysis was carried out ( Figure 2B ) . Increasingly large random subsets of the initial group of OTUs were analyzed for OTU number , and the totals plotted . If all the OTUs in the sample had been sequenced multiple times , a stable estimate would be reached at OTU values less than the number present in the full data set . As can be seen in Figure 2B , the estimates are still climbing even at the highest numbers of OTUs analyzed , indicating that substantial numbers of unseen OTUs exist in the samples and would only be detected after determining larger numbers of sequences . In an attempt to estimate the total number of OTUs in each data set , the Chao 1 estimator was used , which uses frequency of isolation information to estimate the number of unseen OTUs present in the original sample . For most of the samples , the rarefaction curves on the Chao 1 estimates did not reached a stable value , indicating that the true numbers of OTUs in the samples are larger even than the Chao 1 estimates ( 53–1185 OTUs per sample; 97% identity criteria ) . Overall the richness of the bacterial taxa in the macaque GI microbiota was very high . A comparison of the estimated diversity in all 100 samples was carried out by computing the Shannon Diversity Index from the OTU data for each sample ( Figure 2C ) . To investigate the relative diversity at different anatomical sites , the 100 communities were grouped by sample type and their relative diversity compared . Rarefaction analysis indicated that most of the Shannon Diversity estimates had reached stable values . Separating the communities by sample type indicated that the upper GI mucosal samples from the jejunum were notably less diverse than the other groups . A comparison of the macaque GI microbiota to that of humans [4] and mice [8] is shown in Figure 3 . To compare the global compositions of microbial communities , we used UniFrac [38–40] , which measures the similarity among bacterial communities based on phylogenetic distances . To carry out a UniFrac analysis , we used the augmented ARB database described above . To compare two communities using UniFrac , sequences from the two communities are marked on a common phylogenetic tree that contains all the sequences from the communities to be analyzed , and the fraction of the branch length on the tree unique to each community is then computed . This procedure provides a measure of the similarity between the two communities in terms of the total amount of evolutionary history that separates the sequences in the two communities . UniFrac assigns only a small difference to changes in representation of closely-related taxa , but larger value for changes in representation of more distant taxa , in contrast to OTU-based methods that assume that all taxa are equally distinct . To compare multiple communities , all the pair-wise distances between communities were computed , then Principal Coordinate Analysis ( PCoA ) was used to cluster the communities along axes of maximal variance ( Figure 3 ) . To compare human and mouse samples to the macaque pyrosequencing data , sequence reads determined by the Sanger method from human and mouse were first truncated to match the length and position of the macaque 16S rRNA gene sequences . The UniFrac comparison showed strong clustering by species of origin . Similar separation by species was obtained when pyrosequencing data was used for both the rhesus and murine samples ( unpublished data ) . For the human and macaque samples , the communities clustered by species of origin even though samples from diverse anatomical sites were included for each species . The taxonomic groups from GI communities of each species were then compared . The bacterial taxa detected are summarized in Figure 4A . The most prominent bacterial classes were Clostridia ( Phylum Firmicutes ) , Bacteroidetes ( Phylum Bacteroidetes ) and Spirochaetes ( Phylum Spirochaetes ) . Present in lesser amounts are Bacilli and Molicutes ( Phylum Firmicutes ) , Alpha , Beta , Gamma , and Epsilon Proteobacteria , and a collection of additional classes . Several of the minor classes were found repeatedly in specific individual macaques ( e . g . Fibrobacteres , Gemmatimonadetes , Deferribacteres ) . All of the animals showed variation over time , in both the classes detected and in their relative abundance . Many of the bacterial taxa identified were not previously known to be present in the macaque intestinal microbiota . The predominance of the phyla Firmicutes and Bacteroidetes were similar in all three vertebrates , and several lower-abundance phyla also overlapped . For example , Proteobacteria and Actinobacteria were found in both macaques and humans . Verrucomicrobia were detected in humans but were rare macaques . A distinctive feature of the macaques was the density of Phyla Spirochaetes , particularly members of the genus Treponema , which were present in abundance in macaques ( Figure 4 ) but mostly absent in the samples from in mice and humans . The abundance of flagellated Helicobater ( EpsilonProteobacteria ) has previously been noted [41] , and Spirochaetes have been identified in the gut microbiota of many vertebrates including humans and non-human primates [42 , 43] . However , the abundance of Treponema in macaques was unexpected and far greater than in human . In humans , within the Class Bacteroidetes , members of the genus Bacteroides have been reported to be a major and functionally significant component of the human intestinal microbiota [4 , 44 , 45] , but of the 94 near full length 16S rRNA gene macaque sequences , only one was genus Bacteroides . More common were genus Prevotella ( 16/94 sequences ) , which is also common in humans , and Rikenella ( 18/94 sequences ) , which is rare or absent in humans [4] . These proportions of Bacteroidaceae and Prevotellaceae were similar in the shorter pyrosequencing reads . In macaques , comparison of microbial communities among animals showed considerable variation among individuals , both in the relative abundance of the major taxonomic groups and in the presence of minor groups ( Figure 4B ) . For some animals , longitudinal samples were available , showing that the composition of the GI microbiota was quite dynamic over the period of sampling . Figure 5 shows a UniFrac clustering diagram comparing the communities from different anatomical sites . Possible clustering by sample type on the first two principal coordinates was assessed using a t-test to compare the within-group and between-group distances , then 1 , 000 label permutations were used to assess significance . Clustering for all four sample types was found to be significant at the p < 0 . 01 level . The samples from the upper GI mucosa formed a distinct cluster to the upper right of the diagram , indicating unique composition . Analysis of the taxa present indicates that the upper GI communities were depleted in bacteria from the Bacteroidetes and Clostridia classes compared to lower GI , colonic contents , or stool , and enriched in Baccili , Molicutes , and Gamma and EpsilonProteobacteria . Several minor groups were particularly common in upper GI samples , including Mycoplasmatales and Streptococaceae . Analyses of biopsies ( with adherent bacteria ) from the lower GI ( ascending colon ) showed that they intermingle with samples of colonic contents taken at necropsy , though a distinctive feature was the abundance of Helicobater at this site . Enterobacteriaceae were far more common in the upper and lower GI samples than in stool or colonic contents , indicative of probable adherence to mucosal surfaces . Stool samples form a cluster continuous with colonic contents but extending to the upper left of the UniFrac plot . Stool samples commonly differed from colonic contents samples by having greater representation of Spirochaetes and several minor groups . In an effort to identify additional parameters affecting the macaque GI microbiota , we asked whether communities clustered detectably in UniFrac analysis when partitioned by a variety of biological parameters . The parameters tested included sex of the animal of origin , age , disease state , antibiotic use , and viral infection . GI communities were analyzed as pools across all sample types , as pools of related samples ( colonic contents plus stool ) , or as single sample types ( stool only or colonic contents only ) . Unweighted UniFrac was used for these comparisons , which is based on the presence or absence of different taxa without regard to abundance . In samples of colonic contents taken at necropsy , or in samples of stool , a difference was seen between males and females . Separate clustering is illustrated for a pool of the two sample types in Figure 6 ( p < 0 . 05 by t-test and label permutation ) . Analysis of the bacterial groups involved showed that several groups of the Lachnospiracea and Bacteroidales differed ( p < 0 . 0001 ) . One Treponema group was far more common in males ( p < 0 . 0001 ) . The physiological mechanism for the observed sexual dimorphism is unknown , though partitioning of the GI microbiota by sex has been noted in mice [46] . The effects of disease states were then examined . Microbial communities from colonic contents were divided by whether host animals were diagnosed with colitis at necropsy ( Table 1 ) and analyzed in unweighted UniFrac ( Figure 7A ) . Seven samples were available for analysis from males and ten from females . Of these , nine were SIV-infected and eight were uninfected . The communities separated along the first principal coordinate by whether the animals were diagnosed with colitis ( p < 0 . 05; t-test with label permutation ) , indicating that the disease and associated treatment resulted in a change in composition of the GI microbiota . An analysis of the relative diversity , as reported by the Shannon Index , revealed that diversity was consistently lower in the communities from colitic animals ( Figure 7B ) . Most of the animals with colitis had a history of multiple bouts of diarrhea requiring medical attention including fluid therapy and in many animals treatment with antibiotics ( Table S1 ) . The antibiotics chosen for therapy differed among the animals and included tetracycline , enrofloxacin , cefazolin , and tylosin . The time of treatment relative to euthanasia and the duration of treatment also varied . Only two animals were on antibiotics ( tetracycline ) at the time of euthanasia . Within the cluster of communities from animals with colitis ( Figure 7B ) , some possible sub-clustering was seen by antibiotic type , suggesting that each antibiotic resulted in characteristic changes in community composition ( though larger sample sizes will be needed to assess this hypothesis definitively ) . An analysis of the bacterial taxa that differed between the two groups revealed the family Campylobacteraceae ( Epsilon-Proteobacteria ) was much more common in animals with colitis–for the major Campylobacter OTU ( 97% criteria ) , five out of ten monkeys with colitis had this OTU , but none of the seven healthy monkeys had this OTU ( G = 6 . 03 , df = 1 , p = 0 . 015 ) . Two monkeys of unknown clinical status were also positive . A variety of additional taxa within the Bacteroidetes and Firmicutes phyla also changed in abundance significantly in assocation with colitis . The Campylobacter genus contains known enteric pathogens of humans ( [47] and references therein ) , consistent with the idea that the presence of these groups was associated with pathogenesis in macaques . Of the animals detected as Campylobacter positive by sequence analysis , only two animals had positive cultures for Campylobacter when analyzed by conventional clinical methods . One explanation for the enrichment of Campylobacter would be that antibiotic treatment created an environment favorable for colonization , as has been suggested for Clostridia difficile in humans . Of the animals with colitis that were positive for Campylobacter , four had histories of recent antibiotic use but three did not , and for the four treated animals three different antibiotics were used ( Table S1 ) . Thus the presence of Campylobacter was not strongly associated with antibiotic treatment , consistent with the idea that Campylobacter was associated with colitis and not antibiotic use . SIV-infected animals were present in both the colitis and normal groups , and no strong clustering of the bacterial communities was associated with SIV infection when SIV infection was analyzed in isolation ( data not shown ) . These data suggest that the alterations in community composition in SIV-infected animals with colitis was attributable to the colitis resulting from viral infection , and not the viral infection itself .
In this study , we describe the composition of 100 uncultured GI microbial communities from healthy rhesus macaques and macaques with chronic colitis . Each community was characterized by an average of ∼1 , 400 reads of 16S rRNA gene of median 264 nt in length . This work provides a detailed picture of the structure of the macaque GI microbiota , its dynamics , and changes associated with colitis with or without SIV infection . Macaque models are used in studying myriad GI diseases , including SIV-induced enteropathy , bacterial enteropathy , and inflammatory bowel disease . The data presented here provides detailed background , hypotheses and methods for assessing possible involvement of the full GI microbiota , and provides a model for investigating changes in the human GI microbiota in healthy and diseased individuals . The pyrosequencing method [30] allows large numbers of 16S rRNA gene sequence reads to be obtained while controlling the costs of data acquisition , greatly increasing the number of bacterial communities and species accessible to analysis compared to culture-based methods . In the bioinformatic approach used here , the pyrosequencing reads were analyzed after first inserting them into pre-existing phylogenetic trees formed from full-length 16S rRNA gene sequences , allowing relatively accurate phylogenetic placement despite the short sequences lengths [29] . Aligning pyrosequencing reads to a pre-existing tree also serves to minimize the effects of pyrosequencing errors , since single nucleotide substitutions that cause a sequence read to align with an incorrect full length sequence will be rare . Communities characterized by 16S rRNA gene sequence reads were compared to each other using UniFrac [38 , 40] , which evaluates the distance between pairs of samples after alignment on phylogenetic trees based on the unique branch length leading to members of each community . One advantage of this approach is that the collection of pair-wise distances between communities can be subject to PCoA , allowing communities to be clustered along orthogonal axes of maximal variance . In a successful study of this type , clustering on each axis can report the effects of different biological variables . Previous studies of the vertebrate GI microbiota have indicated that many factors influence microbial populations , including host genotype [8 , 48] , geography [49] , antibiotic use [50] , and diet [51] . Using UniFrac and PCoA , in combination with case-controlled samples , it is potentially possible to extract the effects of these and other variables and analyze each independently . Our analysis showed that the macaque microbiota differed significantly from that of mouse or human . Even when communities from different anatomical sites were considered , or when samples from healthy hosts were mixed with diseased hosts , the effect of species of origin was still predominant . For all three vertebrates , the Firmicutes and Bacteroidetes comprised the most abundant phyla , but the composition of minor groups differed and the taxa within the Firmicutes and Bacteroidetes also differed . A distinctive feature of the macaque samples was the abundance of Spirochaetes from the Treponema lineage . These Treponema differ from the spiral-shaped Helicobacter reported previously [41] , which were also detected here . Analysis of full-length 16S rRNA gene clones ( Figure 1D ) showed closest matches to Treponema brennaborense and Treponema saccharophilum . T . brennaborense has been associated with digital dermatitis in dairy cows [52] . T . saccharophilum has been identified as a component of the rumen GI flora that aids in digestion of pectin [53] , suggesting a possible role in digesting vegetable matter in the macaque GI tract . The analysis of healthy animals emphasized the many factors affecting composition of GI communities in macaques . The number of types of bacteria involved is very large–when macaque 16S rRNA gene sequences are grouped into OTUs at 97% or greater similarity , a threshold that has been suggested to correspond roughly to the species level , about 5 , 000 OTUs were identified . Microbial communities of individual animals differed from one another , and all animals followed longitudinally showed changes in community composition over time . Similarly in humans , GI microbial communities have been reported to differ among individuals and at different anatomical sites [4] . The macaque GI communities also clustered by the sex of the host animal , paralleling a proposal for sexual dimorphism in the GI microbiota in mice [46] . Samples from colonic contents of animals euthanized due to advanced colitis showed distinctive communities compared to similar samples from healthy controls , linking alterations in the GI microbial communities and GI pathogenesis . Samples from animals with colitis , whether associated with SIV infection or not , were indistinguishable . This emphasized that colitis itself ( and associated therapeutic interventions ) and not the cause of colitis was most tightly linked with altered GI microbiota . The presence of Campylobacter was strongly associated with colitis . The major Campylobacter OTU ( 97% threshold ) was present in five out of ten animals with colitis , but in zero out of seven free of colitis ( p = 0 . 014 ) . Cultureable C . jejuni or C . coli were obtained only from two animals , indicating that the Campylobacter species detected were either too rare to detect by culture , or did not grow under the culture conditions used . Most of the macaques euthanized due to GI-disease were treated with antibiotics at some point during disease progression . Thus these findings model human clinical cases where antibiotic therapy can be indicated in the treatment of colitis , but antibiotic treatment complicates analysis of effects of GI disease alone . For the samples of colonic contents taken at necropsy , there were indications of clustering due to type of antibiotic used for treatment within the larger cluster of samples from animals with colitis , though the number of samples in each antibiotic group was too low for detailed analysis by antibiotic type . Our data are consistent with the idea that the disease state caused a shift in bacterial communities that was further shaped by the antibiotics used for treatment . The sequence-based approach described here has the potential to identify candidate pathogens involved in previously obscure disease conditions . Animal FH09 ( Table S1 ) provides a case study . This animal suffered from prolonged chronic diarhhea of unknown cause . Exhaustive searches for a microbial pathogen by conventional culturing methods were negative . For unexplained reasons , placing the animal on a gluten-free diet helped ameliorate the condition , but eventually the animal declined and was euthanized for humanitarian reasons . Analysis of colonic contents taken at necropsy revealed a substantial number of 16S rRNA gene sequences ( 51 reads ) that clustered with a group containing Campylobacter fetus and Campylobacter hyointestinalis . Evidently Campylobacters of this group are not detected in the usual culture assays . C . fetus has been implicated as an emerging pathogen and could well have been involved in the GI disease of FH09 . These findings suggest that further analysis of the relationship between diet and C . fetus pathogenesis might be useful , and illustrate how the methods described here could be applied in diagnosis of human GI diseases of unknown etiology . In summary , this study presents the first use of DNA bar coding and pyrosequencing to analyze uncultured bacterial communities from the primate gut , and provides the deepest view into the gut microbiome from the largest sample of any non-human species to date . Using the macaque model and the methods reported here , it will be possible to investigate how the interaction among bacterial community members , together with alterations in the GI environment , leads to outgrowth of pathogenic forms and resultant disease . This study also paves the way for broader application of pyrosequencing to characterize the human microbiota in health and disease , which could potentially allow large-scale characterization of thousands of human samples with orders of magnitude less expense and effort than traditional Sanger sequencing . We will thus soon be able to identify those features of the microbiota ( if any ) that are common to all healthy individuals , and to assess the extent to which changes in the microbiota in animal models can help guide the development of therapy for human diseases .
Rhesus macaques ( Macaca mulatta ) were housed singly at the Tulane National Primate Research Center . For longitudinal studies of stool samples , four animals ( CC47 , FH40 , CT64 , DD05; here M1-M4 ) were infected intravenously with 100 TCID50 SIVmac251 on study day 0 . Fecal samples were collected prior to infection ( t1 ) , at day 7 ( t3 ) , day 14 ( t4 ) , day 28 ( t6 ) and day 56 ( t10 ) post infection . These are standard time points for examination of early events in the pathogenesis of AIDS and are associated with peak viremia ( day 14 ) and establishment of viral set point ( by day 56 ) . Stool samples for control animals ( AM87 , DG23 , CC79 , BA02; here C1-C4 ) were collected similarly over an eight week period . For samples of colonic contents , each was collected from the ascending colon at necropsy within one hour of euthanasia with an intravenous overdose of phenobarbital . All samples were immediately frozen to −80 °C . Samples were shipped on dry ice and stored at −80 °C until processing . In addition , intestinal biopsies of the upper ( jejunum ) and lower ( ascending colon ) were obtained by standard techniques . These biopsies were immediately frozen as for colonic contents . Housing and handling of animals were in accordance with the Guide for the Care and Use of Laboratory Animals ( U . S . Public Health Service ) and the Animal Welfare Act . All protocols and procedures were reviewed and approved by the Tulane University Institutional Animal Care and Use Committee . Additional animals studied , their clinical conditions , and detailed ecological descriptions of samples are in Table S1 . Total DNA was extracted from frozen stool using the QIAamp® DNA Stool Mini Kit ( Qiagen , Inc . , Valencia CA ) , following the manufacturer's protocol for pathogen detection . For samples from each animal and at each time-point , the 16S rRNA gene was amplified from extracted DNA using the composite forward primer 5′-GCCTCCCTCGCGCCATCAGNNNNCTGCTGCCTYCCGTA-3′ where the underlined sequence is that of 454 Life Sciences® primer A and in italics is the broad range bacterial primer BSR357 . The reverse primer was 5′-GCCTTGCCAGCCCGCTCAGNNNN AGAGTTTGATCCTGGCTCAG-′3 , where the underlined sequence is that of 454 Life Sciences® primer B and in italics is the broad range bacterial primer BSF8 . The NNNN designates the unique four base bar code used to tag each PCR product . Reaction conditions were as follows: 5 . 0 μl 10× PCR buffer II ( Applied Biosystems , Foster City , CA ) , 3 . 0 μl MgCl2 ( 25 mM; Applied Biosystems ) , 2 . 5 μl Triton X-100 ( 1% ) , 2 . 0 μl deoxyribonucleoside triphosphates ( 10 mM ) , 1 . 0 μl forward primer and 1 . 0 μl reverse primer ( 20 pmol/μl each ) and 0 . 5 μl AmpliTaq® DNA polymerase ( 5U/μl; Applied Biosystems ) and 100 ng of template DNA in a total reaction volume of 50 μl . Reactions were run in a GeneAmp® PCR System 9700 cycler ( Applied Biosystems ) using the following cycling parameters: 5 minutes denaturing at 95 °C followed by 20 cycles of 30 secs at 95 °C ( denaturing ) , 30 secs at 56 °C ( annealing ) and 90 secs at 72 °C ( elongation ) , with a final extension at 72 °C for 7 minutes . Four independent PCR reactions were performed for each sample along with a no template negative control . Each PCR product was gel purified from a 0 . 8% agarose gel . DNA was isolated using the QIAquick® Gel extraction kit ( Qiagen , Inc . , Valencia CA ) . 100 ng of each of the 100 gel purified DNAs was added to a master pool of DNA which was sent for pyrosequencing with primer A as described [30 , 54] . Several studies have analyzed sources of error in 454 sequencing runs , which informed our choices for quality control here [37 , 54 , 55] . For a sequence to pass quality control , it needed to ( 1 ) show a perfect match to the bar code and 16S rRNA gene primer , ( 2 ) be at least 50 nt in length , ( 3 ) have no more than two undetermined bases in the sequence read , and ( 4 ) find at least a 75% match to a previously determined 16S rRNA gene sequence after alignment with NAST ( http://greengenes . lbl . gov/ ) . The sequences were inserted into the 16S rRNA gene tree constructed by Hugenholz et al . [56] using the parsimony insertion tool from ARB software ( http://www . arb-home . de/ ) . A “termini” filter was used for the parsimony insertion . After applying this criteria , 36 , 652 , 141 bases of sequence were available for analysis . All sequence data will be deposited at NCBI upon acceptance of this manuscript for publication . OTU clustering and analysis was carried out using OTUPicker ( M . Hamady and R . Knight , unpublished ) . Clustering and principal coordinate analysis were conducted using UniFrac [29 , 38 , 39] . UniFrac analysis can be carried out based on the presence and absence of bacterial taxa ( unweighted UniFrac ) , or taking into account abundance information on each group ( weighted UniFrac ) ; Figures 3 , 4 , 6 , and 7 report unweighted UniFrac results . To perform permutation tests within UniFrac , we randomized the labels of each group and repeated the cluster analysis . We then compared all distances between points that both came from the same group to all distances between points that came from different groups using a t-test . In the permutation test , we obtained a nonparametric distribution for the t statistic that takes into account the correlations introduced by the distance matrix structure . We used 1 , 000 permutations , so we cannot specify p-value more precisely than “<0 . 001” if none of the permuted sets gave a more extreme result than the actual set . We note that the principal coordinate analysis assumes that the relationships between taxon abundance and environmental gradients is linear . In choosing the Monte Carlo methods used for significance testing , we accepted reduced power to avoid using parametric methods , which assume random distribution in the error terms . The taxonomy assignments were based on the group names in Arb . Ecological parameters in Table S1 were calculated using OTUPicker and PAST [57] . Errors in pyrosequencing may occur at a rate of about 0 . 25% [37] , suggesting that the most of the 260-nucleotide sequences that remain after filtering will contain either 0 or 1 errors . Single- nucleotide errors will not affect either of the analyses we present ( high-level taxonomic breakdowns or UniFrac ) substantially , as they are unlikely to cause assignment of pyrosequence reads to the wrong taxonomic group and contribute almost no branch length to the phylogenetic tree used for UniFrac analyses . However , these sequencing errors could affect estimates of the total number of OTUs at a given threshold , so some caution in interpreting the total number of species-level taxa in the samples is required . Using the Poisson model , we would expect only 4 . 4 × 10−5% of the reads to contain the seven errors that would be required to form a new species-level at the 97% OTU threshold . Thus , it is unlikely that a single OTU in the analysis was generated through that mechanism .
|
Bacterial mutualists within the gastrointestinal tract aid digestion , promote development of the gut immune system , and provide competitive barriers to pathogen invasion . The host , in return , provides bacteria with safe housing and food during lean times . The composition of the gut microbiota is controlled in part by the host immune system . In a variety of disease states , immune function can be altered , and gut morbidity is often associated , leading to the hypothesis that alterations in the GI microbiota may contribute to disease . In this study , the gut microbiota was characterized in 100 samples from rhesus macaques using pyrosequencing , which allowed 141 , 000 sequences from 16S rRNA genes to be generated and analyzed . Healthy animals were compared to animals with gut disorders , induced , for example by advanced simian AIDS . Many factors contributed to changes in the microbiota , including the sex of the animal of origin . Animals with chronic colitis showed differences in composition of the GI microbiota compared to healthy animals , providing an association between altered microbiota and disease .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biotechnology",
"primates",
"infectious",
"diseases",
"pathology",
"virology",
"microbiology",
"computational",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics",
"eubacteria"
] |
2008
|
The Macaque Gut Microbiome in Health, Lentiviral Infection, and Chronic Enterocolitis
|
Circadian entrainment is necessary for rhythmic physiological functions to be appropriately timed over the 24-hour day . Disruption of circadian rhythms has been associated with sleep and neuro-behavioral impairments as well as cancer . To date , light is widely accepted to be the most powerful circadian synchronizer , motivating its use as a key control input for phase resetting . Through sensitivity analysis , we identify additional control targets whose individual and simultaneous manipulation ( via a model predictive control algorithm ) out-perform the open-loop light-based phase recovery dynamics by nearly 3-fold . We further demonstrate the robustness of phase resetting by synchronizing short- and long-period mutant phenotypes to the 24-hour environment; the control algorithm is robust in the presence of model mismatch . These studies prove the efficacy and immediate application of model predictive control in experimental studies and medicine . In particular , maintaining proper circadian regulation may significantly decrease the chance of acquiring chronic illness .
Control theoretic tools have been used to model mRNA transcriptional/translational regulatory feedback mechanisms [1] , to analyze nonlinear phenomena [2] , [3] , and to control complex biological behavior [4] , [5] . In our research , we couple systems theoretic tools ( such as sensitivity analysis ) with model predictive control , to better address phase resetting properties of nonlinear biological oscillators . Our work aims to alleviate circadian-related disorders ( such as jet lag and advanced/delayed sleep phase syndromes ) by investigating the phase resetting properties of an example circadian mathematical model . More specifically , we manipulate multiple control inputs ( or target parameters ) to drive the dynamic behavior of the system . Many researchers have shown that the systematic application of light pulses may reset the phase of circadian clocks . This light pulse ( input ) to induced phase-shift ( output ) mapping is most notably characterized by the phase response curve ( PRC ) . Daan and Pittendrigh studied the PRC to establish a relationship among circadian behavior ( nocturnal vs . diurnal activity ) , free-running period , and maximum phase advance/delay [6] . The free-running period of an organism reflects its circadian behavior without the influence of entrainment factors such as environmental light∶dark cycles . The free-running period of nocturnal animals , for instance , is often less than 24 hours such that dusk triggers a phase delay and the onset of activity . Conversely , diurnal animals often exhibit free-running periods greater than 24 hours such that dawn triggers a phase advance and the onset of activity [6] . Other researchers have made use of PRCs to establish light as a means to accelerate circadian entrainment [7] , or as a means to start , stop , and reset the phase of simplified circadian models [8]–[11] . In a previous study , we develop a closed-loop nonlinear model predictive control ( MPC ) algorithm that minimizes the phase difference between a reference and a controlled system ( each modeled as a single deterministic oscillator ) through the systematic application of continuous light . Through use of MPC , circadian phase is recovered in almost half the time required by the natural open-loop sun cycles [4] . Next , we investigated how the MPC algorithm's tuning parameters might affect the model's phase resetting dynamics [12] . Here , we make use of sensitivity analysis to identify additional control inputs ( or drug targets ) that , when used by the MPC algorithm , outperform light-based circadian phase resetting . The target identification of single and multiple control inputs , coupled with the analysis of their respective performance , parallels efforts in the pharmaceutical industry to yield the greatest behavioral response with respect to the smallest system perturbation . In other words , our methodology may be used to identify optimal ( and arguably non-intuitive ) drug targets for therapy . To establish an upper bound relating to the time required to recover phase differences , we begin by evaluating the open-loop control algorithm in the Open-Loop Phase Recovery section . The identification and manipulation of a set of single , dual , and triple control inputs are then used to minimize phase recovery dynamics of a wild-type circadian system ( as described in the Single , Dual , and Triple Target Phase Resetting sections , respectively ) . This case is most similar to resetting a healthy organism's phase when subject to an environmental disturbance such as jet lag . In the Short and Long Period Mutants section , we further investigate how MPC may be used to alleviate chronic circadian disorders . More specifically , we apply the algorithm to circadian oscillator models that exhibit either short or long-period mutant phenotypes . Results suggest that organisms with such syndromes may track regular 24 hour rhythms through the systematic application of light . Our findings support this unique application of systematic drug target identification coupled with model predictive control for use in medicine and pharmacology ( see the Discussion section ) . In the Methods section , we describe the employed model predictive control algorithm and the state-based sensitivity analysis used to identify single and multiple parametric control inputs .
Due to the inherent nonlinear phase response of circadian rhythms when subject to environmental/parametric perturbations , phase recovery dynamics are characterized as a function of the initial condition ( IC , the circadian time at which control or entrainment begins ) , and initial phase difference ( IP , the amount of circadian time to be recovered ) . To establish a phase resetting set point or upper bound ( the maximum amount of time required to recover a given phase difference ) , we evaluate the open-loop control algorithm , where environmental light∶dark cycles serve as the only mechanism for phase re-entrainment . The phase recovery surface ( Figure 2 ) displays the time required for the open-loop case to recover from any possible initial condition and initial phase difference . The asymmetry of the surface may be attributed to the nonlinear effects of light , as characterized by the PRC . The input ( light ) to output ( induced phase shift ) mapping of the PRC is seldom symmetric . In Drosophila melanogaster , a 15 minute pulse of light has shown to induce up to 3 . 6 hours of phase advance and 4 . 2 hours of phase delay [13] . Recent studies suggest that the change in phase is less sensitive to the duration of the light , and more sensitive to its time-profile [17] . Phase recovery times ( for both open and closed-loop simulations ) are evaluated with respect to initial conditions and phase differences discretized at 3 hour intervals . Thus , given the integers i , j ∈ [0 , 7] , IC = 3i and IP = 3j . The open-loop entrainment strategy requires at most 183 hours to reset the observed states of the controlled system ( cumulative protein complex concentrations ) to within 15% of the reference trajectories . Mandating the convergence of state trajectories is a tighter constraint than mandating only phase trajectories , since it incorporates amplitude characteristics . The algorithm , however , may be tuned to consider only strict phase measures . The maximum open-loop recovery time refers to a 9 hour initial phase difference whose control action begins at an initial condition of 15 hours . The initial condition , or start of entrainment , is described with respect to circadian time ( CT ) . Interestingly , there is a stark difference between resetting a 3 to 6 hour initial phase difference versus an 18 to 21 hour initial phase difference ( a −6 to −3 hour phase difference ) . In the former , phase recovers in over 100 hours; in the latter , phase recovers in fewer than 60 hours . Additionally , the open-loop algorithm recovers 9 hour phase differences in a fraction of the time required to correct for smaller phase difference . These properties may be attributed to the nature of the phase response curve and are discussed further in the Discussion section . Experimental studies in mammalian SCN cells support this asymmetry: Reddy et al . show that circadian clock resetting from a 6 hour phase advance ( IP6 ) is accompanied by dissociation of cellular gene expression and may take up to 1 week to recover [18] . Conversely , resetting a 6 hour phase delay ( IP18 ) is accompanied by coordinated gene expression and requires only 2 days of recovery [18] . Our simulations support these experimental conclusions as the cumulative protein concentrations in the former case diverge and require several days to converge to the nominal trajectory . In the latter , cumulative protein concentrations oscillate with smaller amplitude until they converge to the nominal trajectory within a couple days . An example of the corresponding simulations is presented in Figure S1 . The MPC algorithm ( described in the Model Predictive Control section ) minimizes the normalized difference between the cumulative protein complex concentration over a prediction horizon of 48 hours , by admitting control action during the first 8 hours of the simulated trajectory . This control action is multiplicative , allowing the algorithm to increase/decrease the nominal parameter by a factor of 2 . The control profile defined within the move horizon is updated every 2 hours . Through use of MPC , the re-synchronization rate of the controlled system is increased nearly 3-fold through the control of light , or νdT . Although light serves as a powerful control input , we show that the manipulation of parameters such as transcription and mRNA degradation rates ( νs and νm , respectively ) may provide more immediate phase resetting . Since we make use of the symmetric version of the mathematical model [13] , we do not differentiate between per or tim specific functions . Instead , we assume that the isolated control of νsP is equivalent to the isolated control of νsT , for instance . Parametric sensitivity analysis quantifies the relative change of system behavior with respect to an isolated parametric perturbation . A large sensitivity to a parameter , for instance , suggests that the system's performance is subject to greater change with small variations in the given parameter . We make use of the Fisher Information Matrix ( FIM ) to evaluate the effect of parametric perturbations on the circadian system's state trajectories [19] . Investigation of the diagonal values , off diagonal values , and singular value decomposition of the FIM points out the relative order , or rank , of parametric sensitivity measures . This relative ordering highlights sets of control inputs whose manipulation may further reduce phase recovery times . The three greatest diagonal values , for instance , identify the most prominent individual control targets ( ranked from most to least sensitive ) ; Recall that νdT is the target parameter of environmental light in Drosophila . Interestingly , the rate of mRNA transcription is the target of environmental light in Mus ( via per genes ) [20] , [21] and Neurospora ( via frq genes ) [22] . Furthermore , in our previous studies of Mus and Drosophila circadian networks , mRNA transcription rates were among the most sensitive parameters with respect to both the state- and phase-based sensitivity analysis of two independent network representations [2] . The greatest off diagonal values identify the most prominent pairs of control targets ( ranked accordingly ) ; Since the manipulation of more than 1 parameter voids the symmetry argument , we target tim specific parameters in the implementation of multiple control targets . The greatest input directions of the singular value decomposition identify the most prominent set of three control targets ( ranked accordingly ) ; We investigate the phase recovery dynamics corresponding to four independent isolated control inputs with respect to the initial condition and initial phase difference ( Figure 3 ) . Results show that control targets identified via sensitivity analysis ( Figure 3 ( A ) –3 ( C ) ) serve as more effective re-entrainment factors than light ( Figure 3 ( D ) ) . More specifically , the maximum recovery time corresponding to a control input of νs is 44 hours ( at IC9 and IP12/IP15 ) , νm is 50 hours ( at IC21 and IP15 ) , ks is 59 hours ( at IC12 and IP15 ) , and νd ( the light target ) is 60 hours ( at IC12 and IP15 , or IC9 and IP12 ) . The control profiles and state response dynamics relating to the phase recovery of IC9 and IP12 are provided in Figure S2 and Figure S3 . There is a subtle similarity among the single-input phase recovery data; namely , the sudden drop in recovery time with respect to the initial condition for initial phase differences of 0 to 15 hours . We attribute this steep recovery gradient to the PRC as it depicts a greater region of phase delay than it does a phase advance . For this reason , it is more beneficial if the organism delays its phase to recover from a 12 hour initial phase difference . Furthermore , recall that a phase delay is incurred if the organism is to receive a photic input in the late evening hours . Hence , recovering from a phase difference via a set of delaying control inputs is most efficient if control action begins around the late subjective evening . Thus , if we observe phase resetting behavior corresponding to a small phase difference ( such that the subjective day of the controlled system and reference are similar ) , we expect it to have the shortest recovery time near an initial condition of 12 hours , or dusk ( Figure 3 ( D ) ) . Interestingly , each of the control inputs exhibits this property . We attribute this similarity to the unique PRC of each control input ( Figure 1 ) . Mutant phenotypes of the circadian oscillator represent cases in which nominal light∶dark cycles are unable to maintain synchrony . For this reason , the MPC tuning parameters must be re-evaluated according to this phase resetting problem . In wild-type , for instance , we can afford to be more aggressive with control penalties since nominal light∶dark cycles ( or , no control ) will eventually entrain the system . In mutants , the weights used to penalize the state error and control inputs prove to be more influential since nominal light∶dark cycles will not entrain the system . Therefore , we set both the move and prediction horizon to 24 hours and reduce the penalty of state error and control to ones . To counter the computational expense incurred with a longer move horizon , we set the time step to 4 hours . Through MPC , we identify a more suitable light∶dark cycle that synchronizes organisms exhibiting abnormally short and long free-running periods ( 22 and 27 hours , respectively , as shown in Figure 5 ) . Determining the complete range of entrainment ( which is likely wider than the 22 to 27 hour period ) is non-trivial . In a previous study , we found that ( i ) the predicted range of entrainment may be very sensitive to the employed performance metric , and ( ii ) the control/light input strength may also play a dominant role in defining the bounds of this range [23] . Given that the PRC characterizing the behavior of Drosophila melanogaster consists of phase delays during the late subjective evening , we expect short-period mutant phenotypes to require bright light after subjective dusk . Similarly , we expect long-period mutant phenotypes to require bright light in the early subjective morning to advance the cycle . Our results confirm this hypothesis . In Figure 2 , we demonstrate how bright light , admitted during the environmental night , resets the phase of short-period mutants such that it matches that of its environment . Given that the controlled system is 2 hours short , the occurrence of light during the night overlaps with the advance region of the system's PRC . Similarly , the onset of bright light at dawn overlaps with the delay region of long-period mutant PRCs ( Figure 2 ) . Our ability to maintain appropriate phase relationships between mutant phenotypes ( models characterized by non-nominal parameters ) and the environment ( the nominal case ) further proves the robustness of the algorithm despite model mismatch .
As implied by the PRC ( Figure 1 ) , a 3 hour phase difference may be recovered immediately through admission of light at CT15 . Hence , for open-loop control action to be most effective , environmental daylight should occur during the controlled system's subjective night ( at CT15 ) . In cases with small initial phase difference ( such that the subject's internal time is nearly equal to environmental time ) , however , daylight begins entrainment once the subjective day is around CT12 , by inducing small phase delays . This delay reduces the overlap between environmental daylight and the subjective night since re-entrainment of the initial phase difference began before subjective night . The opposite occurs with small negative phase differences , where an 18 hour ( or −6 hour ) phase difference may be recovered via a light pulse admitted at CT21 . In this case , environmental daylight affects the controlled system at the start of day while it has not yet begun entrainment , maximizing the phase advancing effect of light . For this reason , open-loop entrainment via phase advances requires less recovery time despite the fact that a single pulse of light may induce a greater phase delay than advance . More generally , we find that any given initial phase difference is more readily recovered if open-loop entrainment begins between CT0 and CT9; the rate of re-entrainment depends on the initial condition . To correct initial phase differences of 0 to 9 hours ( by inducing a phase delay ) , daylight is most effective at the end of the day , suggesting greater performance if the algorithm were to begin control action around CT6 . To correct initial phase differences of 0 to −6 hours ( by inducing phase advances ) , daylight is most effective at the start of the day , suggesting greater performance if the algorithm were to begin around CT0 . In the former case , daylight overlaps with the delay region of the subject's PRC , while in the latter it overlaps with the advance region . Resetting an initial condition of 12 to 15 hours , however , presents an interesting control dilemma as environmental daylight may induce both a phase delay and phase advance . For this reason , the open-loop control algorithm requires several days to correct for such phase differences . If light were accessible to entrain the system continuously throughout the day and night ( in other words , if we were to close the loop ) , phase recovery dynamics would be less extreme since phase resetting would rely less on the initial condition . Additional phase resetting properties may be inferred through investigation of the simulated PRCs . For instance , in the single input case , νs and ks exhibit similar recovery dynamics with the exception that νs is more effective at resetting initial phase differences of 15 to 21 hours . This quality may be associated with the fact that manipulating ks exhibits a strikingly similar phase response as νs where their input to output mapping is shifted by about 5 hours ( Figure 1 ) . This similarity may be attributed to the fact ks and νs are directly involved with the irreversible production , and transcriptional/translational regulation , of clock-specific genes/proteins . Additionally , the “active” region of the νs and νm PRCs are wider than those of ks and νd ( or , their dead zones are shorter than those of ks and νd ) , suggesting that their perturbation-induced phase shifts are accessible throughout a greater portion of the circadian day . Of the single control input results , the manipulation of νs , identified as the most sensitive parameter , provides the shortest phase recovery times . Despite these results , νd or light-based control is most efficient . In Figure 6 , we relate the cumulative control input ( a unitless measure that integrates the multiplicative control target action ) to the convergence of phase via the PER-TIM complex state error . The data shown reflects the recovery of an initial phase difference of 15 hours from IC12 . Analyzing this relationship may provide a basis from which the pharmaceutical industry might select one drug over another . If two different drug targets demonstrate similar response , the one that requires the least number of doses should be admitted , minimizing cost and the potential for drug related side-effects . Moreover , if the symptoms of illness are more severe than the potential for side effect , the drug that minimizes the state error may be preferred over others . The assessment of system convergence and the corresponding admitted control is key to the identification and application of control targets . In our modern “24/7” work world , social and commercial pressures often oppose our natural circadian timekeeping , causing a source of circadian stress that may lead to chronic illnesses such as cardiovascular disease and cancer [24] . Numerous studies seem to show the effect of circadian rhythms on processes such as cell proliferation and apoptosis that eventually lead to proper growth control [25]–[27] . For instance , components of the cell cycle that dictate the G1-S and G2-M transition phase have been associated with circadian transcriptional regulation [28] , [29] . Also in certain conditions , cancer can be a direct consequence of the absence of the circadian regulation [25] , [26] , [30] . A review of circadian related clinical disorders describes how mutations in some clock genes are associated with alcoholism , sleeping disorders , hypertension , and morbidity [24] , [31] . Most commonly , poor circadian regulation leads to advanced sleep phase syndrome , delayed sleep phase syndrome , non-24-hour sleep-wake syndrome , and irregular sleep-wake pattern [32] . In each of these cases , poor circadian phase resetting may be achieved through the systematic admission of controlled light pulses . Assuming we have access to drugs that specifically target circadian genes , we can identify the targets whose manipulation yields the most effective and immediate response through investigation of each control's phase dynamics ( as shown in Figure 1 ) . Or , it is possible to minimize the use of control and choose targets that require the least number of doses . We may also tailor the MPC algorithm to correct phase more readily through simultaneous manipulation of multiple control targets . Even further , we may reduce the computational expense by enumerating the control solutions over a grid in the solution space ( light magnitude as a function of time ) , and choosing the optimal control sequence via an exhaustive search . The algorithm approaches a globally optimal solution as the total possible quantization steps of the control input increases . We tested the efficacy of the algorithm with respect to a quantization of 2 , 4 , 8 , and 16 steps [12] . Results suggest that the shorter recovery time associated with the finer-grid enumeration may not outweigh the increase in computation time . Therefore , we may dramatically reduce computational expense by investigating control solutions for as few as 2 possible control values . Our methods show great promise for use in the pharmaceutical industry as our theoretical phase entrainment of mutant phenotypes demonstrates the robustness of the algorithm in the presence of model mismatch . This robustness alleviates concerns in the pharmaceutical industry to tailor mathematical representations of bio-chemical pathways to individual people . The study of controlled light pulses as a means of correcting phase is a common area of interest . Studies have shown that humans are much more sensitive to light than initially suspected since room light can significantly reset the phase of the human circadian clock [33] , [34] . Furthermore , the admission of morning light has been considered as an antidepressant by realigning the internal clock with the environment [35] . Additional studies suggest that the human circadian clock mechanism functions similarly to those of other mammals [34] . This similarity may be attributed to shape/amplitude characteristics of their respective phase response curves . Humans show phase-delay shifts of up to 3 . 6 hours and phase-advance shifts of up to 2 . 01 hours ( with respect to a 6 . 7 hour pulse of bright light ) [36] , which is both quantitatively and qualitatively similar to other mammalian species . This parallel motivates the experimental application of controlled light pulses for phase resetting in mammals . We have taken this first step by assessing the efficacy and computational utility of model predictive control as applied to a detailed 71-state Mus musculus circadian model [37] . Furthermore , melatonin has proven to be a key circadian phase resetting agent for totally blind people who cannot synchronize to environmental day∶night cycles ( or do so at an abnormal time ) [35] . Therefore , melatonin may be used individually ( in cases to treat the totally blind ) , or in combination with light to provide more effective phase resetting . Therapies designed to alleviate circadian load would have an important impact on morbidity and mortality across the developed world . Aside from correcting mutant phenotypes , phase resetting would increase performance in many healthy , or wild-type , cases such as frequent flyers avoiding jet-lag or astronauts maintaining a rigorous schedule during space exploration [17] . The real-time application of the proposed algorithm , however , may be a major issue; in practice , it will not be feasible to collect the corresponding protein concentration data at the molecular level . However , behavioral and/or physiological parameters that are controlled by ( and correlated with ) the circadian clock's dynamics are easily accessible . Such data may include actograms such as wheel running data for rodents [6] . Hence , a missing link in the current work concerns the development of corresponding ( non-linear ) state estimators for reconstructing the molecular dynamics . Given the discrete nature of MPC ( sampling every 4 hours ) , the proposed strategy is feasible in practice since sampling rates of such physiological circadian markers may be much higher .
Model predictive control [38] is used to increase the re-synchronization or entrainment rate of circadian oscillators through the systematic application of specified control inputs . The algorithm follows a sample and hold strategy , updating the prediction and control input every ts = 2 hours , where the discrete time index , such that a function g ( kts ) = g[k] . For simplicity , we refer to k as being equivalent to and ignore its rounding component . The manipulated control variable , u[k] , optimizes an open-loop performance objective on a time interval extending from the current time to the current time plus a prediction horizon of P = 48 hours , where . This horizon allows the algorithm to take control action at the current time in response to a forecasted error . The move horizon , M = 12 hours , limits the number of control inputs within the prediction horizon such that u[k] spans a time interval . Beyond hours of simulation , the predictive model defaults to u[k] = 1 . Future behaviors for a variety of control inputs are computed according to the mathematical model of the system [13] . The efficacy of the algorithm was evaluated with respect to a sample and hold time interval of 1 , 2 , and 3 hours ( reflecting a move horizon of 3 , 6 , and 9 hours , respectively ) . Although shorter light pulses offer a more dynamic manipulated variable profile , it shortens the move horizon and may reduce the utility of model predictive control . Conversely , a longer pulse may reduce the possible control profiles since extended exposure to light leads to arrhythmic behavior [39] . Thus , we set the sampling rate to 2 hours . The fitness function penalizes the normalized predicted state error between the reference and controlled trajectories , ē[k] , and the net control , ū[k] , over the prediction horizon . The system output used to evaluate circadian performance ( or , phase entrainment ) is the trajectory defined by the total period and timeless protein complex concentrations . This state error , e[k] , is normalized with respect to the nominal amplitude of oscillation while the time dependent control input , u[k] , is normalized with respect to the nominal set of values , 1:where the state dynamics r[k] characterize the nominal reference . Note that the vector e[k] is , while the matrix ū[k] is m×c ( and c denotes the number of control inputs ) . To avoid penalizing transient effects , the state error is weighted uniformly over the move horizon ( reflected in the first m diagonal values of the p×p matrix Q ) , and with increasing weight of slope 2 over the prediction horizon ( reflected in the p−m to p diagonal values of Q ) . The cost of applying a light input is weighted uniformly with a magnitude of 100 as reflected in the diagonal values of the m×m matrix R . We can afford to be conservative with the cost of control in the wild-type case , since we can ensure that the lack of control ( the open-loop algorithm ) will eventually entrain the system . The values contained in R will be re-evaluated when the algorithm is designed to entrain mutant phenotype models . The performance of an m-length control input is measured byOnly the first move of the lowest cost control sequence evaluated at time k , , is implemented . Therefore , the sequence of actually implemented control moves may differ significantly from the sequence of control moves calculated at a particular time step . This discrepancy disappears as the prediction and move horizons near infinity . Feedback is incorporated by using the next measurement to update the optimization problem . Once the controlled state trajectories converge to within 15% of the reference state trajectories , the system is considered to have recovered its phase in Tr = mink[|e[k]|∞≤0 . 15] hours . At this point , the algorithm defaults to no control since nominal light∶dark cycles will keep the system synchronized to the new environment . Optimization of the phase synchronizing control sequences is completed through use of a genetic algorithm [40]–[42] . Parametric sensitivity analysis quantifies the relative change of system behavior with respect to an isolated parametric perturbation . Parametric state sensitivity analysis assigns a value to each system parameter that defines how its perturbation affects state dynamics: . This tool is often used to identify the robustness and fragility tradeoffs of regulatory structures [3] , and may be tailored to evaluate specific output performance such as period , amplitude , or phase characteristics [2] . Assuming the model has n states and ρ parameters , the FIM is a ρ×ρ matrix describing how any two parametric perturbations might affect state dynamics . More notably , the diagonal values of the FIM describe how any single parameter may affect state dynamics . As a result , we sort the values of the FIM from greatest to least magnitude and choose the top three individual parameters ( reflected by the sorted diagonal values ) and top three pairs of parameters ( reflected by the sorted off-diagonals ) whose perturbations yield the greatest change in output . We further analyze the FIM via the singular value decomposition [43] . Assuming FIM = F , it may be decomposed as F = UΣVT , where Σ is an n by p diagonal matrix of non-negative singular values , σ , n is the number of states , and p is the total number of system parameters . Matrices U and V contain the eigenvectors of FFT and FTF , respectively . U , Σ , and V are ordered according to the magnitude of the singular values . Thus , the first column vector of U ( and V ) represents the output ( and input ) direction with largest amplification . The next most important direction is associated with the second column vector , and so forth . We determine the top three parameters associated with the three greatest input directions in ν1 and ν2 as ideal inputs for studying the multiple control input strategy .
|
The robust timing , or phase , of the circadian clock is critical in directing and synchronizing molecular , cellular , and organismal behaviors . The clock's failure to maintain precision and adaption is associated with sleeping disorders , depression , and cancer . To better study and control the timing of circadian rhythms , we make use of systems theoretic tools such as sensitivity analysis and model predictive control ( MPC ) . Sensitivity analysis is used to identify key driving mechanisms without having to fully understand or investigate the detailed mechanistic interconnections of the large complex circadian network . Contrary to intuition , sensitivity analysis of the circadian model highlights several non-photic control inputs ( such as transcriptional regulation ) that outperform light-based circadian phase resetting – light is known to accelerate protein degradation . Aside from targeting individual parameters as control inputs , our methods identify combinations of control targets that may further the efficiency of entrainment . We compare the phase resetting performance of our MPC algorithm among cases involving individual and multiple simultaneous control targets ( in wild-type simulations ) . We then tailor the algorithm to correct continuously the phase mismatch that occurs in short and long period mutant phenotypes . Through use of the presented tools , our algorithm is robust in the presence of model mismatch and outperforms the natural in silico sun-cycle–based phase recovery strategy by nearly 3-fold .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"computational",
"biology/systems",
"biology",
"computational",
"biology/transcriptional",
"regulation"
] |
2008
|
Circadian Phase Resetting via Single and Multiple Control Targets
|
Errors throughout gene expression are likely deleterious , hence genomes are under selection to ameliorate their consequences . Additional stop codons ( ASCs ) are in-frame nonsense ‘codons’ downstream of the primary stop which may be read by translational machinery should the primary stop have been accidentally read through . Prior evidence in several eukaryotes suggests that ASCs are selected to prevent potentially-deleterious consequences of read-through . We extend this evidence showing that enrichment of ASCs is common but not universal for single cell eukaryotes . By contrast , there is limited evidence as to whether the same is true in other taxa . Here , we provide the first systematic test of the hypothesis that ASCs act as a fail-safe mechanism in eubacteria , a group with high read-through rates . Contra to the predictions of the hypothesis we find: there is paucity , not enrichment , of ASCs downstream; substitutions that degrade stops are more frequent in-frame than out-of-frame in 3’ sequence; highly expressed genes are no more likely to have ASCs than lowly expressed genes; usage of the leakiest primary stop ( TGA ) in highly expressed genes does not predict ASC enrichment even compared to usage of non-leaky stops ( TAA ) in lowly expressed genes , beyond downstream codon +1 . Any effect at the codon immediately proximal to the primary stop can be accounted for by a preference for a T/U residue immediately following the stop , although if anything , TT- and TC- starting codons are preferred . We conclude that there is no compelling evidence for ASC selection in eubacteria . This presents an unusual case in which the same error could be solved by the same mechanism in eukaryotes and prokaryotes but is not . We discuss two possible explanations: that , owing to the absence of nonsense mediated decay , bacteria may solve read-through via gene truncation and in eukaryotes certain prion states cause raised read-through rates .
Errors throughout transcription , translation , and post-translational modification can , and do , happen all the time [1–5] . Whilst an invaluable source of novelty that drives evolution [6] , the majority of these errors are likely deleterious [1–3 , 6–8] . Genomes may therefore be under selection to mitigate their consequences . This has been supported by bioinformatic studies of stop codon usage in gene locations other than that of the canonical stop . For example , it has been suggested that adenine enrichment at the fourth coding sequence residue in bacterial genes may promote translation termination following a frameshift event at the initiating ATG that allows an out-of-frame stop codon to be read [9 , 10] . In 5’ leading regions , in-frame stop codons are enriched and postulated to rapidly terminate premature translations [11] ( i . e . those that occur before the ribosome reaches the recognised start codon of the mRNA ) . Selection on the primary stop codon is also thought to be error related [12–14] . Experimental evidence from bacterial studies suggest the three stops differ in their read-through rates [14–21] . Notably the least leaky of the three , TAA , is the preferred codon , especially in the most highly expressed genes [13] . In this study , we consider the hypothesis that additional stop codons ( ASCs ) occur after the primary stop codon as a fail-safe mechanism to minimise the costs of stop codon readthrough [22] . This question is important both as a means to address the importance of error-proofing to genome evolution but potentially also for optimal transgene design . Although ribosomes normally terminate translation at stop codons there is a chance that an ectopic amino acid is inserted , allowing translation to continue in the same frame for the generation of extended polypeptides [23 , 24] . The primary cause appears to be aberrant recognition by near-cognate tRNAs [25 , 26] or other tRNA species [27] . While read-through rates vary depending both on the stop codon and local sequence context [28] , read-through rates are typically orders of magnitude higher than the mutation rate [29–33] , rendering read-through a potentially significant fitness-modifying trait . While there may be beneficial consequences , such as increased proteome diversity [34] , the best evidence suggests that it is largely non-adaptive [8] . Selection for the least leaky stop in highly expressed genes [13] provides strong support for the notion that selection acts to reduce read-through rates as it is most commonly a deleterious error . Possible costs include energetic wastage owing to unnecessary translation [35] and creation of potentially toxic or sticky novel peptides . Resource wastage can be acute if the ribosome needs to be recovered , as can happen , for example , if it moves into a polyA tail as both RNA and protein can be targeted for destruction [36–38] . In theory , the presence of ASCs downstream may alleviate some of these costs by reducing the amount of additional amino acids added to erroneous polypeptide chains [39] and preventing polyA associated destruction . We herein refer to such a system as the ‘fail-safe’ hypothesis . The fail-safe hypothesis has been most thoroughly examined in eukaryotes , notably in yeast [39] , and two ciliate species which have reassigned their genetic code such that TGA is the only stop codon [40] . In yeasts , a statistical excess of UAA at the third codon downstream of TAA-terminating genes points towards a maintenance of ASCs by selection in a manner dependent on expression level [39] . This was corroborated in ciliates , where ASCs appear downstream of the primary stop more often than expected by chance given the base composition of 3’ regions [40] . Given that the excess is larger in ciliates than in yeast , it was proposed that ASCs are under variable selection intensity dependent on readthrough rate , which in turn may vary between species [40] . This , however , remains post hoc speculation . In bacteria tests of the fail-safe hypothesis are lacking . One study found tandem ASCs ( those which immediately follow the primary stop ) are over-represented , being seen in 7% of E . coli genes [41] . However , the experimentally estimated termination efficiency of tandem stops were below the expected rate and it was postulated that prima facie over-representation in the genome could be attributed instead to the preference for a tetranucleotide containing +4U , thought to improve the termination efficiency of the primary stop [41–45] . +4U in this context refers to the base immediately after the primary stop . A +4U base biases the first codon after the primary stop towards a second stop codon as all stops start T/U . More recently , one study has widened the investigation to ASCs in the following 5 in-frame codon positions . Such ASCs are reported in 8% of E . coli genes [13] , however , although this figure concords with the findings of Major and colleagues [41] , the authors do not comment on whether this is higher , lower or the same as expected given more codon positions are being considered . More generally , it is unknown whether ASC frequency downstream is higher than expected under a GC-controlled null in any eubacteria . Preliminary data weakly argue against the fail-safe hypothesis as there is no preference for UAA , UGA or UAG as an ASC downstream of the primary stop [13] . While , however , one might imagine selection that favours ASCs might also be strong enough to bias usage towards the strongest stop ( UAA ) , this is a second order effect compared with selection for any ASC in leaky genes . Differential leakiness of stop codons in eubacteria provides a foundation for testing the fail-safe hypothesis . While UAA [29] , UGA [30 , 31] , and UAG [29 , 32 , 33] are all subject to read-through , they do so to differing degrees . The mechanistic basis for this variation is thought to relate to the specificity and abundance of release factors . The stop codons are recognised by a class I release factor [46–49] , with their dissociation mediated by class II release factors following peptide release [50] . In bacterial lineages decoded according to translation table 11 ( TT11 ) , the class I release factors responsible are RF1 and RF2 . UAG is recognised by RF1 , UGA is recognised by RF2 , and UAA is recognised by both RF1 and RF2 [48 , 51 , 52] . It is thought that the ability of UAA to bind both RF1 and RF2 contributes to it being the least ‘leaky’ stop . No matter what the mechanism , the selection of ASCs is likely to be highest in UGA-terminating genes and weakest for UAA , all else being equal . In addition to termination efficiency , there are at least two other predictors of stop codon usage , GC pressure and expression level , when comparing across genes and genomes . While between genomes genomic GC is a strong predictor of UAA and UGA alone , UAG and UGA , with identical nucleotide contents , show dissimilar trends , UGA usage being positively correlated with genomic GC while UAG usage is uncorrelated [13 , 14 , 53] . This is conjectured to relate to co-evolution between RF1:RF2 ratios and GC content [14] . Within genomes it is considered that highly expressed genes should be under selection to employ UAA this being the least leaky . Indeed , while across bacteria UAA usage is well predicted by GC pressure , it is found to be enriched in highly expressed genes ( HEGs ) even in GC rich genomes [13 , 14] . The resistance to GC pressure in HEGs is consistent with the notion that the net effect of readthrough is a combined function of the per translation leakage rate and the number of translation events any given transcript is subject to . Here we provide the first systematic test of the fail-safe hypothesis applied to eubacteria . We interrogate the 3’ UTRs of a large sample of phylogenetically relatively independent bacterial species for enrichment of ASCs . In acknowledgment of prior studies , we control for GC pressure [13 , 14 , 53] . We ask whether we can detect ASCs at rates higher than expected given underlying nucleotide content , and whether 3’ UTR codon switches seen in closely related species are biased towards ASC deposition compared to null ( determined by out of frame rates ) . Further , we ask whether highly expressed genes have more ASCs and whether expression level and primary stop usage predicts ASC usage . The most extreme difference should be between highly expressed TGA ending genes , which should have strong ASC selection , and lowly expressed TAA ending genes in which fail-safe selection should be the weakest . We also ask if the presence of an ASC predicts the downstream presence of further ASCs and whether mollicutes employing only two stops under-employ the codon that isn’t a stop . The tests are , however , complicated by the fact that stop codon efficiency is also dictated by local genomic context [28] . Indeed , it has been observed that nucleotide substitution rate increases with downstream distance from the stop codon with no obvious plateau within the next six downstream ‘codons’ [12] , bringing attention to this region as a potential influencer of termination efficiency . Such regions may be directly involved in the formation of termination complexes that include the ribosome [45] . As noted , one downstream element thought to affect termination is the nucleotide at position +4 [41 , 42 , 54 , 55] . In eukaryotes , +4C is associated with an increase to ca . 3% readthrough in certain genomic contexts [55] , whereas +4U is highly preserved in all three domains of life and thought to reduce readthrough rate via improved cross-linking with RF2 [42] . This is problematic as it tends to increase the frequency of 3’ in-frame stops at the first downstream codon compared to the simplest null model . At a greater scale , at least a hexanucleotide sequence may affect termination efficiency [44 , 55 , 56] . Whilst this evidence was found in eukaryotes , it cannot be discounted that the local genomic context affecting readthrough rates in bacteria could extend beyond the fourth site nucleotide . Thus , we attempt to control for downstream motif preferences , in addition to GC content , in our assessment of whether ASCs are selected for error-control . We find that , in contrast to eukaryotes , the great majority of our evidence argues against the notion that 3’ ASCs are selectively favoured . We speculate as to why this might be .
A prediction of the fail-safe 3’ stop hypothesis is that stop codons should be enriched immediately after the primary stop . Thus , we assessed genomes for ASC enrichment through comparison against a null model where downstream 3’ codons are chosen according to dinucleotide content only . This was achieved by the simulation of 10 , 000 dinucleotide-controlled 3’ UTRs per genome , the calculated mean ASC frequencies being the ‘expected’ value and the Z-score being the deviation from this mean normalised to the standard deviation of the simulations . A positive Z-score is an instance where ASCs are overused compared to null . The null neutral expectation was that there is no difference between the ASC frequencies of the real genomes and simulated sequences hence 50:50 split of positive and negative Z-scores . We instead find there to be significant variation from this ratio when considering the UTR as a whole but , unexpectedly , with an excess of instances of under-usage of stops ( from codon position +1 to +6; 13/644 Z > 0 , p < 2 . 2 x 10−16 , two-tailed binomial test ) . The same under usage is seen at all sites when considered individually ( p < 2 . 2 x 10−16 for all positions , two-tailed binomial tests; 89/644 Z > 0 at position +1 , 56/644 Z > 0 at position +2 , 36/644 Z > 0 at position +3 , 35/644 Z > 0 at position +4 , 48/644 Z > 0 at position +5 , 40/644 Z > 0 at position +6 ) . All significant findings survive multi-test correction ( p < 0 . 05/6 ) . These results accord with what we see if we consider the proportion of genomes showing significant deviation compared to null ( |Z| > 1 . 96 ) . In this instance , the null expectation of the binomial test is no longer 50:50 , rather that 95% of genomes will not be significantly deviated and 5% will . There is a significant variation from this ratio when considering UTR en mass ( p < 2 . 2 x 10−16 for the whole UTR ( 553/644 genomes ) and at each position ( p < 2 . 2 x 10−16 at position +1 ( 177/644 genomes ) , p < 2 . 2 x 10−16 at position +2 ( 129/644 ) , p < 2 . 2 x 10−16 at position +3 ( 136/644 ) ;p < 2 . 2 x 10−16 at position +4 ( 113/644 ) , p < 2 . 2 x 10−16 at position +5 ( 92/644 ) , p = 3 . 1 x 10−12 at position +6 ( 77/644 ) , two-tailed binomial tests ) , all surviving multi-test correction ( p < 0 . 05/6 ) . Closer examination again indicates that significant enrichment ( one-tailed test , therefore we now use Z > 1 . 64 ) occurs less than expected by chance ( p < 1 . 6 x 10−13 for the whole UTR ( 1/644 ) , p = 2 . 8 x 10−10 at position +1 ( 4/644 ) , p = 1 . 6 x 10−13 at position +2 ( 1/644 ) , p = 4 . 5 x 10−15 at position +3 ( 0/644 ) , p = 4 . 5 x 10−15 at position +4 ( 0/644 ) , p = 1 . 6 x 10−13 at position +5 ( 1/644 ) , p = 4 . 5 x 10−15 at position +6 ( 0/644 ) , one-tailed binomial tests ) . Indeed , when we consider under-enrichment ( Z < -1 . 64 ) , we find more significant results than expected by chance ( p < 2 . 2 x 10−16 for whole UTR ( 570/644 ) , p < 2 . 2 x 10−16 at position +1 ( 230/644 ) , p < 2 . 2 x 10−16 at position +2 ( 206/644 ) , p < 2 . 2 x 10−16 at position +3 ( 204/644 ) , p < 2 . 2 x 10−16 at position +4 ( 176/644 ) , p < 2 . 2 x 10−16 at position +5 ( 135/644 ) , p = 2 . 9 x 10−12 at position +6 ( 119/644 ) , one-tailed binomial tests ) . These results provide no prima facie support for the fail-safe hypothesis and , if anything , argue for ASC avoidance . Is there anything peculiar about the genomes for which we find under usage of ASCs ? As all three stop codon variants are AT-rich by nature , they are more likely to appear in AT-rich genomes by chance . The fail-safe hypothesis therefore predicts selection to retain ASCs most strongly in GC-rich genomes , where a dearth of ASCs is expected in the absence of selection . Our results are contra to this prediction , as we find a significant negative correlation between Z-score and GC3 content ( p < 2 . 2 x 10−16 , ρ = -0 . 64 , Spearman’s rank correlation ) ( Fig 1 ) . This trend is consistent at all positions +1 to +6 ( S1 Fig ) with the magnitude of the gradient decreasing with 3’ distance ( S2 Fig ) . This result is repeated when considering raw ASC frequency instead of Z-score ( S3 Fig ) . Indeed , it appears that it is where ASCs are predicted to be most needed that they most under-employed . Above , we not only find no evidence for ASC enrichment but for ASC avoidance . Could this be because genomes specifically remove ASCs at a higher rate than chance ? Alternatively , perhaps switches from non-stop to stop occur at a lower rate than expected . We investigate both of these possibilities through analysing codon switches from stop to non-stop , and vice versa , in orthologous gene triplets . We employ 29 sets of triplet species ( a paired ingroup and an outgroup ) and consider the results en mass . For null expectations we employ the comparable rate ( stop->non-stop , non-stop->stop ) in the +1 reading frame of the 3’ domain . Considering all codons ( Table 1 ) regardless of position in our dataset of the orthologous genes , we find the frequency of in-frame codon switches from non-stop to stop in 3’ UTR codons to be no different to the same switch in out-of-frame codons of the same sequences ( p = 0 . 31 , χ2 = 1 . 0 , Chi2 test ) . Consistent with selection to avoid ASCs , switches from stop to non-stop occur significantly more often in-frame than out-of-frame ( p = 0 . 0024 , χ2 = 9 . 2 , Chi2 test ) . Hence not only are in-frame stops not deposited in 3’ UTRs more than chance , they are if anything avoided . Both of these results corroborate the findings of our initial binomial tests and argue strongly against the fail-safe hypothesis in bacteria . Considering each position individually tells a similar story . For the vast majority of codon switches at each position , there is no difference in switch rate between in-frame and out-of-frame codons ( S1 Table ) . Exceptions to this are found at position +4 , where switches from stop to non-stop are significantly more common in-frame than out-of-frame , and at position +5 , where switches from non-stop to stop are significantly less common in-frame than out-of-frame . Both results are consistent with rejection of the fail-safe hypothesis , however do not survive even generous Bonferroni correction ( p > 0 . 05/6 ) . At position +1 , switches from non-stop to stop are significantly more common in-frame than out-of-frame , though this is likely explained by selection for +4T . The above tests provide no support for the fail-safe hypothesis but consider genes equally , regardless of the primary stop codon and expression level . Selection for termination efficiency is thought to be highest in HEGs [13 , 57] under the assumption that the net effect of readthrough is a function of the number of translation events the transcripts of any given gene are subject to . If the fail-safe hypothesis of ASCs is true , we therefore expect ASC frequencies to be significantly higher in HEGs than LEGs . This , however , does not seem to be the case . Unlike what is seen in yeast [39] , there were no significant differences between the ASC frequencies of HEGs and LEGs at any position even before multi-test correction ( p = 0 . 95 for whole UTR , p > 0 . 05 for all positions , Wilcoxon signed-rank tests; S4 Fig ) , suggesting that either expression level has no influence over the negative effects of readthrough or ASCs do not significantly affect the ability of a transcript to avoid these consequences . This test is however limited by small genome sample size . Through manually adding enrichment of stop codons to our data we find that a ~35% increase in HEGs compared to LEGs is required to retrieve a signal . Hence , we can be confident that ASC frequencies in our HEGs dataset do not exceed those seen in LEGs by this margin . We cannot investigate codon switches in highly and lowly expressed gene groups , as the PaxDb database does not contain compatible data to match the ATGC data we used for this analysis . The HEG/LEG analysis , whilst also negative , does not allow for covariance between expression level and usage of different stop codons . Notably the least leaky stop ( TAA ) is also the preferred one in the highly expressed genes [13 , 14] , which has the potential to dampen any differences between HEGS and LEGs . Under the fail-safe hypothesis , we expect TGA-terminating HEGs ( high readthrough , high expression ) to have the strongest selection for ASCs and TAA-terminating LEGS ( low readthrough , low expression ) to have the weakest . However , we find no significant difference between these groups when considering the whole UTR ( p = 0 . 36 , Wilcoxon signed-rank test ) . Aside from position +1 , there is no significant difference between TGA-terminating HEGs and TAA-terminating LEGs at a single position scale ( p = 0 . 060 for position +2 , p = 1 for position +3 , p = 0 . 83 for position +4 , p = 0 . 60 for position +5 , p = 0 . 62 for position +6 , Wilcoxon signed-rank tests ) . Even at position +1 the enrichment of ASC in the TAG/HEG class is a barely significant trend ( p = 0 . 041 , Wilcoxon signed-rank test ) that does not survive Bonferroni correction ( p > 0 . 05/6 ) ( Fig 2 ) . We thus find no evidence to support the notion of ASC selection , apart from a possible very weak effect at position +1 . If the above exceptional result at position +1 is owing to selection we might also expect the enrichment to be seen in other TAG and TAA highly expressed expressed genes . If , alternatively , it is a motif preference associated with TGA , we might expect it to be seen in lowly expressed TGA terminating genes but not necessarily elsewhere . To examine these possibilities we consider all combinations of expression level and primary stops in the assessment of ASC frequency ( Fig 3 ) . Considering the whole UTR ( +1 to +6 ) we find evidence for heterogeneity when considering all genes regardless of expression level ( p = 0 . 01 , χ = 8 . 79 , Kruskal-Wallis ) . However , if we remove position +1 from this analysis , significant heterogeneity cannot be recovered ( p = 0 . 57 , χ = 1 . 12 , Kruskal-Wallis ) . Indeed , we find that ASC enrichment is particular to position +1 and a peculiarity of TGA terminating genes weakly seen at all expression levels . We established this by first testing for heterogeneity between ASC usage dependent on the primary stop at position +1 . When considering all genes ( p = 1 . 9 x 10−15 , χ = 67 . 81 , Kruskal-Wallis ) and LEGs ( p = 0 . 032 , χ = 6 . 91 , Kruskal-Wallis ) we see evidence for such heterogeneity . For HEGs ASC usage is highest for TGA terminating genes but not significantly so ( p = 0 . 14 , χ = 3 . 97 , Kruskal-Wallis ) . Similarly the significance at position +1 in LEGs does not survive Bonferroni correction ( p < 0 . 05/6 ) . With some evidence for heterogeneity , we proceed to post-hoc Wilcoxon signed-rank tests for the two significant cases these indicating in each case , enrichment is highest in TGA-terminating genes ( position +1 all genes: TGA > TAA , p < 2 . 2 x 10−16; TGA > TAG , p < 2 . 2 x 10−16; position +1 LEGs: TGA > TAA , p = 3 . 0 x 10−3; TGA > TAG , p = 1 . 3 x 10−3 , Wilcoxon signed-rank tests ) . That we do not find significant deviation between primary stops at position +1 in HEGs is surprising , however likely comes as a direct consequence of small sample size . Confirming the lack of signal outside of position 1 , for all such positions , there is no significant difference in any expression group ( p > 0 . 05 , Kruskal-Wallis ) , with one exception , this being significant enrichment at position +2 in HEGs ( p = 0 . 029 , χ = 7 . 09 , Kruskal-Wallis ) . Here too the effect is most pronounced for TGA terminating genes ( position +2 HEGs: TGA > TAA , p = 6 . 9 x 10−3; TGA > TAG , p = 0 . 027 , Wilcoxon signed-rank tests ) , but neither the original test nor the subtest survive Bonferonni correction . Above we have shown that TGA-terminating genes are commonly immediately followed by ASCs . There are two hypotheses for this: ( i ) a general enrichment of thymine at the fourth coding residue that enables more effective termination [13 , 41 , 42] , most especially true for TGA due to its unique recognition by RF2 alone , and ( ii ) an enrichment of ASCs in response to TGA leakiness . Several lines of evidence argue in favour of the former . First , we sought to establish whether there was general +4T enrichment . To this end we calculated the frequency of T-starting codons at position +1 and compared it to the average T-starting codon frequency from positions +1 to +6 . T-starting codons at position +1 were found to be enriched compared to other downstream positions ( p < 2 . 2 x 10−16 , Wilcoxon signed-rank test ) . However , this is not necessarily attributable to the presence of position +1 ASCs . In repeating the same methodology , we find the frequency of all non-stop T-starting codons to be significantly enriched at position +1 compared to the UTR average in genes that don’t have a position +1 ASC ( p < 2 . 2 x 10−16 , Wilcoxon signed-rank test ) . This effect is most heavily influenced by TGA-terminating genes , in which T-starting non-stop codons are more enriched at position +1 compared to the UTR average ( p < 2 . 2 x 10−16 , Wilcoxon signed-rank test ) than seen in TAA-terminating genes ( p = 4 . 7 x 10−14 , Wilcoxon signed-rank test ) and TAG-terminating ( p = 0 . 9951 , Wilcoxon signed-rank test ) genes . Second , we find ASC frequencies at position +1 in HEGs and LEGs are not significantly different ( p = 0 . 66 , Wilcoxon signed-rank test ) . In absolute terms the enrichment in LEGS is if anything higher . This is contra to the fail-safe prediction that ASCs should be most greatly enriched in HEGs . Third , if the effect is owing to translation termination signals favouring +4T , then +4T enrichment might be expected to be most profound in TGA terminating genes and weakest in TAG terminating genes as RF2 crosslinking [43 , 44] would be irrelevant for RF1-recruiting TAG . As TAA can use RF2 or RF1 it should be intermediate . To investigate this , we analysed the relative usage of thymine against adenine , cytosine , and guanine at the fourth site as a function of primary stop usage ( Fig 4 ) . Considering all genes this not only confirmed T enrichment compared to the next most frequent nucleotide , unique to TGA-terminating genes ( T > A: p < 2 . 2 x 10−16 , Wilcoxon signed-rank test ) but , consistent with the RF2 crosslinking hypothesis , the +4T usage was in the order TGA>TAA>TAG . +4T frequency is significantly higher in TGA-terminating genes than TAG-terminating genes ( p < 2 . 2 x 10−16 , Wilcoxon signed-rank test ) and TAA-terminating genes than TAG-terminating genes ( p = 7 . 5 x 10−5 , Wilcoxon signed-rank test ) . The strength of +4T enrichment in TGA and weakness in TAG-terminating genes is underscored when we consider HEGs and LEGs separately . Thymine frequency at the fourth site significantly exceeded the next highest nucleotide regardless of the primary stop in HEGs , in the predicted order ( T > A , p = 2 . 9 x 10−4 in TGA-terminating genes; p = 0 . 013 in TAA-terminating genes; p = 0 . 045 in TAG-terminating genes , Wilcoxon signed-rank tests ) . The signal in TAG-terminating genes in this instance does not withstand multi-test correction ( p > 0 . 05/3 ) . In LEGs , too , raw +4T frequency is found in the expected order TGA>TAA>TAG , with enriched frequencies of thymine evident only in TGA-terminating genes ( T > A: p = 1 . 2 x 10−4 , Wilcoxon signed-rank test ) . The above results suggest that any weak stop excess at codon position +1 is not owing to selection for stops per se . Is the enrichment for T-starting codons the same for all such codons , stops included , or might some classes be especially preferred , suggesting some further motif structures ? To investigate this , we calculated an enrichment score for each T-starting codon ( Fig 5 ) . We notice an enrichment of TC and TT-starting codons at position +1 , particularly in HEGs and TGA-terminating genes . Indeed , we propose that there may be a fifth nucleotide site preference for thymine or cytosine in +4T-containing genes as part of a wider motif beneficial for translation termination . Consistent with this , the enrichment of stop codons at position +1 is unremarkable compared to other T-starting codons . This is , too , consistent with our +4T-controlled simulation experiment ( S5 Fig ) , which finds that increased ASC frequencies at position +1 are the direct consequence of +4T enrichment . Further analysis suggests that TT is preferred in HEGs regardless of the primary stop . This partially reflects an AT bias in our genome set and more generally the preference for TT in AT rich genomes and TC in GC rich ones ( S6 Fig ) . As stops appear to prefer a +4T to enable stop codon recognition , we can ask whether this is also true of ASCs . We thus test the null that ASCs are as likely to have a downstream T as primary stops . For all genes , ASCs have significantly less chance to be immediately followed by a T than do primary stops ( p < 2 . 2 x 10−16 , Wilcoxon signed-rank test ) . The same is seen in HEGs ( p = 2 . 4 x 10−6 , Wilcoxon signed-rank test ) , though not LEGs ( p = 0 . 078 , Wilcoxon signed-rank test ) . The fail-safe hypothesis however does not necessarily predict selection termination functionality at ASCs to match that of primary stops . A more generous null is to ask whether ASCs have more T at the +4 site than do non-ASC codons in the 3’ region . We actually find for all genes that ASCs have lower chance of having this ( p < 1 . 5 x 10−12 , Wilcoxon signed-rank test ) . The same is seen in both HEGs ( p = 7 . 0 x 10−3 , Wilcoxon signed-rank test ) and LEGs ( p = 0 . 027 , Wilcoxon signed-rank test ) . A more specific approach assesses each stop codon variant individually . As +4T enrichment appears to be peculiar to TGA-terminating genes , we expect TGAT to be more common as an ASC than TAAT , and even more so compared to TAGT . All three stop variants are significantly less likely to be followed by T when in-frame downstream than when located at the primary stop site ( TAA p < 2 . 2 x 10−16; TGA p < 2 . 2 x 10−16; TAG p < 4 . 6 x 10−6 , Wilcoxon signed-rank tests ) . Though whilst TAA ( p < 2 . 2 x 10−16 , Wilcoxon signed-rank test ) and TAG ( p < 2 . 5 x 10−12 , Wilcoxon signed-rank test ) are less likely to possess a 3’ neighbouring T than non-ASC codons , TGA is significantly more likely to ( p < 2 . 2 x 10−16 , Wilcoxon signed-rank test ) . Hence there is exceptionalism of TGA which falls in line with the expectations of the fail-safe hypothesis . Indeed , ASC +4T frequencies are found in the expected pattern TGA > TAA > TAG ( TGA > TAA p < 2 . 2 x 10−16; TGA > TAG p < 2 . 2 x 10−16; TAA > TAG 6 . 0 x 10−5 , Wilcoxon signed-rank tests ) . We do , however , find a contradictory result in that TGAT is no more common in HEGs than LEGs ( p = 0 . 56 , Wilcoxon signed-rank test ) , though this is affected by low genome sample sizes . One might suggest that the enrichment of T following TGA in 3’ positions compared to other non-stop codons could be attributed to dinucleotide preference . We control for this by comparing 3’ TGA to non-stop codons with third nucleotide A , finding again that TGAT to be significantly more common ( p = 2 . 9 x 10−5 , Wilcoxon signed-rank test ) . The above analyses provide little support for the fail-safe hypothesis as any weak site +1 trends appear better explained by +4T motif presence . The observation of ASC enrichment at codon +2 in TGA terminating HEGs ( sensitive to Bonferroni correction ) and the enrichment of 3’ TGAT are the only results that doesn’t obviously fit with this otherwise profound rejection of the hypothesis . Given this , and the difficulties allowing for complex GC pressure and motif issues , we consider alternative tests . In theory , if ASCs function in the termination of translation , it is unlikely that an ASC will be followed by another . The combined action of the primary stop and the ASC should terminate translation such that net readthrough rates are negligible and there is no selection for a third stop . We thus test the null hypothesis that ASC-containing genes , where the stop codon lies before ( and including ) codon +N , have an equal chance of possessing a further ASC downstream . We compare downstream ASC frequencies of ASC-containing and ASC-absent genes and see no evidence that possession of a stop predicts low rates of downstream stops ( p = 0 . 83 where the focal codon is position +1 , p = 0 . 76 for position +2 , p = 0 . 77 for position +3 , p = 0 . 78 for position +4 and p = 0 . 92 for position +5 , one-tailed Wilcoxon signed-rank tests ) . This provides no support for the fail-safe hypothesis . We can ask a more detailed question , namely whether having a stop at position N predicts the absence of a stop at the next codon position ( +N+1 , rather than generically downstream ) . In this case ‘N’ refers to each position from +1 to +5 ( position +6 could not be tested in this instance as this would require analysis of ASCs at position +7 , which is not considered ) . Where we consider all genes , ASC-absent genes demonstrate no significant excess of ASCs at position +N+1 over ASC-containing genes at all positions ( p > 0 . 05 ) , except where the focal codon position was position +1 ( p = 3 . 6 x 10−3 , Wilcoxon signed-rank test; Fig 6 ) . Were this owing to selection , we expect to find a stronger signal in HEGs than in LEGS . However , there is no significant difference between HEGs and LEGs at any position ( p > 0 . 05 for all positions +1 to +5 , Wilcoxon signed-rank tests ) . A significant signal can only be found in HEGs at position +2 ( p = 0 . 027 , Wilcoxon signed-rank test ) , although this result does not survive multi-test correction ( p > 0 . 05/5 ) . We do however notice that the magnitude of the effect is actually greater in LEGs , which is contra to the fail-safe hypothesis . We conclude that these tests provide no robust evidence that the presence of a stop codon predicts the presence/absence of further stops and if any such effects exist they are specific to the domain in the immediate vicinity of the primary stop , suggesting that hidden motifs might be a viable alternative explanation . The mollicute bacteria provide for a “natural” experiment as some genomes employ TT4 in which only TAA and TAG are used for chain termination . Hence , as TGA functions as a stop codon in TT11 genomes , it is expected under the fail-safe hypothesis that TGA frequency 3’ of the primary stop in TT4 genomes should be consistently lower than that in TT11 genomes . We tested this hypothesis by fitting a LOESS model ( span = 2/3 ) for positions +1 to +6 usage of TGA against genomic GC3 in TT11 genomes of the full genome set . These models allowed the prediction of TGA frequencies of TT4 mollicute genomes at each position given their genomic GC3 content . TGA frequency was significantly reduced in TT4 genomes compared to predicted by the LOESS model at positions +3 and +5 ( p = 1 . 5 x 10−1 for position +1; p = 9 . 8 x 10−2 for position +2; p = 9 . 7 x 10−4 for position +3; p = 7 . 9 x 10−1 for position +4; p = 3 . 4 x 10−4 for position +5; p = 7 . 7 x 10−2 for position +6 , Wilcoxon signed-rank tests ) . For comparison , TT11 mollicute genomes do not significantly under-use TGA at any position ( p > 0 . 05 , Wilcoxon signed-rank tests ) . In TT4 genomes , lack of underrepresentation at position +1 possibly accords with the utility of +4T and similar motifs adjacent to the primary stop . The poverty of TGA at positions +3 and +5 survives multi-test correction and is consistent with the possibility that TGA maintains a function in TT11 genomes beyond its role in TT4 genomes . Why TGA is not underused at positions +2 , +4 and +6 is unexplained . We do , however , find that when considering the whole UTR ( positions +1 to +6 ) TGA is used significantly less often in TT4 genomes than predicted ( p = 3 . 8 x 10−6 , Wilcoxon signed-rank test ) . We acknowledge the limitations of LOESS modelling , which include those relating to the arbitrary nature of kernel/span function , and therefore validate this result with a different test design ( S7 Fig ) . Given the above we also asked whether TAA , TGA , and TAG codon switches occur at different rates in TT4 genomes . We find no significant differences ( S2 Table ) but strongly caution that the results are limited by drastically reduced gene sample size . The above results are consistent with the hypothesis that TGA is underused in 3’ domains when it isn’t employed as a stop codon , compared with its usage in genomes of similar GC content when it can function as a stop . However , if TGA is underrepresented in TT4 decoded genomes due to its selection for error-proofing in TT11 decoded species , we expect the magnitude of this under-enrichment to consistently surpass all other codons . We thus investigated all 64 codons using the same LOESS methodology and ranked them by their one-tailed Wilcoxon signed-rank test p-value ( S3 Table ) . We find TGA to be just the 25th most under-enriched codon at position +1 , 20th at position +2 , 4th at position +3 , 49th at position +4 , 2nd at position +5 , and 16th at position +6 . Instead , we find codons CCG ( 1st at positions +1 , +4 , +6 ) , GTG ( 2nd at position +1 , 3rd at positions +4 and +6 ) , and TAT ( 1st at position +2 , 2nd at position +3 , 4th at position +1 ) among the more commonly underrepresented codons at specific positions . Assertions that there is something special about TGA , specifically relating to translational termination , therefore remains speculative . The disconnect between TAG and TGA usage as a primary stop has been attributed to co-evolution between RF1:RF2 ratios and GC content [14] . If true , this renders stop usage tightly coupled to the mechanistic basis of translational termination . Are then these trends in TAA , TGA and TAG usage also seen downstream ? First , we analysed the relative usage of TAA , TGA and TAG at the primary site so as to repeat the findings of Korkmaz and colleagues ( 2014 ) with our genome set ( S8 Fig ) . As expected we find that TAA-usage is negatively correlated with genomic GC3 ( ρ = -0 . 92 , p < 2 . 2 x 10−16 , Spearman’s rank correlation ) , TGA-usage is positively correlated with genomic GC3 ( ρ = 0 . 88 , p < 2 . 2 x 10−16 , Spearman’s rank correlation ) , and TAG-usage shows no significant correlation and remains at low levels regardless of genomic GC3 ( ρ = -0 . 017 , p = 0 . 663 , Spearman’s rank correlation ) . We then returned our focus to downstream . Surprisingly , we find that trends in TGA and TAG usage remains clearly decoupled despite their equal GC content . Indeed , trends in stop codon usage are remarkably similar between positions +0 to +6 ( S4 Table ) . That stop codon usage at the primary stop consistent in 3’ positions implies either a ) that the release factor hypothesis [22] regarding the decoupled usage of TGA and TAG usage is wrong or b ) ASCs are , despite all the other negative data , under selection as fail-safe codons . We can investigate this by considering all three reading frames: should the relative codon usage of ASCs remain consistent in +1 and +2 frame-shift environments we can be relatively confident that usage is not controlled by selection relating to translational readthrough or termination . This is exactly what we find ( Fig 7 ) , and this is consistent with the bulk of the evidence described in our study . Thus , we suggest that the RF1:RF2 ratio is not the correct explanation for the differential stop usage as a function of GC and we are instead missing some important information regarding TGA and TAG usage . The above bacterial evidence against ASC enrichment is in contrast to that seen in yeast and ciliates [39 , 40] . Do prokaryotes and eukaryotes truly differ in their propensity to use ASCs to control translational read-through rates ? Alternatively , might there be a reporting bias in which only significant effects surface in the published literature , thereby giving a skewed view of the commonality of fail-safe stops ? Additionally , there are several ways to evaluate the fail-safe hypothesis and it could be that our methods would fail to report effects in the eukaryotic species within which ASC enrichment has been observed . For example , while we employ a dinucleotide control , Adachi and Cavalcanti in the prior ciliate analysis [40] employ a method that considers the rate of occurrence of the first 3’ stop as a function of downstream position given an underlying rate at which stops are observed in 3’ UTR . To ask whether our method would recover enrichment where previously claimed , we consider ASC enrichment in T . thermophila , P . tetraurelia and S . cerevisiae via the calculation of Z-scores , i . e . using the same method described earlier ( Fig 8 ) . Significant enrichment ( Z > 1 . 64 , for one tailed test of enrichment ) is detectable using whole 3’ UTR frequencies in the two ciliates but not in yeast . The latter negative result is not surprising as , unlike in ciliates , yeast enrichment is only detectable at position +3 and predominantly only when the primary stop is TAA [39] . Indeed , we find that position +3 is unusual in being enriched ( Z>0 ) in ASCs in all genes and in TAA-terminating yeast genes ( Fig 8 ) , although in neither is the effect significant ( Z = 0 . 93 for all genes , Z = 0 . 70 for TAA-terminating genes ) . These results suggest that our method can capture some but not all of the prior claims . Nevertheless , we extend the Z-score analysis to a set of 68 single-celled eukaryotes to investigate whether the spread of Z-scores matches that of the bacteria . We propose that single-celled eukaryotes are the fairest comparators to eubacteria as they are likely to both have large effective population sizes and , being single celled , would suffer the immediate consequences of any fitness costs of read-through . Multi-cellular organisms , by contrast , might be able to buffer fitness loss in one cell , for example by apoptosis and cell replacement . A genome is considered ‘enriched’ if it contains significant ASC enrichment at one or more positions ( Z > 2 . 33 , Bonferroni corrected one tailed ) . Interestingly , we find 20/68 of our eukaryotic genomes to be enriched , compared to 0/644 of our bacteria , these proportions being significantly different ( p < 0 . 0001 , χ2 = 184 . 3 , Chi2 test ) . An alternative metric is to consider the number of genomes showing enrichment , defined by chi-squared , above dinucleotide controlled null frequencies at each position . For this we employ a Chi2 p-value < 0 . 05/n , where n is the number of positions tested , and apply this to our TT11 bacterial set of genomes and our set of 68 single-celled eukaryotes . Having defined positional ASC enrichment as p < 0 . 01 ( 0 . 05/5 ) as we analyse five positions ( +2 to +6 ) , the probability of a genome not possessing significant ASC enrichment at one or more positions is 0 . 995 ( approximately 0 . 951 ) . There is hence a 1–0 . 995 ( approximately 0 . 049 ) probability that a genome will contain significant enrichment at one or more positions . Hence , our null is that 4 . 9% of our genomes are expected to show ASC enrichment by chance alone . In our eukaryotic set , we find over-representation of genomes containing significant ASC enrichment compared to this null prediction ( 21/68 , p = 6 . 12 x 10−12 , one tailed-binomial test with p = 0 . 049 , expected = 3 ) . Such a result supports evidence for ASC enrichment in eukaryotic systems [39 , 40] , however we note that whilst ASC enrichment is commonplace , it is not universal nor consistent in its position . By contrast in bacteria , using this same method , we find that significantly fewer bacterial genomes show enrichment than expected by null ( 21/644 , p = 0 . 028 , one-tailed binomial test with p = 0 . 049 , expected = 32 , Fig 9 ) , consistent with a broad claim that eubacteria seem to avoid ASCs . Moreover , the observed proportions of 21/644 in bacteria and 21/68 in eukaryotes are significantly different ( p < 0 . 0001 , χ2 = 79 . 6 , Chi2 test ) , corroborating the results of our Z-score analysis . We also repeat the Chi2 comparison using an alternative null model as proposed by Adachi and Cavalcanti [40] . This too confirms the same results ( Fig 10 ) , namely avoidance of ASCs in bacteria , enrichment in single-celled eukaryotes . Indeed , this mode of analysis reports enrichment at one or more positions in 32 of 68 eukaryote genomes and only 7 of 644 bacterial genomes , these proportions being different ( χ2 = 242 . 3 , p < 0 . 0001 , Chi2 test ) . The conclusions that there is indeed a discrepancy between bacterial and eukaryotic propensity to select for ASCs is hence both real and largely resilient to methodological nuance . With respect to the eukaryotes , we corroborate significant ASC enrichment ( using at least one methodology ) in the previously analysed yeast [39] ( S . cerevisiae , plus C . albicans ) and ciliates [40] ( P . tetraurelia , T . thermophila ) . We note that the two ciliate species analysed in the prior study [40] possess a re-assigned translation table ( TGA is the only stop ) . We not only recover ASC enrichment in these re-assigned ciliates ( plus S . lemnae ) , but a translation table 11 ( TGA , TAA and TAG are all stops ) ciliate as well ( S . coeruleus ) . Of our methodologies , the two dinucleotide-controlled analyses ( Z-score: 20 enriched genomes , S7 Table; Chi2 analysis: 21 enriched genomes , S8 Table ) appear to be the most stringent in detecting eukaryotic ASC enrichment . Identification of enrichment using the Adachi and Cavalcanti null model [40] is more generous ( 32 enriched genomes , S9 Table ) . We do , however , note that ASC enrichment at one or more positions is recovered by all three methods in 15 eukaryotic genomes , indicating reasonable overlap between the tests .
Our results suggest that , unlike in yeast , ciliates and some other protists , the error-proofing role of ASCs in bacteria is minimal at best . We began by testing the most obvious prediction of the fail-safe hypothesis , that stop codons should be enriched downstream of the primary stop codon . Having found no evidence for this at a genome-wide level , we considered the conservation of ASCs and found evidence that stops are less preserved than expected , this too being consistent with apparent avoidance . Additionally , we compared highly expressed and lowly expressed genes , seeing no differences . Comparing TGA-terminating HEGs and TAA-terminating LEGs we found TGA-terminating HEGs do not contain significantly higher ASC frequencies , except at position +1 . The effect seen at +1 is not the result of selection for stops , but rather a knock-on consequence of selection for T-starting codons at the first codon downstream , the trend being seen for non-stop T-starting codons too . Indeed , in the context of other T-starting codons stop codon usage is not simply unremarkable , the trend seems to be enrichment for non-stops , TT and TC being preferred residues . While it is suggestive that the leakiest codon ( TGA ) is the one associated with ASC enrichment at site +1 , this trend is better explained by reference to the notion that RF2 cross-links with the adjacent +4T and TGA uses only RF2 . Perhaps an informative test would compare species with defective/absent RF2 to those without , however we find no such genomes in our genome set . These results suggest bacteria and eukaryotes are different in the usage of fail-safe stops . Using several alternative methodologies to compare ASC enrichment in bacteria to protists , we validate that ASC enrichment is found in single celled eukaryotes more often than in bacteria . Our findings therefore highlight a discrepancy in the way that bacterial and eukaryotic genomes evolve in response to translational read-through . With respect to bacterial transgenes , our results thus do not support any major adjustments to their design or experimental protocols , beyond using TAA or TAAT[T/C] for termination . A few results were consistent with the fail-safe hypothesis but not overwhelmingly so . While having a stop codon at any given position doesn’t predict a dearth of downstream stops , if there is a stop at position +1 there is less likely to be one at position +2 . However , the magnitude of this effect is greater in LEGs than HEGs questioning the overall relevance of this to the fail-safe hypothesis . Given too that the effects are seen exclusively in proximity to the primary stop , selection on unrecognised motifs is a viable and probably better alternative explanation . That TT4 mollicutes contain fewer TGAs in their 3’ domain than expected is also enigmatic . That TT4 mollicutes contain less 3’ UTR TGA than TT11 genomes ( after control for GC content ) is consistent with selection impacting TGA levels in 3’ domains of TT11-decoded species . However , in TT11 genomes we see no evidence for ASCs beyond null levels and indeed , prima facie they seem to be avoided more often than enriched compared to GC controlled nulls . Furthermore , that some sense codons are even more consistently under-used at 3’ UTR sites , for reasons that are unknown , suggests that there is a gap in our knowledge of the biology of these 3’ ends . A third possibly consistent result is that ASC usage of the three stops as a function of GC content matches that of the primary stop . The patterns for the primary stop were speculated to reflect co-evolution between GC content and RF1:RF2 ratios [14] but this remains to be verified . That we see the same broad trends at all downstream positions , despite all the other evidence against these ASCs being functional stops , we suggest more profoundly questions the RF1:RF2 ratio model than it supports ASC functionality . In accord with the under usage of TGA and other codons in TT4 genomes , perhaps more complicated dinucleotide or trinucleotide preferences should be considered . This leaves one outstanding observation , namely that 3’ TGA tend to be followed by T more than expected , even given the rate of T-starting codons with an A immediately prior . How can we explain this ? We suggest a hypothesis that might explain the curious observations against a backdrop of a large body of negative evidence . First , we wish to discount the possibility that the lack of evidence for selection on ASCs relates to read-through not being a strong enough selective force . Experimental estimates in E . coli and S . typhimurium suggest that the read-through rates are really very high . A read-through event at a TAA-terminating site can occur at frequencies between >1 x 10−5–9 x 10−4 [29] , and at a TAG between 1 . 1 x 10−4–7 x 10−3 [28 , 29 , 32 , 33] . If ASCs do meaningfully function in chain termination , one would have expected to find a signal in TGA-terminating genes , where readthrough may occur at rates of 1 in 1000 translation events up to 1 in 100 [15 , 30 , 31] . Thus , numbers suggest a potentially high rate of readthrough . Second , it is most likely because of this that stop codons are themselves subjected to selection for efficient termination . This is probably why TGA-terminating genes are rarely highly expressed–where such selection is expected to be strongest and TAA is over-represented in the set of highly expressed genes even in GC rich genomes [13] . Consistent with this , Belinky and colleagues ( 2018 ) found that stop codon switches occur significantly more frequently than the equivalent substitutions in non-coding DNA . Given this we assume that selection against read-through is a significant force . We can then question whether , if read-through is the problem , ASCs are the expected solution in bacteria . Evidence from stop codon usage , especially in highly expressed genes suggests that there is selection for TAA enrichment as the stop . We could presume that in many cases this means simply a non-TAA stop mutates to be TAA and is selectively favoured , especially if the gene is highly expressed . However , there are other possibilities . For example , imagine that we have a highly expressed gene using TGA and so possesses high read-through rates . Imagine too that upstream are sense codons which could mutate in one step to TAA or indeed any stop . This would introduce a premature stop ( assuming the context is otherwise fine ) with , importantly , a guaranteed fail-safe stop downstream i . e . the original primary stop . There would be a benefit from lower net read-through rates ( we presume nearly all genes will terminate at or before the second , original , stop ) and a benefit from reduced translation costs when the new earlier stop functions . Moreover , the sequence now immediately 3’ of the new stop will , if read-through happens , be sense codons of a recently functional protein , so there should be no toxicity of this additional sequence . All of these benefits suggest this is a viable path for evolution , the major cost owing to reduced gene length affecting protein function . However , tolerating such a cost appears to be possible , with stop codon shifts in 5’ directions now thought to have an under-appreciated influence on gene shortening [58] . If the net benefits of reduction in read-through is greater than this cost then the system will have evolved towards reduced net readthrough . Could this also explain why we detect no enrichment of stop codons in the 3’ domain as , until the first ASC , selection would have recently been on this to perform as coding sequence ? Might this explain the apparent general rarity of downstream T following an ASC ? A stop lacking the +4T would be especially leaky and so especially favour rescue by creation of an earlier stop . The one exception could be TGAT . If this , like TGA , remains relatively leaky ( unlike TAAT ) then selection could still favour 3’ stop creation . Might this also go some way to explaining the mollicutes result ? If TGA wasn’t a stop there is no reason it would by necessity feature in the 3’ domain as the abandoned stop and so might appear at low frequency in the mollicutes . An alternative trajectory to rescue a leaky TGA would be for TGA to mutate but to a sense codon . This could be favoured if the run on then meets a less leaky stop codon shortly downstream . The shortening process we suggest would be more common than the lengthening for several reasons . First , especially in highly expressed genes , addition of amino acids is likely to be costly , whereas loss would come with an energetic saving . Second , in the shortening process there are multiple potential sites that could mutate to a new upstream stop , while in the latter the mutation is required at the stop codon . Third in the gene shortening mode , at the time of mutation , at least one downstream site will be an ASC ( the old primary stop ) , thus the system comes with guaranteed ASC protection . By contrast , gene extension could replace a leaky stop with , at best a less leaky stop , but no guaranteed fail-safe ASC . Fourth , there is no guarantee on extension that the extension isn’t toxic , while for read-through after shortening this would not be an issue . Thus , we suggest there may be a process to shorten highly expressed genes to enable evolution of protection from read-through that might be particular to prokaryotes . The difficulty with this model seems to be that the rate at which this would need to occur might have to be rather high . Whether this predicts any pattern is unclear as genes cannot continue to shorten indefinitely . Might a propensity to gene shortening as a mechanism to cope with read-through also explain why ASC enrichment isn’t seen in bacteria but is in eukaryotes ? In eukaryotes the mutation creating this new upstream stop could be trapped by eukaryote-specific nonsense mediated decay ( NMD ) making gene shortening a non-viable solution . Perhaps for eukaryotes ASCs are the only viable solution ( although how NMD knows a 3’ stop isn’t the true stop and the real primary stop not a premature stop is unknown ) . The model is consistent with HEGs generally being shorter ( S6 Table ) but this is not a discriminating prediction as a simple translational cost argument would predict the same . Arguing against such a model however is the finding that stops in the vicinity of the true stop might not trigger NMD , the stops having to be 3’ of the last intron , at least in some species [59–61] . An alternative possibility to explain the eukaryote-prokaryote divide concerns the possibility that in some eukaryotes read-through rates can be greatly increased . Notably , the yeast prion [PSI+] state has been linked to extensive read-through via the misfolding of release factor Sup35p [6 , 62 , 63] . It is tempting to speculate that this provides a possible mechanism for increased selection of ASCs in yeast not seen in bacteria . Though the [PSI+] system in yeast is possibly best studied , it now appears that prion-like systems are present throughout the tree of life [64 , 65] , including bacteria [64] . Not all prion-like states affect translation termination , however . The identification of species susceptible to prion-induced increased translational read-through rate could provide a means to test the fail-safe hypothesis in the future . Such a model predicts co-incidence between genomes with ASC selection and prion-like systems affecting translation . Indeed , we are unaware of any bacterial prion system disrupting translational termination which would be consistent with the absence of ASC selection . The closest resemblance that we are aware of is with a system in Clostridium botulinum affecting a domain of transcription ( not translation ) termination factor Rho ( Cb-Rho ) [66] . Above we have presumed that read-through rates are the same in all genomes , with the possible exception of prion mediated read-through . In this context we note a further striking peculiarity , that ASC rates ( Z-score deviation from dinucleotide controlled null ) are especially low in GC-rich organisms . GC-rich organisms are typically thought to be those with stronger selection as the underlying mutational bias is towards AT [67 , 68] . Assuming this reflects higher effective population sizes in GC-rich organisms , the lower Z-scores in GC-rich organisms is enigmatic—if anything one might expect selection to favour more ASCs if selection is strong . It is also enigmatic as in GC-rich genomes the span to the next random stop in the 3’ domain is likely to be longer as stops are AT-rich , hence GC-rich genomes should also be under selection to conserve ASCs . However , this assumes all else is equal . If AT-rich bacteria are subject to higher read-through rates , the GC-trend might make some sense . Such a model would fit in the broader context of the possibility of stronger selection against error creation when populations are large and selection efficient [69] . Comparably , GC-rich organisms have a broader spectrum of tRNAs thought to reduce ribosomal frameshifting rates [70] . Might this also reduce read-through rates ? An alternative possibility is that in GC-rich genomes , random ASCs are less likely to function as stops . If for example AT-richness in the vicinity of a stop is needed to enable stop functioning , then a random ASC in a GC-rich genome is , for example , unlikely to have a +4T and might thus be ineffective . Indeed , experimentally tandem stops appear not to have the expected level of read-through suggesting particular context requirements [41 , 42] . We suggest that experimental determination of read-through rates in organisms with different tRNA profiles would be informative .
All analyses were performed using bespoke Python 3 . 6 scripts . Statistical analyses and data visualisations were performed using R 3 . 3 . 3 . Scripts can be found at https://github . com/ath32/ASCs . Whilst it is acknowledged that stop codons function at the mRNA level , in this analysis we have analysed chromosomal DNA sequences and henceforth refer to the three stops as TAA , TGA and TAG and to +4U enrichment as +4T . Please note that in all other contexts +1 , +2 etc refer to the position of downstream codons , not nucleotides , with +1 being the codon immediately after the primary stop . Whole-genome sequences for 3 , 727 bacterial genomes were downloaded from the European Molecular Biology Laboratory ( EMBL ) database ( http://www . ebi . ac . uk/genomes/bacteria . html , last accessed 1st August 2018 ) . For the majority of the analyses , genomes were filtered to include only one genome per genus , so as to prevent over-sampling from the very well surveyed groups and hence to reduce any bias attributable to phylogenetic nonindependence . So as to exclude plasmids , incomplete genomes or very small genomes we retained only those genomes larger than 500 , 000 base pairs . This generated a sample of 650 genomes , 644 that employ translation table ( TT ) 11 and 6 using TT4 , in which TGA no longer functions as a stop . The exception to this filtering was the specific analysis of mollicute and TT4 genomes , which were filtered directly from the raw sample of 3 , 727 genomes ( 106 and 94 genomes respectively ) . Of these genomes , only those with > 100 genes were considered for analysis . For every gene in each genome , a sequence inclusive of the primary stop followed by 27 nucleotides of the 3’ UTR was extracted by applying coding sequence coordinates to the total genomic sequence attainable in the EMBL files . Only genes with 3’ intergenic space of >30 base pairs were considered for analysis , thus ensuring a sample of genes with sufficient 3’ UTR length . Resultant sequences were filtered to retain only those 3’ sequences made up exclusively of A , T , G and C , those from genes with one stop after the initiating codon , and those from a gene body with a nucleotide length that is a multiple of three . Genomic GC values were calculated from the whole genome sequence . GC3 values are unweighted means of per gene GC3 value . Our single-celled eukaryotic set were downloaded and filtered much in the same way . 70 eukaryote genomes of unique genus were downloaded from the full Ensembl Protist set ( https://protists . ensembl . org/species . html , last accessed 8th August 2019 ) . Similar to the ciliates analysis by Adachi and Cavalcanti [40] , we extracted a sequence inclusive of the primary stop followed by 97 nucleotides of the 3’ UTR from each gene . As with the bacterial genomes , we do this by applying annotated coding sequence coordinates to the total genomic sequence . Only genes with 3’ intergenic space of >100 base pairs were considered for analysis to ensure a sample of genes with sufficient 3’ UTR length . Extracted 3’ UTR sequences were subjected to the same filters as with the bacterial ones . We increased our sample with the addition of two yeast species via bespoke downloads—S . cerevisiae ( yeastgenome . org ) and C . albicans ( candidagenome . org ) . For S . cerevisiae , annotated 3’ UTR coordinates were applied to the whole genome sequence to extract the appropriate sequence . For C . albicans , 3’ UTR sequences were located downstream from the first in-frame stop codon of downloadable ORFs ( that contain intergenic sequence ) . We exclude genomes with < 500 qualifying 3’ UTR sequences , leaving a final sample of 68 genomes . Experimental protein abundance data were downloaded for all genomes available from PaxDb [71] . Corresponding whole genome sequence files were downloaded from the European Molecular Biology Laboratory ( EMBL ) database . PaxDb external IDs and EMBL locus tags were extracted and matched to generate a sample of genomes and genes for which both PaxDb and EMBL sequence data were available ( n = 24 ) . In these genomes , qualifying genes that feature in the top and bottom quartiles of PaxDb data were defined as highly expressed genes ( HEGs ) and lowly expressed genes ( LEGs ) respectively . Only genomes with >100 qualifying HEGs and >100 qualifying LEGs were considered ( n = 22 ) . In reporting our results , we refer to the analysis of three gene groups: HEGs and LEGs which contain the qualifying genes of the 22 genomes for which there was available gene expression data , and ‘all genes’ where the qualifying genes of all filtered genomes are considered regardless of expression level . ASC frequencies for codon positions +1 to +6 were compared to expected frequencies generated from a null model where sequence is dictated solely by 3’ UTR dinucleotide content . To achieve this , we simulated 10 , 000 UTR sequences for each genome using Markov models to preserve reading frame context at the dinucleotide level . ASCs are likely to occur by chance in every genome at a given rate that is dependent on its dinucleotide content . Hence the observation of ASC frequencies that exceed our null represents enrichment beyond chance . Nucleotide frequencies used in the Markov decision process were determined by generating a string containing the 3’ UTRs of all qualifying genes from a given genome . The raw frequencies of each nucleotide within this string were calculated for the selection of the first base of each simulation . Overlapping dinucleotide frequencies were calculated for the selection of following simulated nucleotides according to the previously selected nucleotide . Simulations were complete once 21 nucleotides in length ( equivalent to a primary stop followed by 6 downstream codons ) . For each genome , ASC frequencies were calculated and compared to the mean ASC frequencies from the 10 , 000 simulated sequences at each of the 6 downstream codon positions . Comparison to null was established through the calculation of Z-scores under the assumption of a normal distribution to assess the magnitude of deviation from null in standard deviation units . Z-scores were used to complete various binomial tests using the binom_test function from the SciPy stats R package [72] . The mollicute group contains both TT11 and TT4 genomes , allowing a side-by-side comparison in closely related species . TGA is not used as a stop codon in TT4 genomes . Hence , if observed TGA frequency is lower in TT4 genomes than in TT11 genomes , this implies selection upon TGA as an ASC in TT11 genomes . We design two tests to investigate whether TGA is underused in TT4 genomes . ( i ) Frequency of TGA at codon positions +1 to +6 was plotted against genomic GC3 content in TT11 genomes from the full genome set ( n = 644 ) . A LOESS model was fit to allow the prediction of TGA frequency of TT11 and TT4 mollicute genomes according to their GC3 content at each position . TGA frequencies at each position for mollicute genomes were calculated and compared to their predicted values . The fail-safe hypothesis predicts under enrichment of TGA in the TT4 genomes , but not TT11 ones . ( ii ) Frequencies of TGA at positions +1 to +6 were calculated for TT4 mollicute genomes and compared to those of GC3 content-matched TT11 genomes from the full genome set . TT11 genomes were selected for comparison if their genomic GC3 content lies within 3 . 5% of the focal TT4 genome . Mean TGA frequencies for each position were calculated for selected TT11 genomes and compared with the corresponding TT4 genome frequency . Genes with an ASC were compared to those without . The null expectation is that those containing an ASC before ( and including ) position +N have an equal chance of possessing another ASC downstream as genes without one . Two groups of genes were thus extracted for each position–those with an ASC up to position N and those without . ASC frequencies of each group were calculated for downstream positions up to position +6 and compared . Given the nature of this experiment , no data is available for position +6 ( as there is no further downstream position to use to calculate ASCs within our chosen intergenic range ) . To consider more localised nucleotide preferences , we also repeat this methodology considering just the following base ( +N+1 ) instead of all downstream positions . For the analysis of the fourth nucleotide site of the primary stop codon , raw nucleotide frequencies ( A , T , G , C ) were calculated . Fourth site T enrichment relative to null was investigated through the comparison of T-starting codon frequency at position +1 to the mean frequency of T-starting codons throughout the 3’ UTR ( +1 to +6 ) using a Wilcoxon signed-rank test . The analysis of the fifth nucleotide site of +4T-containing genes was completed in a similar manner . Raw nucleotide frequencies at nucleotide position +5 of genes were calculated , plotted for visual comparison and used in the completion of statistical analysis . Fifth site T and fifth site C enrichment relative to null was investigated through the comparison of TT/TC-starting codon frequency at position +1 to the mean frequency of TT/TC-starting codons throughout the 3’ UTR ( +1 to +6 ) of the given genome using a Wilcoxon signed-rank test . Analysis of stop codon switches ( from non-stop to stop , or vice versa ) was completed by adapting a methodology described in previous studies [12 , 73 , 74] . Orthologous gene information for closely related species were downloaded from the Alignable Tight Genome Clusters ( ATGC ) database [75] . Corresponding whole genome sequence data was downloaded from NCBI [76] . Where possible , the same triplets ( containing two closely related ingroup species and one outgroup to allow the reconstruction of mutations by a parsimony approach ) were downloaded as used in previous studies [12 , 73 , 74] . In total , 29 ATGC triplet clusters were considered in the analysis ( 8 of the 37 clusters used in prior studies were ineligible ) . All gene sequences from each ATGC-COG ( Cluster of Orthologous genes ) were aligned using the -linsi parameter of MAFFT [77] . Aligned genes without gaps downstream of the primary stop , from all genomes , were considered together in the codon switch analysis . Ancestral codons were inferred where the outgroup codon matched at least one of the ingroup codons . A switch was recorded where one of the ingroup codons differed from both the other ingroup codon and the outgroup codon ( and thus the inferred ancestral codon ) . Frequencies of switches from non-stop to stop and stop to non-stop amongst ‘in-frame’ 3’ UTR codons were calculated . These were compared to null frequencies , calculated through the analysis of the same sequences but with +1 frameshift . To provide fair comparison between prokaryotes and eukaryotes we adopt the same set of methodologies for both genome sets . Due to the nature of ASC enrichment in eukaryotes not being universally specific to a particular codon position , we count the number of genomes in each set that possess ASC enrichment at one or more site ( between +2 to +6 ) . ASCs at a particular position were considered to be enriched if they produced a positive Chi2 value and a p-value below 0 . 05/5 ( after Bonferroni correction ) when compared to the mean from a dinucleotide controlled null ( see ‘Simulations’ section of these methods ) . As we set our p-value threshold at 0 . 01 , the probability of a genome possessing significant ASC enrichment at one or more positions by chance is 0 . 995 ( approximately 0 . 951 ) . Therefore , there is a 1–0 . 995 ( approximately 0 . 049 ) probability that a genome will contain significant enrichment at one or more positions by chance . We determined whether the number of genomes containing enrichment in each set was higher , lower , or as expected by using binomial tests under the null expectation that 4 . 9% of genomes possess enrichment purely by chance . We additionally repeat the analysis using the null model proposed by Adachi and Cavalcanti [40] . In their analysis of ASC enrichment in ciliates , they consider the probability of finding the first in-frame stop codon as a function of 3’ distance from the primary stop . The probability of finding the first stop at position +1 is equal to the probability of finding a stop at any position , p . The probability p is calculated for each genome by concatenating the first 100 non-coding nucleotides downstream of each gene , scanning this sequence for in-frame stops , and dividing the total number of stops by the total number of codon positions considered . At position +2 , the probability of finding the first stop is the probability of not finding a stop at any position upstream , in this case position +1 , multiplied by the probability of finding a stop at any position . This concept is recursively applied with each position downstream such that first ASC probability = p[1 –p] ( n– 1 ) , where n is the focal codon position . For each position +1 to +6 we calculate ASC probability and multiply this by the total number of UTR sequences analysed to determine the expected number . We then apply a Chi2 test . To determine whether the number of genomes showing significant enrichment at one of more sites is higher , lower , or as expected , we apply a binomial test as described above .
|
In all organisms , gene expression is error-prone . One such error , translational read-through , occurs where the primary stop codon of an expressed gene is missed by the translational machinery . Failure to terminate is likely to be costly , hence genomes are under selection to prevent this from happening . One proposed error-proofing strategy involves in-frame proximal additional stop codons ( ASCs ) which may act as a ‘fail-safe’ mechanism by providing another opportunity for translation to terminate . There is evidence for ASC enrichment in several eukaryotes . We extend this evidence showing it to be common but not universal in single celled eukaryotes . However , the situation in bacteria is poorly understood , despite bacteria having high read-through rates . Here , we test the fail-safe hypothesis within a broad range of bacteria . To our surprise , we find that not only are ASCs not enriched , but they may even be selected against . This provides evidence for an unusual circumstance where eukaryotes and prokaryotes could solve the same problem the same way but don’t . What are we to make of this ? We suggest that if read-through is the problem , ASCs are not necessarily the expected solution . Owing to the absence of nonsense-mediated decay , a process that makes gene truncation in eukaryotes less viable , we propose bacteria may rescue a leaky stop by mutation that creates a new stop upstream . Alternatively , raised read-through rates in some particular conditions in eukaryotes might explain the difference .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"bacteriology",
"3'",
"utr",
"messenger",
"rna",
"microbiology",
"untranslated",
"regions",
"bacterial",
"genetics",
"microbial",
"genetics",
"bacteria",
"microbial",
"genomics",
"bacterial",
"genomics",
"gene",
"expression",
"mollicutes",
"comparative",
"genomics",
"biochemistry",
"rna",
"eukaryota",
"nucleic",
"acids",
"protein",
"translation",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"computational",
"biology",
"organisms"
] |
2019
|
In eubacteria, unlike eukaryotes, there is no evidence for selection favouring fail-safe 3’ additional stop codons
|
Homologous recombination is central to repair DNA double-strand breaks , either accidently arising in mitotic cells or in a programed manner at meiosis . Crossovers resulting from the repair of meiotic breaks are essential for proper chromosome segregation and increase genetic diversity of the progeny . However , mechanisms regulating crossover formation remain elusive . Here , we identified through genetic and protein-protein interaction screens FIDGETIN-LIKE-1 INTERACTING PROTEIN ( FLIP ) as a new partner of the previously characterized anti-crossover factor FIDGETIN-LIKE-1 ( FIGL1 ) in Arabidopsis thaliana . We showed that FLIP limits meiotic crossover together with FIGL1 . Further , FLIP and FIGL1 form a protein complex conserved from Arabidopsis to human . FIGL1 interacts with the recombinases RAD51 and DMC1 , the enzymes that catalyze the DNA strand exchange step of homologous recombination . Arabidopsis flip mutants recapitulate the figl1 phenotype , with enhanced meiotic recombination associated with change in counts of DMC1 and RAD51 foci . Our data thus suggests that FLIP and FIGL1 form a conserved complex that regulates the crucial step of strand invasion in homologous recombination .
Homologous recombination ( HR ) is critical for the repair of DNA double-strand breaks ( DSBs ) in both mitotic and meiotic cells [1] . Defects in HR repair causes genomic instability , leading to cancer predisposition and various inherited diseases in humans [2] . During meiosis , HR promotes reciprocal exchange of genetic material between the homologous chromosomes by forming crossovers ( COs ) . COs between the homologs constitute a physical link that is crucial for the accurate segregation of homologous chromosomes during meiosis [3] . COs also reshuffle parental genomes to enhance genetic diversity on which selection can act [4] . Failure or errors in HR at meiosis lead to sterility and aneuploidy , such as Down syndrome in humans [5 , 6] . During meiosis , HR is initiated by the formation of numerous programmed DSBs catalyzed by the topoisomerase-like protein SPO11 [7] . DSBs are resected to form 3’ single-stranded DNA ( ssDNA ) overhangs . A central step of HR is the search and invasion of an intact homologous template by the broken DNA end , which is catalyzed by two recombinases , RAD51 and its meiosis-specific paralog DMC1 [8] . Both recombinases polymerize on 3’ ssDNA overhangs to form nucleoprotein filaments that can be cytologically observed as foci on chromosomes [9 , 10] . At this step , meiotic DSB repair encounters two possibilities to repair DSB by HR , either using the sister chromatid ( inter-sister recombination ) or using the homologous chromosomes ( inter-homolog recombination ) . The invasion and strand exchange of ssDNA displaces one strand of the template DNA , resulting in a three-stranded joint molecule ( d-loops ) . D-loops are precursors for different pathways leading to either reciprocal exchange ( CO ) or non-reciprocal exchange ( non-crossovers ) between the homologous chromosomes . Two pathways of CO formation , classified as class I and class II , have been characterized , with variable relative importance in different species [3] . Class I COs are dependent on the activity of a group of proteins collectively called ZMM ( for Zip1-4 , Msh4-5 , Mer3 ) [11] , which stabilize D-loop intermediates to promote formation of the double-Holliday junction intermediates [12] . MLH1 and MLH3 in conjunction with EXO1 promote the resolution of double-Holliday junctions as class I COs [13 , 14] . The formation of a Class I CO reduces the probability of another CO forming in the vicinity , a phenomenon termed as CO interference [15] . Additionally , recombination intermediates can be resolved by structure specific endonucleases including MUS81 , producing class II COs , which are not subjected to interference [16–18] . In Arabidopsis , class I COs constitute 85–90% of COs , while remaining minority are class II COs [19][20] . Like in most eukaryotes , DSBs largely outnumber COs in Arabidopsis [21] . This suggests that active mechanisms prevent DSBs from becoming CO . Accordingly , several anti-CO factors are identified in different species [10 , 22–31] . Previously , our forward genetic screen identified FIDGETIN-LIKE-1 ( FIGL1 ) as a negative regulator of meiotic COs in Arabidopsis [10] . Mutation in Arabidopsis FIGL1 increases meiotic CO frequency by 1 . 8-fold compared to wild type and modifies the number and/or dynamics of RAD51/DMC1 foci . FIGL1 is widely conserved and is required for efficient HR in human somatic cells through a direct interaction with RAD51 [32] . Altogether , this suggests that FIGL1 is a conserved regulator of the strand invasion step of recombination , both in somatic and meiotic cells . FIGL1 belongs to the large family of AAA-ATPase proteins that are implicated in structural remodeling , unfolding and disassembly of proteins and oligomer complexes [33 , 34] . Here , we identified a new factor limiting COs in Arabidopsis that interacts directly with FIGL1 , which we named FIDGETIN-LIKE-1 INTERACTING PROTEIN ( FLIP ) . FLIP and its interaction with FIGL1 are conserved from plants to mammals , suggesting that the complex was present at the root of the eukaryotic tree . While this manuscript was under evaluation , the homologue of FLIP in rice ( MEICA ) was also shown to regulate meiotic recombination [35] . We further showed that FLIP and FIGL1 act in the same pathway to negatively regulate meiotic CO formation , which appears to act on the regulation of the recombinases DMC1 and RAD51 . Finally , we showed that both Arabidopsis and human FIGL1-FLIP complexes interact with both RAD51 and DMC1 . Overall , this study identified a novel conserved protein complex that regulates a crucial step of homologous recombination .
We previously identified FIDGETIN-LIKE-1 ( FIGL1 ) as an anti-CO protein [10] . To better understand the role of FIGL1 during meiotic recombination , we searched for its interacting partners by tandem affinity purification coupled to mass spectrometry ( TAP-MS ) using overexpressed FIGL1 as a bait in Arabidopsis suspension culture cells [36] ( Table 1 ) . After filtering co-purified proteins for false positives ( see Materials and methods and [36] ) , we recovered , in two independent experiments , peptides from FIGL1 itself and a single additional protein . This single interacting protein is encoded by a gene of unknown function ( AT1G04650 ) , and we therefore named it as FIDGETIN-LIKE-1 INTERACTING PROTEIN ( FLIP ) . Reciprocal TAP-MS experiments using FLIP as bait recovered only FLIP and FIGL1 peptides , further suggesting that FLIP and FIGL1 belong to the same complex in vivo ( Table 1 ) . Direct interaction between FLIP and FIGL1 was further supported by yeast two hybrid ( Y2H ) assay using full length proteins ( Fig 1 ) . To map the interaction domains , we truncated FIGL1 and FLIP proteins and tested their interaction in Y2H assays ( Fig 1 ) . The N-terminal region of FIGL1 ( 1–271 amino acids ) , which lacked both the AAA-ATPase domain and the sequence similar to the human FIGNL1’s RAD51 binding domain ( FRBD ) , was sufficient to mediate the interaction with FLIP . Conversely , the N-terminal half of FLIP ( 1–502 aminoacids ) was sufficient to mediate an interaction with FIGL1 . Further , the full length or the N terminal half of FLIP was able to interact with itself , suggesting that it could oligomerize ( Fig 1 ) . Moreover , the human orthologs of FLIP ( C1ORF112 , hFLIP ) and FIGL1 ( hFIGNL1 ) also showed interaction in our Y2H assays , suggesting that this interaction is evolutionarily conserved ( Fig 2 ) . hFIGNL1 and C1ORF112/hFLIP proteins were previously showed to co-purify in pull-down assays [32 , 37] and the mouse corresponding genes are strongly co-expressed [38] , further supporting the conservation of the FIGL1-FLIP interaction from plants to mammals . The N-terminal region ( 1–290 aminoacids ) of hFIGNL1 , lacking the AAA-ATPase domain and the FRBD , was able to mediate the interaction with hFLIP , consistent with the Arabidopsis data ( Figs 1 and 2 ) . In addition , the FRBD of hFIGNL1 showed an interaction with hFLIP , suggesting that the FRBD domain could also contribute to the interaction ( Fig 2 ) . Finally , hFLIP was able to interact with itself , as shown for the Arabidopsis FLIP ( Figs 1 and 2 ) . The distribution of FLIP orthologs in eukaryotic species was analyzed using remote homology search strategy ( see Methods ) . Orthologs of FLIP could be unambiguously detected in a wide range of species including mammalia , sauria and plants but also in arthropods and unicellular species such as choanoflagellate ( Fig 3 , S3 Fig for a larger number of species , and as interactive tree http://itol . embl . de/tree/132166555992271498216301 ) . The FLIP orthologs showed low conservation at the sequence level ( e . g . AtFLIP and hFLIP sharing only 12% sequence identity ) , but they all harbor a specific DUF4487 domain ( Domain of Unknown Function ) [39] , further supporting their orthology . No FLIP ortholog could be detected in alveolata , amoebozoa and fungi . FLIP systematically co-occur with FIGL1 , which is consistent with FLIP supporting the function of FIGL1 ( Fig 3 , S3 Fig ) . The reverse is not true since there are a number of species with FIGL1 ortholog detected but no FLIP ( as in D . melanogaster and C . elegans ) . Structural predictions using RaptorX server[40] and HHpred [41] do not converge towards the same predicted fold but are both in agreement with FLIP likely folding as a long helical bundle over its full sequence . Such folds are often seen in protein recognition scaffolds suggesting FLIP could act as a FIGL1 adaptor module . Given the wide range of species harboring both FLIP and FIGL1 orthologs , the origin of this complex is probably quite ancient at the root of the eukaryotic tree suggesting that absence of FLIP-FIGL1 in some eukaryotic clades ( such as Dikarya that regroups the fungi Basidiomycetes and Ascomycetes ) is due to independent gene loss events . In parallel to the protein complex purification approach , FLIP was independently recovered in a genetic screen aiming at identifying meiotic anti-CO factors that previously uncovered FIGL1 . Using fertility ( fruit length ) as a proxy for CO formation , we screened for ethyl methane sulfonate-generated mutations that restored COs in class I CO deficient mutants ( zmm ) . As COs provide a physical link between pairs of chromosomes ( bivalents ) , mutation of an anti-CO factor is expected to restore bivalent formation in CO-deficient mutants , thus improving balanced chromosome segregation and fertility [22] . This genetic screen led to the identification of several anti-CO factors , defining three pathways that limit COs in Arabidopsis: ( i ) The FANCM helicase and its cofactors [22 , 23]; ( ii ) The AAA-ATPase FIDGETIN-LIKE-1 ( FIGL1 ) [10]; ( iii ) The RECQ4 helicase-Topoisomerase 3α-RMI1 complex [24 , 25] . Here , we isolated an additional suppressor of hei10 , one of the zmm mutants that are deficient in class I COs [42] . This suppressor , hei10 ( S ) 320 showed longer fruit length compared to hei10 and bivalent formation was restored to an average of 3 . 7 bivalents per cell compared to 1 . 5 in hei10 and 5 in wild type ( Fig 4 ) , suggesting a partial restoration of CO formation . Whole genome sequencing and genetic mapping of hei10 ( S ) 320 defined a genetic interval containing five putative causal mutations . One of them resulted in a stop codon in the gene AT1G04650 , which encodes FLIP ( flip-1 W305>STOP ) ( Fig 4 ) . An independent mutation in FLIP ( T-DNA Salk_037387/ flip-2 ) , was also able to restore bivalent formation in hei10 ( Fig 4 ) . Further , flip-1/flip-2 hei10 exhibited restored bivalents ( Fig 4 ) , demonstrating that flip-1 and flip-2 are allelic and that mutations in FLIP are causal for the restoration of bivalents in hei10 . The flip-1 mutation was also able to restore bivalent formation in msh5 ( Fig 4 ) , another essential gene of the class I CO pathway , suggesting that effect of the flip mutation is not specific to hei10 but allows the formation of COs in absence of the class I pathway . No growth or development defects were observed in the flip mutants . Meiosis progressed normally in single flip-1 and flip-2 , except that a pair of univalent was observed at metaphase in ~14% of the cells ( n = 12/99 in flip-1; n = 9/50 in flip-2 ) . ( Fig 4B and 4C ) . Similarly , we observed a low frequency of univalents in figl1-1 ( n = 6/82 cells ) that has been missed in previous analyses [10] , and in figl1-1 flip-1 ( n = 5/66 cells ) . This suggests a slight defect in implementation of the obligate COs in absence of FLIP or FIGL1 . We also observed a moderate increase in the frequency of pollen death ( wild type 1 . 1% ±0 . 3 , figl1-1 5 . 2% ±1 . 8 , flip-1 5 . 8% ±0 . 6 , figl1-1 flip-1 4 . 3% ±1 . 1; n = 5 plants per genotype , ≥300 pollen grains/plant ) and a decrease in the number of seeds per fruit was observed in the single and double mutants ( Fig 4D ) . We next monitored the direct effect of FLIP mutation on CO frequency by tetrad analysis and measured recombination in six genetic intervals defined by fluorescent tagged markers that confer fluorescence in pollens [43] . CO frequencies in flip-1 were significantly increased in four intervals out of six tested , in the range of +15% to +40% compared to wild type ( Fig 5 ) . This increase in CO frequencies due to loss of FLIP is consistent with the restoration of bivalent formation in zmm mutants and implies that FLIP limits COs during meiosis in wild type . FLIP physically interacts with FIGL1 ( see above ) , suggesting that they can act together to limit COs . We therefore compared recombination in flip-1 , figl1-1 and the double mutant by tetrad analysis . On the four intervals tested , figl1-1 showed an average of ~70% CO increase compared to wild type , corroborating previous findings ( Fig 5 ) , which is significantly higher than flip-1 . Combining flip-1 and figl1-1 mutations did not lead to a further increase in recombination suggesting that FIGL1 and FLIP act in the same pathway to negatively regulate CO formation ( Fig 5 ) . However , FIGL1 may be partially active in absence of FLIP as flip-1 increases CO frequencies to a lesser extent than figl1-1 . We next explored the origin of extra COs in flip . In the flip-1 spo11-1 double mutant , bivalent were completely abolished and 10 univalents were observed at metaphase I , ( Fig 4B ) , showing that all COs in flip-1 are dependent on SPO11-1 induced DSBs . Two classes of COs exist in Arabidopsis: class I COs are dependent on ZMM proteins and are subjected to interference , while class II are insensitive to interference and involve structure specific endonucleases including MUS81 [21] . The flip-1 mutation restored CO formation in two zmm mutants , hei10 and msh5 ( see above ) . Further , tetrad analysis of three pairs of intervals showed reduced interference in flip-1 compared to wild type ( Fig 6A ) . Finally , we examined meiosis in the flip-1 mus81 double mutant . While no chromosome fragmentation is observed in single flip-1 or mus81 mutants , chromosome fragments were observed at anaphase I in the flip-1 mus81 double mutant ( n = 31/31 cells . Fig 6B ) . This suggests that MUS81 is required for resolution of recombination intermediates formed in flip-1 . Altogether , the extra COs produced in flip-1 appeared to be dependent on the class II pathway , as previously shown for figl1-1 [10] . Based on genetic and physical interactions between FIGL1 and FLIP , we next hypothesized that FLIP might regulate RAD51 and DMC1 foci during meiosis , as previously shown for FIGL1 [10] . We thus performed RAD51 ( Fig 7 ) and DMC1 ( Fig 8 ) immunolocalization in figl1 , flip and figl1 flip in combination with staining of the chromosome axis ( ASY1 ) and the synaptonemal complex ( ZYP1 ) to follow their localization at early , mid and late prophase stages . In wild type , RAD51 foci appear at leptotene and increase at zygotene ( Fig 7 ) . The number of RAD51 foci at leptotene is increased by ~2 fold in figl1 and figl1 flip . An increase is also observed in flip at leptotene , but to a lesser extent and at the edge of significance . At zygotene the number of RAD51 foci was not significantly different between the two single mutants and the wild type , but appeared increased in figl1 flip . This suggests that FIGL1/FLIP negatively regulates the formation or the turnover of RAD51 foci . In wild-type , DMC1 foci first appear at leptotene , peak at zygotene and almost disappear at pachytene ( 33/46 had less than 10 foci ) ( Fig 8 ) . At both leptotene and pachytene , a large increase of DMC1 foci was observed in figl1 and figl1 flip . The same trend was observed in flip , but with a lesser increase and barely significant . At zygotene , only the single figl1 had a significantly higher number of DMC1 foci . Altogether , this suggests that FIGL1/FLIP regulate the kinetics of appearance and disappearance of DMC1 foci , with FIGL1 playing a more central role than FLIP . Persistence of DMC1 foci may represent unrepaired DSBs that are eventually repaired ( possibly by MUS81 ) , as no chromosome fragmentation was observed at anaphase I in figl1 or flip mutant . One known positive regulator of DMC1 in plants is SDS , a meiosis-specific cyclin-like protein [44 , 45] . In absence of SDS , DMC1 foci do not form , synapsis and COs are abolished , but DSBs and RAD51 foci are formed and repair is completed , presumably using the sister as template [44 , 45] . We previously showed that mutation in FIGL1 restores DMC1 focus formation , synapsis , and bivalent formation in sds [10] . These results argued for antagonistic functions of SDS and FIGL1 , the former positively and the latter negatively regulating DMC1 foci formation and DMC1-mediated homolog engagement . Here , we similarly showed that DMC1 foci and synapsis are partially restored in sds flip double mutants as compared to sds ( Fig 9A , 9B and 9C ) . Moreover , 4 to 5 bivalents per metaphase I were observed in sds flip ( n = 57 ) while their formation is almost completely abolished in sds ( 0 . 12 bivalents per metaphase I , n = 50 ) ( Fig 9D ) . However , recombination is not completely restored in sds flip as chromosome fragmentation is observed at anaphase I . Accordingly , fertility is only partially restored in sds flip compared to sds ( Fig 9E ) . Taken together , this strongly suggests that FIGL1 and FLIP antagonize SDS in the regulation of DMC1 focus formation and DMC1 mediated inter-homolog interactions and crossover formation . In both figl1 sds [10] and sds flip ( Fig 9D ) , bivalents at metaphase I had slightly aberrant shape and chromosome fragmentation was observed at anaphase I . This suggests that FIGL1 and FLIP may have a function in DSB repair downstream of homologous template invasion or that the recombination intermediates restored in absence of both sds and figl1/flip are aberrant . Our genetic interaction and immuno-localization studies in Arabidopsis suggest that the FIGL1/FLIP complex might regulate the function of RAD51 and DMC1 , directly or indirectly . In addition , it was shown that human FIGNL1 interacts with human RAD51 through a domain called FRBD [32] . Hence , we set out to examine whether Arabidopsis and human FIGL1 and FLIP interact with RAD51 and DMC1 , using Y2H assays . Consistent with published data , the Y2H assay detected an interaction between the FRBD domain of human FIGNL1 and RAD51 , though it was weak and only positive in one direction ( Fig 2 ) . Similarly , we detected an interaction between Arabidopsis FIGL1 and RAD51 , mediated by the predicted FRBD domain ( Fig 1 ) . In addition , we observed a clear interaction between human FIGNL1 and DMC1 , mediated by the FRDB domain ( Fig 2 ) . Arabidopsis FIGL1 interacted also with DMC1 , although the interaction was detected only in one direction ( Fig 1 ) . This suggests that FIGL1 can interact directly with both RAD51 and DMC1 and that these interactions are conserved in plants and mammals . Next , we tested interaction between FLIP and the two recombinases , with both plant and human proteins . Human FLIP interacted with DMC1 , suggesting that FLIP could reinforce the interaction of the FIGL1-FLIP complex with DMC1 ( Fig 2 ) . However , our Y2H assay did not reveal any interaction between Arabidopsis FLIP and DMC1 ( Fig 1 ) . No interaction was detected between FLIP and RAD51 , for both human and Arabidopsis proteins ( Figs 1 and 2 ) .
We identified , by two different approaches , FLIP as a new factor that genetically and physically interacts with FIGL1 [10] and regulates meiotic recombination . We showed that ( i ) FIGL1 and FLIP form a conserved complex; ( ii ) FLIP and FIGL1 are anti-CO factors that act in the same pathway to regulate meiotic recombination; ( iii ) kinetics of DMC1 and RAD51 foci are modified in figl1 and , to a lesser extent , in flip; ( iv ) flip and figl1 restore DMC1 focus formation and inter-homolog interactions ( synapsis ) in the sds mutant; ( v ) FIGL1-FLIP complex interacts with RAD51 and DMC1 , and this interaction is evolutionarily conserved in both plants and mammals . FIGL1 was previously shown to be involved in meiotic recombination in Arabidopsis , and in recombination-mediated DNA repair in human somatic cells [10 , 32 , 46] . In contrast and despite the conservation in many eukaryotes , FLIP was of unknown function . We propose a model wherein FIGL1 and FLIP act as a complex that negatively regulates the strand invasion step of HR by interacting with DMC1/RAD51 and modulating their activity/dynamics . FIGL1 belongs to the AAA-ATPase group of proteins , which typically function by dismantling the native folding of their target proteins [33 , 34] . Therefore , it is tempting to suggest that the FLIP-FIGL1 complex may directly disrupt DMC1/RAD51 filaments using the unfoldase activity of FIGL1 . Supporting this possibility , both Arabidopsis and human FIGL1 physically interact with DMC1 and RAD51 . We showed that FLIP and FIGL1 act together to limit meiotic COs in Arabidopsis , but the increase in CO frequency is lower in flip than in figl1 ( ~30% and ~70% increase compared to wild type , respectively ) . This difference in CO frequency could be attributed to the catalytic activity of the complex being supported by FIGL1 . We suggest that FLIP could only be partially required for FIGL1 enzymatic functions in vivo , acting as a co-factor or reinforcing the affinity and/or the specificity of the interaction of the FIGL1/FLIP complex with the target . In our assay , human FLIP interacted with DMC1 , suggesting that FLIP could indeed function to facilitate FIGL1 activity towards DMC1 . We could not detect an interaction between FLIP and RAD51 but we cannot rule out the possibility that FLIP facilitates also interaction of the complex with RAD51 . Indeed , several lines of evidence suggest that FLIP could act in conjunction with FIGL1 in its role in somatic HR [32]: Down-regulation of hFLIP induces reduced growth of HeLa cells [38] . FLIP in mouse is strongly co-expressed with cancer related genes and the knock out mouse is not viable [38 , 47] . Finally , FIGNL1 and hFLIP are strongly co-regulated in mouse expression data [38] . Overall , this argues for a conserved role of the FIGL1/FLIP complex in regulating RAD51/DMC1 activities during both somatic and meiotic HR . Beyond Arabidopsis and humans , FIGL1 and FLIP are conserved in all vertebrates and land plants examined in the current study . FIGL1 and FLIP can be also detected in species from other distant clades , suggesting that this complex emerged early in the evolution of eukaryotes ( Fig 3 ) . However , some clades appear to have lost both FIGL1 and FLIP , most notably the Alveolata and Dikarya ( which regroups the fungi Basidiomycetes and Ascomycetes ) . In those species , RAD51/DMC1 might be regulated independently of FIGL1-FLIP . Species with a FLIP ortholog also systematically have a FIGL1 , but the reverse is not true , several species/clades having FIGL1 but no detectable FLIP orthologs . This is consistent with our experimental data that argue for FIGL1 being the core activity of the complex and FLIP as a dispensable factor for FIGL1 activity . While RAD51 appears to be universally conserved , DMC1 is absent in a number of species ( Fig 3 ) . Moreover , we could not find any correlation between presence/absence of FIGL1 or FLIP with DMC1 . Some species have DMC1 but no FIGL1/FLIP ( e . g . many fungi ) , while others have DMC1 and FIGL1 but not FLIP ( e;g some nematodes ) , or FIGL1 and FLIP without DMC1 ( e . g . Chrophyta ) . Altogether , our phylogenic analysis supports that neither FIGL1 nor FLIP are specific to DMC1 , and that the FIGL1-FLIP complex can regulate the activity of both RAD51 and DMC1 . The FIGL1 complex may also have additional functions unrelated to HR [48] . We suggest that FIGL1 and FLIP could limit strand invasion mediated by RAD51 and DMC1 . How could the lack of this function lead to an increase in the frequency of meiotic COs as observed in flip and figl1 ? One conceivable explanation is that the absence of FLIP and FIGL1 changes the equilibrium between invasions on inter-sister versus inter-homolog , leading to the formation of higher numbers of inter-homolog joint molecules and eventually more COs . However , DSBs and presumably inter-homologous joint molecules are already in large excess to COs in wild type [21] , making it hard to believe that a simple increase in their number would increase CO frequencies . We favor another possibility in which the lack of the FLIP / FIGL1 activity generates aberrant recombination intermediates through either multi-chromatid invasions or invasion of both ends of a break . The observation that the structure specific nuclease MUS81 becomes essential for completion of repair in figl1 and flip suggests that indeed some novel class of intermediates arise in these mutants . Thus , we favor the hypothesis in which the absence of FLIP and FIGL1 leads to excessive and/or late activity of DMC1/RAD51 , generating aberrant joint molecules such as multi-chromatid joint molecules [49 , 50] . Such unusual structures would need structure specific endonucleases to be resolved , leading to increased COs . Therefore , the function of FLIP-FIGL1 in wild type context could prevent formation of aberrant recombination intermediates by functioning as a quality control of strand invasion . Intriguingly , some univalents are observed at metaphase in figl1 and flip . This suggests that the implementation of the obligate CO is slightly affected in absence of FIGL1/FLIP . One possibility is that some recombination intermediates designated to become COs fail to mature into actual COs because they have aberrant structures generated by unregulated DMC1/RAD51 . In such scenario , these intermediates would be eventually repaired as non-crossovers , as no chromosome fragmentation is observed in the mutants . While this manuscript was under evaluation , the homologue of FLIP in rice ( MEICA ) has been shown to regulate meiotic recombination [35] . The mutation of meica restores COs in msh5 , suggesting that the anti-CO function of FLIP/MEICA is conserved in plants . However , both Osfignl1 [51] and meica [35] mutants in rice show significant chromosome fragmentation at anaphase I , suggesting that the FIGL1-FLIP/Os FIGNL1-MEICA complex is more crucial for the completion of DSB repair in rice than in Arabidopsis . In conclusion , we uncovered a conserved FIGL1-FLIP complex that directly binds to RAD51/DMC1 and could negatively regulate strand invasion during homologous recombination . It would be of particular interest to further study the function of this complex in mammalian systems and in biochemical assays . Unraveling proteins playing a role in HR pathway would provide better understanding related to various inherited diseases in humans pertaining to defects in HR repair proteins [2] . Targeting HR protein could increase the sensitivity of cancer cells to anti-cancer drugs [52] . Thus , FIGL1-FLIP could represent potential targets for cancer therapy .
The Arabidopsis lines used in this study were: hei10-2 ( N514624 ) [42] , msh5-2 ( N526553 ) [53] , mus81-2 ( N607515 ) [18] , spo11-1-3 ( N646172 ) [54] , sds-2 ( N806294 ) [44] , figl1-1 [10] , zip4-2 ( N568052 ) [55] . Tetrad analysis lines ( FTLs ) used were as follows: I2ab ( FTL1506/FTL1524/FTL965/qrt1-2 ) , I3bc ( FTL1500/FTL3115/FTL1371/qrt1-2 ) and I5cd ( FTL1143/FTL1963/FTL2450/qrt1-2 ) . FTLs were obtained from Gregory Copenhaver [43] . Suppressor hei10 ( s ) 320/flip-1 was sequenced using iIlumina technology at the Genome Analysis Centre , Norwich , UK . Mutations were identified through MutDetect pipeline [23] . The flip-1 causal mutation was C to T substitution at the position chr1:1297137 ( Col-0 TAIR10 assembly ) . flip-2 ( N662136 ) T-DNA mutant was obtained from the Salk collection , distributed by the NASC . The primers used for genotyping are listed in the S4 Table . Meiotic chromosomes from anthers were spread and DAPI stained as previously described [56] . For cytological detection of meiotic proteins , male meiotic chromosome spreads from prophase I were prepared as described in Armstrong et al . [57] . Spread slides were either immediately used for immuno-cytology or stored at -80°C before immunostaining . Chromosome axis protein ASY1 and synaptonemal complex protein ZYP1 staining were performed to define substages of prophase I . Leptotene stage had only ASY1 signal , while zygotene and pachytene cells showed partial stretches of ZYP1 signal or 95–100% of ZYP1 signal in the nucleus , respectively . Primary antibodies used for immunostaining were: anti-DMC1 ( 1:20 ) [58] , anti-RAD51 ( 1:500 ) [59] , anti-ZYP1 raised in rat ( 1:250 ) [60] or rabbit ( 1:500 ) and anti-ASY1 raised in guinea pig ( 1:250 ) or chicken ( 1:50 ) [57] . Secondary antibody: Alexa fluor 488 ( A-11006 ) ; Alexa fluor 568 ( A-11077 ) ; Alexa fluor 647 ( A-11006 ) , anti-rabbit 647 ( 6444–31 Southern Biotech ) and super clonal Alexa fluor®488 , ( A-27034 ) obtained from Thermo Fisher Scientific were used in 1:400 dilution . Images were obtained using a Zeiss AxioObserver microscope and were analyzed by Zeiss Zen software . In case of DMC1 and RAD51 staining , all images were acquired at 2s exposure , and DMC1 and RAD51 foci were counted by using Fiji software after exporting images in tiff format [61] . Briefly , DAPI or ASY1 images were binarized using the ‘triangle’ intensity thresholding method followed by a binary morphological closing operation to localize meiotic chromosomes and to mark them as regions of interest ( ROI ) . In parallel , a white top-hat transform was applied to DMC1 or RAD51 images . Significant peaks located within chromosome ROI were counted as foci . Scatter dot plots and statistical analysis were performed using the software GraphPad Prism 6 . We used FTLs [43] to estimate male meiotic recombination rates at three pairs of genetic intervals I2ab , I5cd and I3bc . For each set of experiment , heterozygous plants were generated for the pairs of linked fluorescent markers and siblings from the same segregating progeny were used to compare the recombination frequency between different genotypes . Slides were prepared as described previously [43] . Tetrads were counted and sorted to specific classes ( A to L ) [43] using a pipeline developed on the Metafer Slide Scanning Platform . For each tetrad , attribution to a specific class was double checked manually . Genetic sizes of each interval was calculated using Perkins equation [62] as follows: D = 100× ( Tetratype frequency+6× Non-Parental Ditype frequency ) /2 in cM . The Interference ratio ( IR ) was measured as described previously [63] [43] . Briefly , in two adjacent intervals I1 and I2 , genetic size of I1 was calculated for the two populations of tetrads in I2 interval–D1 is at least with one CO in I2; D2 is without CO in I2 . The ratio of D1/D2 revealed presence ( when IR<1 ) or absence ( when IR is close to 1 or >1 ) of the interference . A chi square test is performed to test the null hypothesis ( H0: D1 = D2 ) . The average of the two reciprocals is depicted on the graph ( Fig 6A ) . Cloning of the FIGL1 open reading frame ( ORF ) is described in [10] . The AtFLIP ORF was amplified using gene-specific primer ( S4 Table ) on cDNA prepared from Arabidopsis flower buds ( Col-0 accession ) . The full length or truncated ORFs of FLIP were cloned into pDONR207/pDONR201 vectors to produce entry clones . All plasmid inserts were verified by Sanger sequencing . The ORFs for human FIGNL1 ( BC051867 ) , RAD51 ( BC001459 ) , DMC1 ( BC125163 ) were obtained from the human orfeome collection , while human FLIP ( IMAGE clone: 30389801 ) ORF was ordered from Source BioScience , UK For yeast two hybrid assays , AtFIGL1 , AtFLIP , AtRAD51 and AtDMC1 as well as their respective human orthologs ( hFIGNL1 , hFLIP , hRAD51 , hDMC1 ) were cloned into destination vectors pGBKT7 and pGADT7 by the Gateway technology . The fidelity of coding sequence of all clones was verified by sequencing . Yeast two hybrid assays were carried out using Gal4 based system ( Clontech ) [64] by introducing plasmids harboring gene of interest in yeast strains AH109 and Y187 and interaction were tested as previously described [65] . TAP-MS analysis was performed as described previously [36] . Briefly , the plasmids expressing FLIP or FIGL1 fused to the double affinity GSrhino tag [36] were transformed into Arabidopsis ( Ler ) cell-suspension cultures . TAP purifications were performed with 200 mg of total protein extract as input and interacting proteins were identified by mass spectrometry using an LTQ Orbitrap Velos mass spectrometer . Proteins with at least two high-confidence peptides were retained only if reproducible in two experiments . Non-specific proteins were filtered out based on their frequency of occurrence in a large dataset of TAP experiments with many different and unrelated baits as described [36] . Identification of putative orthologs of FLIP , FIGL1 , DMC1 and RAD51 was performed following different strategies based on the sequence divergence and the existence of paralogs . Since FLIP sequence diverged significantly during evolution without detectable paralog , 3 iterations of HHblits [66 , 67] against the uniclust30_2017_04 database were sufficient to retrieve 139 sequences belonging to plants and metazoa species . To get NCBI entries of those proteins , a PSSM generated from the recovered alignment was used as input of a jump start PSI-blast [68] against the eukaryotic refseq_protein database [69] . For DMC1 and RAD51 , reciprocal best hits of blast searches were used to identify the most likely ortholog in every species . First , DMC1 in H . sapiens and S . cerevisiae sequences were blasted against the refseq_protein database to gather a set of DMC1 candidates . Each of these candidates was reciprocally blasted against the protein sequences of six fully sequenced genomes wherein DMC1 and RAD51 genes could be unambiguously identified and which were chosen spread over the phylogenetic tree ( H . sapiens , S . cerevisiae , C . reinhardtii , T . gondii , P . falciparum , T . cruzi ) . Detection of a DMC1 ortholog was considered correct when one of the 6 DMC1 genes was spotted out as best hit with an alignment score at least 10% higher than that of the second best hit , supporting its significantly higher similarity to DMC1 than to RAD51 . The same strategy was followed to assign RAD51 orthologs . In the case of FIGL1 , large number of paralogs such as spastin , fidgetin , katanin or sap1-like proteins render the global analysis more complex . A phylogenetic tree was initially built focused on the AAA ATPase domain of 600 protein sequences belonging to fidgetin , spastin , katanin , sap1 and VPS4 families . They were aligned using mafft einsi algorithm [70] and tree was built with PhyML [69] using the LG model for aminoacid substitution and 4 categories in the discrete gamma model . This prior analysis helped to delineate which homologs could be considered as orthologs of H . sapiens and A . thaliana FIDGETIN-like proteins . For the 373 fully sequenced species presented in Fig 3 , reciprocal blast best hit searches were then performed to retrieve the Fidgetin-like ortholog when present . FIGL1 ortholog candidates were retrieved from a blast of H . sapiens and A . thaliana FIGL1 sequences against the refseq_protein database and were assessed by reciprocal best hit searches using these candidates as query against genomes of H . sapiens and A . thaliana . Detection of FIGL1 orthology was assessed if best hit was FIGL1 sequence with an alignment score at least 10% higher than that of the second best hit . For a limited number of species , orthologs were suspected but not identified in any of the NCBI databases . Targeted blast searches where then performed on their genomes using the Joint Genome Institute ( JGI ) server to further probe the existence of these orthologs which could be detected in 7 cases . All the NCBI and JGI gene entries are listed in S2 Table and can be easily retrieved from the interactive tree ( http://itol . embl . de/tree/132166555992271498216301 ) [71] by passing the mouse over the species names .
|
Homologous recombination is a DNA repair mechanism that is essential to preserve the integrity of genetic information and thus to prevent cancer formation . Homologous recombination is also used during sexual reproduction to generate genetic diversity in the offspring by shuffling parental chromosomes . Here , we identified a novel protein complex ( FLIP-FIGL1 ) that regulates homologous recombination and is conserved from plants to mammals . This suggests the existence of a novel mode of regulation at a central step of homologous recombination .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"taxonomy",
"protein",
"interactions",
"brassica",
"dna-binding",
"proteins",
"phylogenetics",
"data",
"management",
"model",
"organisms",
"phylogenetic",
"analysis",
"experimental",
"organism",
"systems",
"dna",
"plants",
"homologous",
"recombination",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"computer",
"and",
"information",
"sciences",
"genetic",
"interference",
"proteins",
"evolutionary",
"systematics",
"biochemistry",
"eukaryota",
"plant",
"and",
"algal",
"models",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"repair",
"dna",
"recombination",
"evolutionary",
"biology",
"organisms"
] |
2018
|
FIGL1 and its novel partner FLIP form a conserved complex that regulates homologous recombination
|
We compared conscious and nonconscious processing of briefly flashed words using a visual masking procedure while recording intracranial electroencephalogram ( iEEG ) in ten patients . Nonconscious processing of masked words was observed in multiple cortical areas , mostly within an early time window ( <300 ms ) , accompanied by induced gamma-band activity , but without coherent long-distance neural activity , suggesting a quickly dissipating feedforward wave . In contrast , conscious processing of unmasked words was characterized by the convergence of four distinct neurophysiological markers: sustained voltage changes , particularly in prefrontal cortex , large increases in spectral power in the gamma band , increases in long-distance phase synchrony in the beta range , and increases in long-range Granger causality . We argue that all of those measures provide distinct windows into the same distributed state of conscious processing . These results have a direct impact on current theoretical discussions concerning the neural correlates of conscious access .
We adopted a theory-driven approach , trying to test experimentally a set of explicit predictions derived from the global workspace model of conscious access . This model , in part inspired from Bernard Baars' theory [30] , proposes that at any given time , many modular cerebral networks are active in parallel and process information in an unconscious manner [22 , 23 , 31 , 32] . Incoming visual information becomes conscious , however , if and only if the three following conditions are met [23]: Condition 1: information must be explicitly represented by the neuronal firing of perceptual networks located in visual cortical areas coding for the specific features of the conscious percept . Condition 2: this neuronal representation must reach a minimal threshold of duration and intensity necessary for access to a second stage of processing , associated with a distributed cortical network involved in particular parietal and prefrontal cortices . Condition 3: through joint bottom-up propagation and top-down attentional amplification , the ensuing brain-scale neural assembly must “ignite” into a self-sustained reverberant state of coherent activity that involves many neurons distributed throughout the brain . Why would this ignited state correspond to a conscious state ? The key idea behind the workspace model is that because of its massive interconnectivity , the active coherent assembly of workspace neurons can distribute its contents to a great variety of other brain processors , thus making this information globally available . The global workspace model postulates that this global availability of information is what we subjectively experience as a conscious state . Neurophysiological , anatomical , and brain-imaging data strongly argue for a major role of prefrontal cortex , anterior cingulate , and the associative areas that connect to them , in creating the postulated brain-scale workspace . In the present work , we measured the neural correlates of visually masked words and contrasted them with those of consciously visible unmasked words . On each trial , patients were randomly presented with a masked word , a visible word , or with corresponding control stimuli in which the words were replaced by blank screens . In the masked condition , words or blank screens were presented for 29 ms , preceded by a forward mask and followed by a backward mask . In the unmasked conditions , words or blank screens were made visible by simply removing the backward mask ( see Materials and Methods and Figure 1 for details ) . In order to discard activations induced by the masks , we always subtracted from word-present conditions the corresponding blank condition . This subtraction allowed us to isolate the entire processing path evoked by the masked or unmasked word . In the light of our model , the masked–unmasked contrast corresponds to a comparison between a visual representation satisfying only condition 1 and a representation satisfying all three conditions for conscious access listed above . The global workspace model therefore leads to the following four predictions .
Unmasked words were consciously reportable , and were categorized better than chance level in a forced-choice categorization task on the emotional valence of words ( mean discriminability index d′ = 2 . 24 ( +1 . 14 to +3 . 04 ) , all individual χ2 p-values and group analysis Student t-test p-value < 0 . 0001 ) . In sharp contrast , masked words were not consciously visible , and forced-choice performance was at chance level for each of the implanted patients ( mean d′ = 0 . 02 ( −0 . 18 to +0 . 27 ) , all p-values >0 . 2 ) . Response times ( RTs ) were similar across the two masking conditions ( p > 0 . 38 in Student t-test performed on mean RTs; masked mean RT = 1 , 640 ms , unmasked mean RT = 1 , 300 ms ) . We defined masked effects by subtracting the voltages measured on masked blank trials from those associated with masked word trials . This subtraction allowed us to isolate , on a sample-by-sample basis , activations associated with masked word processing ( see Figure 1 and Materials and Methods for our detailed three-step statistical procedure ) . Unmasked effects were defined similarly by subtraction of the unmasked word and unmasked blank conditions . Figure 1 shows the anatomical distribution of the 176 reconstructed bipolar montages ( “electrodes” ) from which we obtained valid data across the ten patients . The bipolar subtraction of nearby recording sites reduced distant influences , including those from the reference electrode , and resulted in a signal tightly localized to the implanted structure . Although measures were obtained for all four lobes , it should be kept in mind that major sectors of dorsolateral and polar prefrontal cortex as well as parietal cortex were not sampled . Among the 176 electrodes , 24 . 4% ( 43 electrodes ) showed a significant effect for masked words . These effects were observed across all implanted structures but with a predominance of effects on occipital electrodes: 22/55 ( 40% ) within the occipital lobe , 11/78 ( 14 . 1% ) within the temporal lobe , 4/24 ( 16 . 7% ) within the parietal lobe and 6/19 ( 31 . 6% ) within the frontal lobe ( χ2 p-value = 0 . 004 ) . Concerning unmasked words , 68 . 8% of all electrodes ( 121 electrodes ) showed a significant effect of word presence—a remarkably high percentage , given that electrodes had been placed at clinically relevant sites without consideration of their relevance to our visual stimuli . Unmasked effects were observed across all implanted structures but with a particular emphasis on the frontal lobe: 42/55 ( 76 . 4% ) within the occipital lobe , 49/78 ( 62 . 8% ) within the temporal lobe , 12/24 ( 50% ) with the parietal lobe , and 18/19 ( 94 . 7% ) within the frontal lobe ( χ2 p-value = 0 . 005 ) . The frontal lobe showed a major difference between trials containing masked and unmasked words: almost all contacts were systematically activated during conscious processing of unmasked words ( ∼95% ) , whereas this was not the case during unconscious processing of masked words ( ∼32% ) . In order to better assess the specificity of this last result , we ran an ANOVA to directly compare the impact of masking on the proportion of activated electrodes between occipital and frontal lobes . A main effect of masking was observed ( 86% versus 36%; p < 10−4 ) , confirming the larger spatial extension of unmasked activations as compared to masked activations . No main effect was observed between frontal and occipital electrodes ( 58% versus 63% , p > 0 . 5 ) . Crucially , we observed a significant interaction between frontal and occipital cortices and masking condition ( p = 0 . 05 ) , assessing the larger differential activation of frontal lobe between masked and unmasked conditions , as compared to the pattern observed in posterior visual cortex . Note that this spatial analysis is affected by the nonhomogenous sampling of brain regions , minimizing the contribution of cortical structures that were less frequently implanted . Nevertheless , masked effects were more frequent on posterior than on anterior electrodes , whereas unmasked effects were homogeneously distributed . To demonstrate this point , we examined the distribution of the anterior-posterior ( y ) coordinate , in Talairach space , of the electrodes showing a significant effect , and compared it to the spatial distribution of all 176 recorded electrodes ( see Figure S1 ) . For masked words , the spatial distribution of significant electrodes was strongly shifted towards posterior sites ( p < 10−6 , Kolmogorov-Smirnov test , relative to the distribution of either the whole set of 176 electrodes or to those showing an unmasked effect ) . The same analysis conducted on the cumulative distribution of unmasked effects showed a spatial distribution statistically indistinguishable from that of the whole set of electrodes . Masked and unmasked words were also distinguished by the temporal extension of their activation . A crude analysis , averaging across all electrodes , revealed that masked effects had a mean duration of 60 ms , much shorter than the mean of 378 ms for unmasked effects ( p < 10−6 ) . Masked effects also showed an earlier onset ( mean = 366 ms; median = 301 ms ) than unmasked effects ( mean = 522 ms; median = 497 ms; t-test , p < 10−5 ) . A more relevant analysis focusing on the first significant effect within the subset of electrodes with both masked and unmasked effects showed similar latencies between these two conditions ( 299 ms and 348 ms , respectively , for masked and unmasked conditions; t-test , p = 0 . 30 ) . Indeed , up to approximately 200 ms after word onset , glass brain visualization of the spatiotemporal dynamics of masked and unmasked effects showed a strikingly similar pattern of activations within posterior occipitotemporal cortical regions ( see Figure 2 ) . This dynamic pattern is very comparable to the “feedforward sweep” described by Lamme in the nonhuman primate visual cortex using multiunit recordings for latencies up to 100 ms after visual stimulus onset [5] . Clear differences between masked and unmasked effects appeared after 150 ms , with a progressive increase in the intensity and spatial extension of unmasked effects , while masked effects decayed and did not show a similar spatial extension ( see Videos S1 and S2 ) . This general pattern was observed on individual electrodes ( see Figure 3 ) . The initial response was often indistinguishable between masked and unmasked effects . This initial common response was usually followed by later effects specifically for the unmasked condition . Out of 14 electrodes showing this pattern with our statistical criteria , 11 of them also showed a polarity inversion of the late sustained effects relative to the polarity of the initial effect . A cortical lobe analysis focusing on the proportion of electrodes showing a significant effect over time ( Figure 4 ) showed a similar proportion of electrodes activated by masked and unmasked words at short latencies , whereas at later latencies , the effects were increasingly specific to the unmasked condition . An analysis of the mean voltage power , averaged across electrodes within one lobe , showed a similar temporal dynamics , and additionally allowed us to detect a progressive time delay in the peak of the initial activation common to masked and unmasked words . The time point at which the first significant divergence between masked and unmasked effects occurred , as estimated by a t-test ( p < 0 . 05 ) , progressively increased from 215 ms to 275 ms and 347 ms , respectively , for the occipital , temporal , and frontal lobes ( see Figure 4 ) . The divergence did not reach significance for the small set of 14 parietal lobe electrodes tested ( those showing at least one significant effect , masked or unmasked ) . We then turned to frequency-domain analyses of the intracranial signals . Figure 5A shows a typical single-electrode example of the time-frequency transform applied to our data . The masks alone evoked a very strong sequence of event-related increase in the beta and gamma bands accompanied by alpha decrease , followed by a reversal of this pattern . Subtraction of each mask-only condition from the corresponding word-present condition , however , isolated the ERSP induced by the word alone , as a function of whether it was masked or unmasked . As can be seen in this example , masked words induced a slight increase in gamma power 100–200 ms after the stimulus , whereas unmasked words induced a much bigger effect that lasted throughout the epoch and was accompanied by alpha suppression . To evaluate the generality and significance of such effects , we averaged the time-frequency diagrams across all electrodes ( Figure 5B ) . Statistical comparisons over time-frequency regions of interest , with Bonferroni correction ( see Methods and Materials ) , identified several significant effects . In the 100–200-ms time window , masked words evoked highly significant power changes ( beta suppression: p = 0 . 0004; high-gamma increase: p = 0 . 0005 ) . In this time period , there was no significant difference with unmasked words , confirming that a volley of activation , reflected primarily in a gamma increase , can propagate nonconsciously while being largely unaffected by masking [7] . In the next time window ( 200–300 ms ) , whereas unmasked words created an even larger power increase in the high-gamma band ( p < 10−11 ) and decreases in alpha ( p < 10−8 ) and beta bands ( p = 10−5 ) , masked words induced only small effects of alpha suppression ( p = 0 . 0014 ) and high-gamma increase ( p = 0 . 038 ) . Beta and high-gamma bands showed significantly stronger changes for unmasked compared to masked words ( all p < 0 . 0007 ) . In the subsequent time window ( 300–500 ) , alpha suppression , beta suppression , and high-gamma increase were very strong in the unmasked condition ( all p < 0 . 0003 ) , but altogether absent in the masked condition , creating a significant difference ( all p < 0 . 0002 ) . In summary , masked words induced significant changes in the power spectrum , particularly increases in the high-gamma band , but these induced oscillations quickly dissipated with time , whereas the ERSPs evoked by unmasked words exhibited a greater power and lasted significantly longer . Note that the above analysis was based on the pooling of ERSP results from all electrodes , regardless of their location . We also replicated this ERSP analysis while separating the electrodes as a function of their lobe of origin . Both the high-gamma increase and late alpha and beta suppression specific to the unmasked condition were replicated within each of the four lobes ( see Figure S2 ) . Interestingly , the high-gamma increase peaked earlier in occipital cortex than in temporal , parietal , or frontal , following an approximate posterior to anterior progression ( see Videos S3 and S4 ) . Furthermore , the lobar analysis showed that the early nonconscious effects were confined to the occipital and temporal lobes: the only significant effects were a high-gamma power increase in occipital cortex in the 100–200-ms and 200–300-ms windows ( respectively , p = 0 . 013 and p = 0 . 016 ) , and decreases in alpha ( 200–300 ms , p = 0 . 007 ) and beta ( 100–200 ms , p < 10−3 ) in temporal cortex . In brief , the early ERPS evoked by nonconscious stimuli originated only from occipitotemporal regions , whereas conscious perception was associated with stronger and longer-lasting power changes spreading towards anterior cortical regions . In this respect , analyses of induced high-gamma power yielded conclusions very similar to those derived from iERP analyses . To better evaluate the relation between induced gamma activity and iERPs , we calculated for each segment for which a significant ERP difference was seen , the absolute value of the iERP effect as well as the mean power evoked in the high-gamma range , averaged over the same time period . A significant positive correlation between these two measures was found , for unmasked words ( r2 = 14 . 9% , 241 time segments , p < 10−8 ) and , crucially , for masked words ( r2 = 9 . 3% , 59 time segments , p = 0 . 028 ) . This means that periods in which a high-gamma–band activity is seen are also periods in which a high-voltage difference between word-present and word-absent conditions exists . In brief , high-gamma activity and iERP are correlated measures that both jointly reflect conscious as well as nonconscious processing stages . Indeed , videos of ERP and of high-gamma activity , provided as supplementary online material , present high similar profiles ( see Videos S1 , S2 , S3 , and S4 ) . Spectral changes are complex phenomena that can be sensitive to local as well as global neuronal synchronization of thalamocortical networks [56] . To evaluate the global workspace model's prediction that access to consciousness is associated with long-distance synchronization , we measured the phase synchrony between all electrode pairs . Phase synchrony can occur independently of changes in induced power: it solely evaluates whether oscillations are reproducibly synchronized across two distant sites in the sense that across trials , they exhibit a systematic phase relationship . Figure 6 shows a time-frequency diagram of the intertrial phase coherence ( ITC ) changes induced by the masked and unmasked words , both in an example electrode and in the mean overall electrodes . All statistics were Bonferroni corrected . Statistical analyses revealed no significant coherence changes induced by the masked words . For unmasked words , an increase in beta synchrony in the 300–500-ms time window was highly significant ( t ( 1 , 282 df ) = 7 . 12 , p < 10−10; difference with masked condition , t ( 1 , 282 df ) = 5 . 43 , p < 10−6 ) . It is particularly interesting to note that ( 1 ) this phase synchrony increase was concomitant with a decrease in induced spectral power ( ERSP ) within the same frequency band ( see Figure 7A and 7B ) ; ( 2 ) no phase synchrony increase was detected in the high-gamma band , although in this band , a highly significant increase in induced power had been detected by ERSP analysis . Thus , ERSP and phase synchrony appear to double dissociate , and beta synchrony appears as a highly selective marker of the late phase of conscious access . Figure 8 shows in graphic form the value of the beta coherence increase due to word presence in the critical time window 300–500 ms , separately for unmasked and masked words . Clearly , unmasked words create a more globally synchronous brain state than masked words . The figure makes apparent that this phase coherence analysis is importantly limited by the available electrodes: we can only analyze coherences between electrodes within a given patient , and these tend to be regrouped within a cortical area , thus preventing a thorough analysis of how coherence evolves across distant anatomical sites . For instance , it was not possible to evaluate the prediction that frontal electrodes should cohere more with posterior sites during conscious processing , because in our sample , these two regions were very rarely recorded simultaneously . Still , to probe long-distance connections , we could analyze a subset consisting of electrode pairs in which the two electrodes lie in different hemispheres , thus imposing a long-distance transfer across the corpus callosum . As predicted by global workspace theory , we observed an increase in long-distance interhemispheric beta coherence selective to unmasked words ( t ( 71 ) = 2 . 50 , uncorrected p = 0 . 015 ) . In fact , interhemispheric beta coherence actually decreased when masked words were presented ( t ( 71 ) = 3 . 14 , uncorrected p = 0 . 003 ) , thus creating a strong difference between visible and invisible conditions ( t ( 71 ) = 3 . 83 , p = 0 . 0003 ) . Conversely , Figure 8 suggests that in the masked condition , there might have been a small local increase in beta coherence within posterior occipitotemporal cortices , which might have been missed in our analysis pooling across all electrode pairs . Indeed , when we restricted only to intrahemispheric electrodes lying within occipital cortex or within temporal cortex posterior to y = −20 , we detected a significant increase in beta coherence for masked words during the 200–300-ms time window ( t ( 734 ) = 2 . 34 , uncorrected p = 0 . 02 ) , which ceased to be significant in the 300–500-ms time window ( t = 0 . 41 , not significant [n . s . ] ) . No such increase was seen in other frequency bands , or in other regions ( e . g . , within frontal electrodes ) . Thus , nonconscious word processing resulted in only small and barely detectable transient increases in phase coherence within visual cortex , whereas conscious words yielded a massive increase in long-distance beta coherence A final measure of conscious processing that we evaluated is Granger causality [60 , 63–65] , a mathematical tool that can estimate the causal influence that one electrode site exerts on another . Global neuronal workspace theory predicted that access to consciousness for unmasked words would be accompanied by a massive web of causal relations among distant cortical sites , not seen in the masked condition . Granger causality and phase coherence are similar in that both estimate the correlations among pairs of electrodes , but Granger causality looks for temporal contingencies inaccessible to coherence analyses . In a nutshell , the method estimates whether past samples of electrode j account for a significant amount of variance in electrode i , over and above a simpler “autoregressive” model using only past samples of electrode i ( see [63] for details ) . It is possible for two time series to be strongly phase coherent , yet not causally related ( for instance , two sine waves with constant phase lag and independent noise ) . Thus , Granger causality analysis is not redundant with phase coherence analysis: finding that Granger causality increases during conscious perception , perhaps simultaneously with the beta coherence increase , would provide additional evidence in favor of a large-scale reverberating neuronal assembly linking distant sites . Furthermore , unlike phase coherence , Granger causality is a directional measure: it is possible for electrode j to causally influence i without i causally influencing j ( although it is also possible for two signals to exert mutual causal influences on each other ) . This analysis therefore provided an opportunity to examine the top-down versus bottom-up propagation of activation during conscious and nonconscious processing . As a concrete example , Figure 9 illustrates the causality analysis of a sample electrode pair consisting of one frontal and one occipital electrode . At the time of stimulus presentation , a massive increase in Granger causality is seen in the feedforward , occipitofrontal direction ( Figure 9A , left panel ) and , to a smaller extent in the top-down , fronto-occipital direction ( Figure 9A , right panel ) . Importantly , the curves showing the evolution of the F-test for causality as a function of time exhibit two successive peaks: one early peak is evoked by the masks alone ( 146 ms after mask onset ) , whereas a second peak ( 325 ms after word onset ) is seen only when a word is present and unmasked . As illustrated in Figure 9B , a strong “causal gain” is therefore observed , approximately 200–450 ms after word onset , when the word-present condition is contrasted to the word-absent condition . This effect is seen mostly in the feedforward direction , thus engendering a “causal imbalance” ( higher causal gain in one direction than in the other ) . Similar increases in causal gain and causal imbalance in the unmasked condition were seen in a large set of electrode pairs . To evaluate their statistical significance , we first averaged the causal gains across all electrode pairs and both causal directions , separately for masked and unmasked conditions , and used t-tests to evaluate the significance of changes within three temporal windows ( 100–200 , 200–300 , and 300–500 ms , similar to the ERSP and phase coherence analyses , with Bonferroni correction over the number of windows tested ) ( see Figure 9C ) . A massive increase in mean causal gain was observed during the 300–500-ms window in the unmasked condition ( t ( 1805 ) = 7 . 60 , p < 10−13 ) , but not in the masked condition ( t = 1 . 47 , n . s . ) , resulting in a significantly larger causal gain during conscious than during nonconscious processing ( t ( 1805 ) = 5 . 46 , p < 10−8 ) . The effect was already perceptible in the 200–300-ms window , though it was much smaller ( unmasked: t ( 1805 ) = 3 . 03 , p = 0 . 0075; masked , t = −0 . 86 , n . s . ; difference: t ( 1805 ) = 2 . 89 , p = 0 . 012 ) . No effect reached significance in the 100–200-ms window . Figure 8B illustrates the anatomical distribution of the mean causal gains during the 300–500-ms window . In the unmasked condition , causal relations increased massively among many distant sites , both within the occipitotemporal cortex , between occipitotemporal cortex and distant frontal and insular sites , and across the corpus callosum . By contrast , increases were very scarce in the masked condition and never reached significance even when restricted to posterior electrodes only . With similar methods , we evaluated the statistical significance of changes in the variable of “causal imbalance , ” which is the subtraction of forward causal gain minus backward causal gain in the same electrode pair . This variable evaluated the dominant directionality of causality ( posterior to anterior = feedforward , or anterior to posterior = feedback ) . During the 300–500-ms time window , in the unmasked condition , there was a small imbalance with a higher causality gain in the feedforward compared to the feedback direction ( t ( 1 , 850 ) = 2 . 07 , p = 0 . 039 before Bonferroni correction ) . Although marginally significant , this finding occurred in the predicted late time window and fits with our prior hypothesis that during this time period , and only in the unmasked condition , perceptual information gains access to consciousness and is therefore able to invade anterior areas in a feedforward manner . Indeed , in this time window , the imbalance was not significant for masked targets ( t = −0 . 90 , n . s . ) , creating a difference for unmasked as opposed to masked targets ( t ( 1 , 805 ) = 2 . 09 , p = 0 . 037 before Bonferroni correction ) . Quite surprisingly , however , in the preceding time window ( 200–300 ms , see Figures 9D and S3 ) , there was a significant imbalance in the converse direction ( higher causality in the top-down or feedback ) . This was true only for the masked condition ( t ( 1 , 805 ) = −2 . 66 , p = 0 . 024 , Bonferroni corrected ) , not the unmasked condition ( t = 1 . 11 , n . s . ) , a significant difference ( t ( 1 , 805 ) = 2 . 70 , p = 0 . 021 Bonferroni corrected ) . This unexpected finding , further discussed below , may indicate that in the masked condition , there is a top-down component of attentional amplification , perhaps relating to an unsuccessful effort to identify the masked word .
It is important to keep in mind the limits of the present paradigm . First , when comparing the neural activity elicited respectively by a masked word and by an unmasked word , as was done here , one is actually comparing the correlates of a conscious representation with those of a degraded nonconscious representation . Early differences are therefore not guaranteed to isolate a neural signature of conscious processing per se [23] . Indeed , Del Cul et al . [1] demonstrated that early visual components such as the P1b , N1 , or N2 are affected by masking , yet do not meet the criteria for a strong one-to-one correlation with subjective reports of conscious experience . It therefore seems likely that the early iERP differences that we observed at occipital sites only reflected the early and not-yet conscious effects of masking ( see , for instance , the peak of activity observed on occipital lobe electrodes on Figure 4 , bottom right panel ) . The three additional markers of conscious processing that occurred later on , around 300 ms , synchronous to the activation of frontal regions , appear to be more consistent with the existing literature on electrophysiological correlates of conscious access [1 , 2 , 61] . Indeed , they concur nicely with ERP results obtained in the attentional blink [2 , 61] , a paradigm that may be more appropriate for answering the early versus late debate because it involves an undegraded stimulus that is only made invisible by central competition . A second limitation relates to task confounds . In the present work , some of the observed differences might correspond to differences between a task being performed ( in the conscious condition ) and a task not being performed ( in the nonconscious condition ) . Once a word was consciously perceived , subjects were able to complete the requested semantic emotional categorization task , a complex decision that presumably requires high-level coordination of multiple brain areas . In the masked condition , on the other hand , although subjects still responded to the task , objective performance was by chance . Some of the neurophysiological differences that we reported between masked and unmasked words might therefore be related to task performance rather than to conscious access itself . Mitigating this criticism , however , is the fact that our task appeared to be more difficult for masked words than for unmasked ones , as indicated by longer RTs and by informal subject reports . Indeed , the observed imbalance in the top-down direction for masked words not observed for unmasked words in the 200–300-ms window suggests that nonconscious processing was not free of attentional effects . Nevertheless , an important direction for future research will be to minimize task differences while still comparing neural activity associated with conscious and nonconscious perception . Passive viewing of stimuli could be a valuable instruction , but does not guarantee an absence of task differences ( indeed , it seems obvious that subjects would continue to read and to memorize the conscious words , but not the nonconscious words ) . The task problem is notoriously difficult because performance levels are very rarely identical under nonconscious and conscious conditions . Nevertheless , using “blindsight” conditions with unseen but better than chance performance , matched with seen but degraded performance , Lau and Passingham [24] used fMRI to identify an activation of dorsolateral prefrontal cortex that was specifically associated with conscious experience . This type of paradigm may be usefully complemented with intracranial recordings . We close this issue with a general consideration , initially put forward by phenomenological philosophers [87]: whenever a subject is conscious , he is necessarily conscious of a given mental content . Consciousness is an transitive or “intentional” process ( it is “about” a certain content ) , and therefore it may be illusory to look for a “pure” form of consciousness independent of its particular contents and of the tasks that it affords . Applied to neuroscientific experiments , this property of consciousness implies that when imaging a brain having some conscious experience , we will necessarily observe activations corresponding to a specific conscious content . Nevertheless , satisfactory solutions to overcome this major limitation may exist . For instance , it could be particularly relevant to replicate the present results while manipulating the subjects' task . This track of research may look for the common and invariant correlates of conscious access , irrespective of the task being performed and of the specific mental content being probed . Plausible candidates include late brain-scale activation and beta synchrony , since they were observed in the present results as well as other EEG and MEG studies of conscious perception [1 , 2 , 61] , with various tasks of letter perception , word reading , or digit comparison . The main motivation of our study was to probe the convergence of multiple neurophysiological measures of brain activity in order to define candidate neural signatures of conscious access . Conscious word processing was associated with the following four markers: ( 1 ) sustained iERPs within a late time window ( >300 ms after stimulus presentation ) ; ( 2 ) sustained and late spectral power changes , combining a high-gamma increase , beta suppression , and alpha blockage; ( 3 ) sustained and late increases in long-range phase coherence in the beta range; and ( 4 ) sustained and late increases in long-range causal relations . Our results suggest that in the search for neural correlates of consciousness , time-domain , frequency-domain , and causality-based electrophysiological measures should not be seen as competing possibilities . Rather , all of these measures may provide distinct glimpses into the same distributed state of long-distance reverberation . Indeed , it seems to be the convergence of these measures in a late time window , rather than the mere presence of any single one of them , that best characterizes conscious trials . For instance , masked words also elicited significant iERPs and significant increases in spectral power in the gamma band , contemporary with short-range synchronies in the beta range during the 200–300-ms window . Yet , those words were not consciously accessed , which implies that neither iERPs ( even those recorded from frontal cortex ) , nor gamma-band activity or beta synchrony per se are unique markers of conscious experience . Our results suggest that only late sustained long-distance synchrony and late amplification ( >300 ms ) may be causally related to conscious-level processing . There are yet other mathematical measures derived from nonlinear dynamics that could have been applied to our dataset , such as dimensional activation [88] or neural complexity [21 , 89] , although some of them remain to be made operational in a computationally tractable manner . We consider it likely that these measures would also show a pattern unique to conscious perception . The present work suggests that , rather than hoping for a putative unique marker ( the neural correlate of consciousness ) , a more mature view of conscious processing should consider that it relates to a distributed pattern of brain activation that occurs at a specific level within a complex anatomical and functional architecture , and that it can therefore be reflected by many partially overlapping physiological measures .
Experiments were approved by the Ethical Committee for Biomedical Research of Pitié-Salpêtrière Hospital in Paris ( agreement #99–04 issued on 15 December 2004 ) , participants gave informed consent , and all clinical investigation have been conducted according to the principles expressed in the Declaration of Helsinki . Ten patients ( five men ) suffering from drug-refractory epilepsy were stereotactically implanted with depth electrodes as part of a presurgical evaluation . One patient was implanted twice . Patient ages ranged from 18 to 47 y . Eight patients were right-handed; one man was left-handed; one woman was ambidextrous . Neuropsychological assessment revealed normal or mild impairment in general cognitive functioning: verbal IQ ranged from 65 to 97 and performance IQ from 64 to 120 . On each trial , patients were randomly presented either with a masked word ( 33% ) , a masked blank ( 17% ) , a visible word ( 33% ) , or a visible blank ( 17% ) . A total number of 548 trials were presented for each patient . Words or blanks were presented for 29 ms and were preceded by a visual forward mask ( string of hash signs [#] ) presented for 71 ms , and were followed by a backward mask ( string of ampersands [&] ) presented for 400 ms ( see Figure 1 ) . Words or blanks were made visible by simply removing the backward mask . Although mask removal in itself affected brain activity , in order to discard the activation induced by the masks , we systematically subtracted each word-present condition with its corresponding blank condition , thus isolating the processing path of the masked or unmasked word . In order to maximize attentional engagement and word processing , participants were engaged in a forced-choice task of categorizing each word as threatening or nonthreatening , even on the masked trials . Subjects responded by manually pressing one of two response buttons with the left and right index fingers , and hand response instructions were inverted halfway through the experiment . To prevent automatic stimulus–response learning , we used two distinct sets of 92 French words each for the masked trials and the visible trials , so that the masked words were never seen consciously . In each list , half of the words were threatening ( e . g . , danger , kill ) , with variable frequencies , lengths ( three to eight letters ) and lexical categories ( verbs and nouns ) . The other half included nonthreatening , emotionally neutral words ( e . g . , cousin , see ) , matched for frequency , length , and category . RTs of less than 250 ms or more than 5 , 000 ms were discarded . Median RTs were calculated for each patient and for each condition . RTs were compared across conditions using analysis of variance ( ANOVA ) F-tests . For each patient , we also computed objective discriminability ( d′ ) separately for each of the two masking conditions . We then assessed better-than-chance performance on d′ using both individual and group statistical criteria . First , individual χ2 statistics were calculated for each condition: proportions of correct and incorrect responses to emotional and neutral words were compared with those in the expected random distribution . Second , group analysis was performed using a Z-test testing distributions of d′ against a zero-centered Gaussian . Patients were implanted intracerebrally with depth electrodes , each bearing four to eight recording sites ( Ad-Tech Medical Instruments ) . Recording sites were 2 . 3-mm long , 1-mm diameter cylinders , separated by a distance of 10 mm . The structures to be explored were defined on the basis of ictal manifestations , electroencephalography ( EEG ) , and neuroimaging studies . For each recording site , the Cartesian coordinates ( x , y , z ) were calculated after normalization of the anatomical three-dimensional spoiled gradient recalled ( SPGR ) anatomical cerebral MRI into Talairach space using SPM2 ( Matlab ) . iERPs were digitized at 400 Hz , referenced to the vertex ( Nicolet-BMSI ) . Epochs were then extracted ( −500 ms plus 1 , 000 ms from word onset ) , submitted to automatic artifact rejection ( ±300-mV threshold ) , visually inspected , and notch filtered ( 50 Hz ) using EEGLAB software ( Matlab ) [90] . Recording sites with more than 5% of rejected trials were discarded from analysis . To maximally prevent the measurement of the same electrical signals by multiple sites , for instance , through the common vertex reference , bipolar montages were calculated by subtracting the signals recorded from adjacent sites belonging to the same-depth electrode . This calculation resulted in a total of 176 bipolar recordings ( here called “electrodes” for simplicity ) across ten patients ( corresponding to 11 distinct implantations ) . For each bipolar montage , the Cartesian coordinates ( x , y , z ) were calculated as the medium location of the two adjacent recording sites , resulting in the following anatomical repartition: 55 electrodes in the occipital lobe , 78 in the temporal lobe , 24 in the parietal lobe , and 19 within frontal lobes . For simplicity , in the remnant of this paper , we refer to these recomputed bipolar montages as “electrodes . ” Eighty-two electrodes were recorded within the left hemisphere and 94 in the right hemisphere . Baseline correction ( from −500 to 0 ms before word onset ) was applied , and potentials were averaged from −500 up to 800 ms in order to keep the same temporal windows for ERPs and time-frequency measures . We used a three-step strategy to assess the statistical significance of our results . First , the ERPs of masked words or of unmasked words were compared with those of masked or unmasked blanks by using sample-by-sample t-tests , with a criterion of significance being set at p < 0 . 001 for a minimum of five consecutive samples . Second , we further checked the statistical significance of the observed effects ( number of consecutive samples with p < 0 . 001 on t-test ) through Monte Carlo permutations . This method provides an estimation of type I error rate ( false positives ) by using resampling procedures . Precisely , for each patient and for each electrode showing a significant effect , we computed 20 , 000 random permutations of the observed trials in two surrogate conditions: trials were randomly assigned to one of the two groups , and then for each permutation , we counted the number of surrogate effects satisfying the observed effect anywhere within a time window of 0–1 , 000 ms after stimulus onset . Third , statistical values obtained in the Monte Carlo procedure were corrected for multiple comparisons across electrodes using the false discovery rate procedure [91] . The false discovery rate of a test is defined as the expected proportion of false positives among the declared significant results . Spectral analysis is exquisitely sensitive to small subcritical epileptic events of small amplitude , which are often hard to detect on conventional voltage time series yet can severely impact on time-frequency diagrams . For spectral analysis , our first stage was therefore a visual inspection of the individual time-frequency diagrams of all electrodes . Artifacts , in the form of large broad-band excursions on occasional trials , were detected in 19 additional electrodes that were therefore eliminated from further analysis . For the remaining 147 electrodes , we used the ‘newtimef' and ‘newcrossf' functions of EEGLAB software [90] to estimate and plot , respectively , the mean event-related ( log ) spectral perturbation ( ERSP ) and ITC around word onset . As illustrated in Figures 5 and 6 , ERSP and ITC were first calculated separately within each of the four conditions , then the word-present and word-absent conditions within each masking condition were subtracted from each other to yield a time-frequency plot of masked and unmasked effects . We used a prestimulus baseline correction over −400 to 0 ms , and all results are expressed as increases or decreases relative to this baseline ( in logarithmic decibel scale for ERSP , in coherence units ranging from 0 to 1 for ITC ) . Calculation was based on successive windows of 128 samples ( = 320 ms ) centered on time points −500 to 1 , 000 ms around stimulus onset . In order to avoid windowing artifacts , we only report results over a window of −400 to 800 ms . We used EEGLAB's default algorithm ( fast Fourier transform with Hanning window tapering ) , a frequency range of 0–100 Hz , and a padding ratio of 1 . Prior to computation , individual trials were linearly detrended , and the mean ERP in each condition was removed . Thus , our analysis was aimed at detecting “induced” oscillatory activity , temporally associated with the stimuli but not phase-synchronized with them , while discarding any phase-synchronous “evoked” activity [92–96] . Because of the larger number of available electrodes , times , and frequencies , calculating a corrected-level statistic that evaluates the level of significance of a given time-frequency change in a given electrode poses a severe statistical problem . For simplicity , we therefore only tested significance of effects across groups of electrodes and time-frequency regions of interest . For statistical purposes , we distinguished three time periods ( 100–200 , 200–300 , and 300–500 ms relative to word onset ) and four frequency bands ( alpha = 8–13 Hz; beta = 13–30 Hz; low gamma = 30–50 Hz; and high gamma = 50–100 Hz ) , thus defining 12 time-frequency regions of interest . For each region , we then computed the mean ERSP for each of the 147 electrodes , and then performed t-tests across electrodes to evaluate the significance of the activity induced by masked and unmasked words relative to zero ( i . e . , relative to prestimulus baseline ) and relative to each other . Unless otherwise stated , p-values are Bonferroni corrected for the 12 regions tested , and we only report Bonferroni-corrected effects at p < 0 . 05 . Identical analyses were performed on mean phase synchrony , except that there were now measures from 1 , 283 electrode pairs . We adapted a Matlab software package developed by Anil K . Seth [63] . Because our goal was to establish whether relations of causality between electrodes change in the course of the trial , we performed our analyses on successive time windows of 320-ms width , spaced every 25 ms , with Hanning window tapering . For each such window , for each of the four experimental conditions , and for each pair of distinct electrodes ( i , j ) , the software computed two linear regressions . The first is an autoregressive model that predicts the signal observed on electrode i at time t based on the signal of the same electrode i at previous times t − δt , t − 2δt , … t − nδt . The second model is a causal model that adds as regressors to the above regression the signals of the other electrode j measured at previous times t − δt , t − 2δt , … t − nδt . Here the value of n was fixed at eight retrospective samples ( = 20-ms retrospective time window ) after piloting showed essentially similar results with n = 4 or n = 16 . An associated F-value then probes whether the second causal regression accounted for significantly more variance than the first autoregressive model , indicating a putative causal influence of electrode j on electrode i ( see Figure 9 for an example ) . Seth ( 2005 ) suggests using the logarithm of the F-test as an index of causality strength , but a problem with this index is that it varies with the number of trials , which differed in the word-present and word-absent conditions [63] . We therefore used a causal index independent of run length: the amount of residual variance that was gained by the second model compared to the first , expressed as a percentage of the residual variance in the second model . A general finding ( see , e . g . , Figure 9 ) was that the presentation of the masks alone already created a considerable increase in Granger causality relative to the baseline . Because we were interested in the further increases due to the presence of the words , over and above the effect of the masks , we calculated a “causal gain” by subtracting the causal index in the word-absent condition from the causal index in the word-present condition . Finally , note that the causal gain is still a directional measure: there are distinct causal gains for the effects of electrode j onto electrode i and vice-versa . In the results section , we therefore perform the analysis in two steps: ( 1 ) analysis of the mean causal gain , averaged across the two directions of causality , thus probing the existence of causal relations independently of their direction; ( 2 ) analysis of the “causal imbalance , ” obtained by subtracting the causal gains in the feedforward and feedback directions , thus probing the existence of a preferred direction of causality . For the latter analysis , we defined the feedforward direction as a causal influence of the more posterior electrode onto the more anterior electrode ( in case their y coordinates were identical , we used the z coordinate to define as feedforward an effect of the lowest onto the highest electrode ) . A positive causal imbalance , therefore , indicated stronger causality in the feedforward or bottom-up direction , and a negative causal imbalance indicated a stronger feedback or top-down causality . Statistical significance was evaluated by t-tests over the 1 , 806 electrode pairs , on causal gains averaged over time windows of 100–200 , 200–300 , and 300–500 ms . Unless otherwise state , p-values are Bonferroni corrected for the three windows tested .
|
What is the neural signature of the conscious perception of a visual stimulus ? To address this question , we recorded neural activity directly from the brains of human subjects ( who were undergoing neural surgery for medical reasons ) . This rare opportunity afforded greater spatial and temporal resolution than noninvasive methods used previously to probe the neural basis of consciousness . We compared neural activity concomitant with conscious and nonconscious processing of words by using a visual masking procedure that allowed us to manipulate the conscious visibility of briefly masked words . Nonconscious processing of words elicited short-lasting activity across multiple cortical areas , including parietal and visual areas . In sharp contrast , only consciously perceived words were accompanied by long-lasting effects ( >200 ms ) across a great variety of cortical sites , with a special involvement of the prefrontal lobes . This sustained pattern of neural activity was characterized by a specific increase of coherence between distant areas , suggesting conscious perception is broadcasted widely across the cortex .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"neuroscience"
] |
2009
|
Converging Intracranial Markers of Conscious Access
|
Ebola virus ( EBOV ) , family Filoviridae , emerged in 1976 on the African continent . Since then it caused several outbreaks of viral hemorrhagic fever in humans with case fatality rates up to 90% and remains a serious Public Health concern and biothreat pathogen . The most pathogenic and best-studied species is Zaire ebolavirus ( ZEBOV ) . EBOV encodes one viral surface glycoprotein ( GP ) , which is essential for replication , a determinant of pathogenicity and an important immunogen . GP mediates viral entry through interaction with cellular surface molecules , which results in the uptake of virus particles via macropinocytosis . Later in this pathway endosomal acidification activates the cysteine proteases Cathepsin B and L ( CatB , CatL ) , which have been shown to cleave ZEBOV-GP leading to subsequent exposure of the putative receptor-binding and fusion domain and productive infection . We studied the effect of CatB and CatL on in vitro and in vivo replication of EBOV . Similar to previous findings , our results show an effect of CatB , but not CatL , on ZEBOV entry into cultured cells . Interestingly , cell entry by other EBOV species ( Bundibugyo , Côte d'Ivoire , Reston and Sudan ebolavirus ) was independent of CatB or CatL as was EBOV replication in general . To investigate whether CatB and CatL have a role in vivo during infection , we utilized the mouse model for ZEBOV . Wild-type ( control ) , catB−/− and catL−/− mice were equally susceptible to lethal challenge with mouse-adapted ZEBOV with no difference in virus replication and time to death . In conclusion , our results show that CatB and CatL activity is not required for EBOV replication . Furthermore , EBOV glycoprotein cleavage seems to be mediated by an array of proteases making targeted therapeutic approaches difficult .
Members of the family Filoviridae , Ebola virus ( EBOV ) and Marburg virus ( MARV ) , are the causative agents of viral hemorrhagic fever in central Africa and a major public health threat in their endemic areas . Worldwide concern relates to the importation of infected individuals and the potential use of filoviruses as biothreat pathogens [1] . All current MARV strains belong to the Lake Victoria marburgvirus species , while Ebola virus ( EBOV ) strains are attributed to five different species: Zaire ebolavirus ( ZEBOV ) , Sudan ebolavirus ( SEBOV ) , Côte d'Ivoire ebolavirus ( CIEBOV ) , Reston ebolavirus ( REBOV ) and Bundibugyo ebolavirus ( BEBOV ) [1] , [2] . The species vary in their pathogenicity for humans with ZEBOV being most pathogenic ( up to 90% case fatality rate ) , followed by SEBOV and BEBOV with about 50% and >25% case fatality rates , respectively . CIEBOV and REBOV cause lethal infections in nonhuman primates , but have not yet been associated with fatal human cases [1] , [2] . Although EBOV ( mainly ZEBOV ) and MARV have been extensively studied in vitro and in vivo , today there is neither a licensed vaccine nor treatment available . EBOV entry into target cells is still not fully understood and remains a focus of ongoing research . While a number of attachment factors seem to facilitate EBOV entry [3] , so far no specific cell surface receptor molecule has been identified . Recently , it was shown that the cholesterol transporter Niemann-Pick C1 ( NPC1 ) is required for ZEBOV infection [4] , [5] . NPC1 is a ubiquitously expressed endosomal membrane protein involved in the fusion and fission of endosomes and lysosomes [6] , and its deficiency has been shown to impact on HIV-1 particle release [7] . Its presence in the endosome fits the current EBOV entry model based on macropinocytosis , a cellular pathway proposed to be the main uptake mechanism of EBOV particles into cells [8]–[10] . Previously , Chandran et al . identified the cysteine proteases cathepsin B ( CatB ) and cathepsin L ( CatL ) , which are also present in endosomes , as important factors for ZEBOV entry [11] . According to the current model , cleavage of the ZEBOV glycoprotein ( GP ) by CatB is necessary for exposure of the core receptor-binding domain and fusion machinery , otherwise buried in the GP structure , to initiate fusion of the viral and the endosomal membrane [12]–[14] . Subsequently , the viral genome along with the replication complex is released into the cytoplasm where replication and virus progeny production occur . CatB and CatL are members of a family of 11 human cysteine proteases . Both proteases are highly abundant , broadly expressed and exhibit nonspecific proteolytic activity within lysosomes [15] , [16] . CatB has been associated with TNF-α induced liver damage and seems to play a critical role for the development of pancreatitis [17] , [18] . Despite these facts , catB−/− mice are phenotypically similar to wild-type control mice and fully immunocompetent [18] . CatL is important for epidermal homeostasis and the regulation of the hair cycle and as such catL−/− mice are hairless [19] . Furthermore , CatL is involved in MHC II-mediated antigen presentation in epithelial cells of the thymus [20] . Consequently , catL−/− mice have reduced numbers of CD4+ T helper cells , which are however fully functional . A double knockout mouse lacking CatB and CatL has been generated but is not viable long enough for experimental use [21] , [22] . To determine the importance of cathepsins in viral infections , various inhibitors of endosomal acidification , cathepsin or specifically CatB and CatL activity have been used in vitro . In these studies the role of cathepsins have been demonstrated to be important for the entry of reovirus , SARS-CoV , henipaviruses and ZEBOV [11] , [14] , [23]–[27] . Here we show that while CatB mediates ZEBOV entry in vitro , it is not important for entry and replication of any other EBOV species . We further show that in vivo replication of mouse-adapted ZEBOV ( MA-ZEBOV ) is independent of CatB and CatL , indicating that both of these cathepsins are not required for ZEBOV replication in vivo .
Animals were handled in the Biosafety Level 4 ( BSL4 ) containment space of the Integrated Research Facility ( IRF ) at the Rocky Mountain Laboratories ( RML ) , Division of Intramural Research ( DIR ) , National Institute of Allergy and Infectious Diseases ( NIAID ) , National Institutes of Health ( NIH ) . Research was conducted in compliance with the guidelines of the NIAID/RML Institutional Animal Care and Use Committee ( IACUC ) . The facility , where this research was conducted , is fully accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care International ( AAALAC ) with approved Office of Laboratory Animal Welfare ( OLAW ) assurance ( #A4149-01 ) . Research was conducted under a protocol approved by the IACUC . All procedures were conducted by trained personnel under veterinary supervision , and all invasive clinical procedures were performed while animals were anesthetized . Endpoint criteria , as specified by the IACUC approved scoring parameters , were used to determine when animals should be humanely euthanized . Vero E6 cells and Mouse Embryonic Fibroblast ( MEF ) cell lines lacking cathepsin B , cathepsin L or both cathepsins were cultured in Dulbecco's Modified Eagle Medium ( DMEM ) ( Sigma , St . Louis , MO ) supplemented with 10% FBS , penicillin/streptomycin and L-glutamine in a 37°C incubator , 5% CO2 . VSV ( serotype Indiana ) , SARS ( strain Tor 2 ) , BEBOV [2] , CIEBOV ( strain Tai Forest ) , REBOV ( strain Pennsylvania ) , SEBOV ( strain Boniface ) , ZEBOV ( strain Mayinga ) and the furin knockout and mouse-adapted variants of ZEBOV ( ZEBOV-Fko and MA-ZEBOV , respectively ) [28]–[30] were propagated in Vero E6 cells . The supernatants were cleared of cell debris by centrifugation at 1 , 500×g for 10 min , aliquoted and stored in liquid nitrogen . Viral titers were determined conducting conventional plaque assay or immunoplaque assay . All infectious work with EBOV was performed in the Biosafety Level 4 ( BSL4 ) laboratories at the National Microbiology Laboratory ( NML ) of the Public Health Agency of Canada ( PHAC ) or the IRF , RML , DIR , NIAID , NIH . The endosomal acidification inhibitor Bafilomycin A1 ( BafA1 ) and the inhibitor CA074 ( specific for CatB ) were obtained from Sigma ( St . Louis , MO ) ; CatL-inhibitor V was purchased from Calbiochem ( via EMD Chemicals Inc . , Gibbstown , NJ ) . All inhibitors were dissolved in DMSO and stored at −20°C . Dilutions were prepared in plain DMEM ( no supplements ) prior to every experiment . For foci reduction experiments Vero E6 cells were seeded in 48-well plates the day before infection . Media was removed and cells were incubated for one hour with 50 µl inhibitor in the following concentrations as described previously [11]: BafA1 200 , 100 , 50 , 20 nM; CA074 ( CatB ) 200 , 100 , 50 , 20 µM; CatL-inhibitor V 20 , 10 , 5 , 2 µM . Then 50 µl plain DMEM with 100 focus forming units ( ffu ) virus were added and kept for one hour at 37°C . After three washes with plain DMEM a 1∶1 mixture of 2 . 4% carboxymethyl cellulose ( CMC ) and 2× Minimal Essential Medium ( MEM ) ( Life Technologies , Carlsbad , CA ) supplemented with 4% FBS , penicillin/streptomycin and L-glutamine was added containing the appropriate concentration of inhibitor ( BafA1 100 , 50 , 25 , 10 nM; CA074 ( CatB ) 100 , 50 , 25 , 10 µM; CatL-inhibitor V 10 , 5 , 2 . 5 , 1 µM ) . SARS infected cells were fixed and stained with crystal violet and plaques were counted three days after infection . Four days after infection EBOV inoculated cells were fixed with 10% neutral buffered formalin and removed from BSL4 following standard operating procedures . Subsequently , the cells were permeabilized and foci were stained with a rabbit anti-VP40 antibody ( kindly provided by Y . Kawaoka , University of Wisconsin , Madison , WI ) or a rabbit serum directed against REBOV-NP ( kindly provided by A . Takada , Hokkaido University , Sapporo , Japan ) followed by a FITC-labeled secondary antibody ( Sigma , ST . Louis , MO ) . Foci were counted using a fluorescent microscope ( Carl Zeiss Microimaging LLC , Thornwood , NY ) . Vero E6 cells were seeded in a 24-well plate the day before the experiment . Pretreatment occurred with 200 µl of 100 nM BafA1 , 100 µM CA074 , 10 µM CatL-inhibitor V or no inhibitor for 1 hour . Thereafter , 200 µl ZEBOVwt ( MOI = 1 ) were added and cells were incubated for another hour . Following three washes with plain DMEM , cells in each well were covered with 1 ml DMEM ( supplemented with 2% FBS ) containing 50 nM BafA1 , 50 µM CA074 , 5 µM CatL-inhibitor V or no inhibitor . MEF cell lines were seeded in a 24-well plate the day before the experiment . For infection , 200 µl ZEBOVwt or MA-ZEBOV ( MOI = 1 ) were added and incubated for one hour . Following three washes with plain DMEM , cells in each well were covered with 1 ml DMEM ( supplemented with 2% FBS ) . At time points 0 , 12 , 24 , 48 , 72 and 96 hours post infection 200 µl supernatant were collected from all infected cells , and 200 µl DMEM with 2% FBS containing the appropriate concentration of inhibitor were added back into each well . Samples were stored at −80°C before titration on Vero E6 cells . Groups of C57BL/6 catB−/− , C57BL/6 catL−/− or C57BL/6 ( control ) mice were infected intraperitoneally ( i . p . ) with 10 ffu MA-ZEBOV ( 1 , 000 LD50 ) or 1×105 pfu VSVwt ( serotype Indiana ) and monitored daily for weight loss and signs of disease . On day 3 and 7 post infection 3 mice of each group were euthanized and blood , liver and spleen samples were collected and stored at −80°C for virus titration . Surviving animals were euthanized at day 28 ( study endpoint ) and final serum was collected to determine antibody titers . Vero E6 cells were seeded in 96-well plates the day before titration . Liver and spleen samples were thawed , weighed , homogenized in 10-fold weight/volume of plain DMEM and serial 10-fold dilutions were prepared . Blood samples and supernatant collected from infected cells to determine viral growth for ZEBOVwt and MA-ZEBOV were thawed and serial 10-fold dilutions were prepared . Media was removed from cells and triplicates were inoculated with each dilution . After one hour DMEM supplemented with 2% FBS , penicillin/streptomycin and L-glutamine was added and incubated at 37°C . Cells were monitored for cytopathic effect ( CPE ) and 50% tissue culture infectious dose ( TCID50 ) was calculated for each sample employing the Reed and Muench method [31] . For the detection of ZEBOV-GP specific antibodies in mouse sera , soluble ZEBOV-GP antigen ( ZEBOV-GPΔTM ) was produced and used in an enzyme-linked immunosorbent assay ( ELISA ) as described before [32] , [33] . For the detection of VSV-specific antibodies VSV was propagated in Vero E6 cells . VSV particles were harvested after 48 hrs , purified by centrifugation through a 20% sucrose cushion and treated with 0 . 05% Triton-X100 in PBS prior to storage at −80°C . The ELISA used VSV antigen in a 1∶100 dilution . Sera from mice infected with MA-ZEBOV were inactivated by γ-irradiation as per standard operating protocol . One-way ANOVA was performed using Prism 5 ( Graph Pad Software Inc . ) .
The observation that entry of ZEBOV into Vero E6 cells is CatB- and CatL-dependent was initially described in 2005 by Chandran and colleagues [11] . For further analysis of ZEBOV uptake into Vero E6 cells we performed a different in vitro assay based on foci reduction using the same inhibitors . Cells were seeded and treated with various lysosome and cathepsin inhibitors for one hour , infected with ZEBOV wild-type ( wt ) for one hour , covered with carboxymethyl cellulose containing inhibitors and fixed after incubation for 4 days . The number of foci was determined by immunostaining . Only the inhibition of CatB resulted in significantly reduced virus entry , whereas CatL proteolytic activity did not appear to be obligatory for ZEBOV uptake into Vero E6 cells , although CatL is highly expressed in these cells [27] . SARS-CoV was used as a CatL-dependent control virus [26] to verify the activity of the CatL inhibitor ( Fig . 1 ) . In addition to ZEBOVwt , we also tested a furin cleavage site knockout ZEBOV ( ZEBOV-Fko ) [28] , [29] and the MA-ZEBOV in this assay to determine whether furin cleavage of the glycoprotein or mutations caused by the adaptation process of the virus to mice would influence the requirement for cathepsin cleavage . Both viruses entered Vero E6 cells in a CatB-dependent manner similar to ZEBOVwt and were not affected by the presence of the CatL inhibitor ( Fig . 1 ) . These results show that the uptake of all tested viruses occurred by utilizing the cellular endocytotis machinery as indicated by the inhibitory effect of Bafilomycin A1 ( BafA1 ) , an endosomal acidification inhibitor . Furthermore , proteolytic processing of the ZEBOV-GP by furin is not a prerequisite for cathepsin cleavage in the virus maturation or entry process . In addition , mutations throughout the viral genome acquired during the adaptation process of ZEBOV to the mouse have no influence on ZEBOV uptake . Our data on ZEBOV only partially support previously published observations [11] , but confirm more recent studies claiming that CatB , rather than CatL , plays a role in ZEBOV entry [34] , [35] . Therefore , we studied the role of cathepsin cleavage on virus entry for BEBOV , CIEBOV , REBOV and SEBOV using a representative strain for each EBOV species . Foci reduction assays were performed and evaluated as described above . Infection of all viruses was reduced in the presence of BafA1 ( Fig . 2 ) , confirming that endosomal acidification plays a critical role during the entry process of all these EBOVs . Interestingly , only the maximum concentration of the CatB inhibitor had an effect on BEBOV uptake into Vero E6 cells , whereas CIEBOV , REBOV and SEBOV produced foci independent of the presence of the inhibitor ( Fig . 2 ) . The entry of none of the EBOVs was affected by the CatL inhibitor . This data suggests that CatB is not equally important for the entry of other EBOVs and seems to have a selective effect on ZEBOV and a limited effect on BEBOV entry . After confirmation that CatB has a role in ZEBOV entry we analyzed virus replication in the presence of cathepsin inhibitors . Vero E6 cells were treated with inhibitors ( BafA1 100 nM; CatB 100 µM; CatL 10 µM ) for one hour prior to ZEBOVwt infection ( MOI = 1 ) . After unbound virus was washed away the cells were covered with medium containing inhibitor ( BafA1 50 nM; CatB 50 µM; CatL 5 µM ) , samples were taken at the indicated time points and stored at −80°C for virus titration . Virus titers were determined on Vero E6 cells using a 50% tissue culture infectious dose ( TCID50 ) assay and calculated using the Reed and Muench formula ( Fig . 3A ) [31] . The results demonstrate that ZEBOV replicates to similar titers in the presence of CatB or CatL inhibitor ( difference less than 1 log ) ; the presence of both inhibitors reduced ZEBOV replication by about 1 log . This indicates that the CatB inhibitory effect on entry is likely compensated for by other cellular proteases . However , BafA1 reduces virus growth by more than 2 logs , showing that endosomal acidification is important for efficient ZEBOV entry and replication ( Fig . 1 , 3A ) . In order to further confirm the data we performed ZEBOVwt and MA-ZEBOV infections in MEF cell lines deficient in the expression of CatB , CatL or both cathepsins . The cells were infected for one hour with ZEBOVwt or MA-ZEBOV ( MOI = 1 ) and samples were taken at the indicated time points . Virus titers were determined on Vero E6 cells using a TCID50 assay and calculated using the Reed and Muench formula ( Fig . 3B , C ) [31] . ZEBOVwt and MA-ZEBOV replicated similarly well in the absence of CatB or CatL or both proteases . Although small differences in titer were observed at certain time points , one-way ANOVA did not find statistically significant p values ( Fig . 3 ) . Finally , the in vitro data were verified in vivo using the well-established lethal mouse model for ZEBOV [36] . The catB−/− and catL−/− mice are well characterized [37] and therefore ideal to determine the importance of CatB and CatL for ZEBOV replication in vivo . MEF cell lines obtained from these mice have been used here and in previous studies to demonstrate that mouse cathepsins are functionally similar to human cathepsins and important for ZEBOV-GP-mediated entry [11] , [38] . Groups of catB−/− , catL−/− and control mice were infected with 1 , 000 LD50 of MA-ZEBOV and monitored daily for signs of disease including weight loss ( Fig . 4A , B ) . On day 3 and 7 post-infection 3 mice in each group were euthanized and blood , liver and spleen samples were taken to determine the viral load ( Fig . 5 ) . All animals in the catL−/− group succumbed to MA-ZEBOV infection between days 7 and 9 , similar to most control mice . All but one of the 22 catB−/− mice challenged with MA-ZEBOV succumbed to infection; the one surviving mouse showed signs of disease and recovered ( Fig . 4A , B ) . As previously observed , the infection of control mice is not always uniformly lethal [33] . Here 3 out of 22 control mice developed disease but survived the challenge ( Fig . 4A , B ) . There were no significant differences between viral titers in liver , spleen and blood samples taken on day 3 post MA-ZEBOV infection from the three different mouse strains ( Fig . 5A ) . At day 7 post Ma-ZEBOV infection liver and spleen titers were similar among all three mouse strains but higher in the blood of catB−/− and catL−/− mice compared with wt mice ( Fig . 5B ) . This indicates that disease progression and outcome were similar in all the animals ( Fig . 4 , 5 ) . In order to exclude the occurrence of mutations in the glycoprotein ( GP ) gene of MA-ZEBOV during in vivo replication , we determined the full GP sequence and found no mutations ( data not shown ) . To exclude that catB−/− and catL−/− mice are in general more susceptible to viral infections , groups of mice were infected with recombinant vesicular stomatitis virus ( VSV ) , strain Indiana , and monitored daily for weight loss and signs of illness ( Fig . 4C ) . VSV did not cause disease in any of the mice demonstrating that neither the catB−/− nor the catL−/− mice do possess an increased susceptibility to virus infection . In blood , liver or spleen samples taken on day 3 and 7 post VSV infection no viral RNA was detected ( data not shown ) , indicating that catB−/− and catL−/− as well as control mice were able to efficiently clear the virus . ELISA performed with serum samples of these mice showed that all animals were infected as indicated by the detection of VSV-specific antibodies ( Fig . 4D ) . This data demonstrates that there is no obvious difference between catB−/− , catL−/− and control mice in susceptibility to viral infections and the development of immune responses .
The present study demonstrates that CatB , but not CatL , mediates ZEBOV uptake into Vero E6 cells . This observation is only partially in line with the initial in vitro studies demonstrating both CatB and CatL dependent uptake of ZEBOV [11] , [14] and supports more recent studies showing that only CatB mediates ZEBOV entry into target cells [34] , [35] . In addition , we could demonstrate that post-translational furin cleavage of ZEBOV-GP into the fragments GP1 and GP2 is not a prerequisite for cathepsin processing . This result does not come as a big surprise considering earlier studies reporting that ZEBOV replication and pathogenicity were independent of furin cleavage [28] , [29] . Interestingly , all other EBOV species tested in our study seem to enter Vero E6 cells in a CatB- and CatL-independent manner suggesting that other endosomal proteases might functionally replace CatB in virus entry . This finding is in disagreement with recently published data showing that cell entry of VSV- and HIV-1-based pseudotype particles expressing different EBOV-GPs is CatB-dependent [38] , [39] . Interestingly , one of these studies also showed that cell entry of infectious SEBOV ( strain Gulu ) was CatB- and CatL-independent as we could demonstrate here for a different SEBOV strain ( strain Boniface ) [38] . The discrepancies among our data and some of the previously published reports might be explained by the differences in size and shape of particles as well as the mechanism of particle uptake . HIV-1 particles are largely spherical and the mechanism of uptake is receptor mediated through the interaction of its surface glycoprotein gp160 with CD4 and CCR5 or CXCR4 [40] . VSV particles are short and bullet-shaped and cell uptake occurs via the endocytic pathway [41] . In the infectious ZEBOV context the interactions of GP with VP24 and VP40 ( missing in pseudotype particles ) may further influence the cellular uptake mechanism [14] , altogether suggesting that HIV-1- and VSV-based pseudotype particle entry could be different from those of filovirus particles , which are extremely long and filamentous in shape and mainly utilize macropinocytosis for particle uptake [8]–[10] . In addition , the CIEBOV-GP used to produce VSV-based pseudotype particles in one study lacked the mucin-like domain , which could have had impact on the GP structure and thus might have affected cleavage and entry [38] . In our view this highlights the need for confirmation of data obtained from pseudotype particle systems by live EBOV infections or at least by the use of EBOV-like particles . Finally , cell type and origin may also influence CatB- and CatL-mediated cleavage as studies were performed in different cell lines . For SARS-CoV entry , which is reported to be highly CatL-dependent , it has been shown that expression of the cellular transmembrane protease serine 2 ( TMPRSS2 ) can overcome the block in SARS-CoV infection and replication caused by CatL inhibitors [42]–[45] . Moreover , signaling of toll-like receptor 9 ( TLR9 ) has initially been associated with CatB , CatL and CatK activities [46] , [47] . However , studies using bone marrow derived macrophages and dendritic cells derived from cathepsin knockout mice did not identify a single cathepsin as an essential factor for TLR9 signaling [48] , [49] and rather point towards a role of other endolysosomal proteases , such as asparagine endopeptidase ( AEP ) [50] for activation . Thus , it seems reasonable to speculate that in the absence of CatB and CatL , such as in corresponding knockout mice , other endosomal proteases will mediate EBOV-GP cleavage enabling cathepsin-independent EBOV entry into target cells . Previous reports have shown that EBOV-GPs were also processed by cathepsins in MEFs indicating that mouse CatB and CatL are functionally active [11] , [38] . Here we have demonstrated that , similarly to ZEBOVwt , cell entry by MA-ZEBOV into Vero E6 cells was CatB-dependent but CatL-independent ( Fig . 1 ) . Furthermore , both EBOVs replicated to high titers in MEF cell lines independent of CatB and/or CatL ( Fig . 3B , C ) . Therefore , we used the mouse disease model to investigate the effect of these cathepsins on ZEBOV replication in vivo . C57BL/6 mice ( genetic background ) , CatB or CatL knockout mice ( catB−/− or catL−/− ) did not show an increased susceptibility to viral infection in general as determined here with VSV ( Fig . 4C , D ) . In contrast , all knockout and C57BL/6 mice infected with MA-ZEBOV succumbed to infection with no difference in disease progression and time to death ( Fig . 4A ) or viral loads in liver , spleen and blood ( Fig . 5 ) demonstrating that MA-ZEBOV replication in vivo is CatB- and CatL-independent . In conclusion , our studies indicate that CatB and CatL are not absolutely required for EBOV replication . For yet unknown reasons , CatB seems to play a more considerable role in ZEBOV uptake than it does for any other EBOV species . EBOV seems to have evolved to use a broader spectrum of endosomal proteases to ensure GP cleavage and thus facilitates successful infection of target cells . Therefore , therapeutic approaches targeting single proteases are unlikely to be beneficial to combat EBOV infections .
|
It is currently believed that Ebola virus ( EBOV ) enters cells via macropinocytosis following which , the cysteine proteases cathepsin B and L ( CatB , CatL ) cleave the viral glycoprotein ( GP ) allowing exposure of its core receptor-binding and fusion domain thus facilitating subsequent infection . We studied the effect of CatB and CatL on in vitro and in vivo EBOV replication . Our results demonstrate a reduction of Zaire ebolavirus ( ZEBOV ) entry upon selective inhibition of CatB , but not CatL in cell culture . Interestingly , all other EBOV species enter the cells efficiently when CatB and/or CatL activity is blocked . Moreover , when wild-type ( control ) , catB−/− and catL−/− mice were infected with a lethal dose of mouse-adapted ZEBOV , all animals were equally susceptible to lethal challenge with no difference in virus replication and time to death . Therefore , we conclude that EBOV replication is dispensable of CatB and CatL , and proteolytic processing of GP can also be mediated by other endosomal proteases .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"emerging",
"infectious",
"diseases",
"virology",
"biology",
"microbiology",
"pathogenesis"
] |
2012
|
Cathepsin B & L Are Not Required for Ebola Virus Replication
|
Rift Valley fever virus ( RVFV ) causes a viral zoonosis , with discontinuous epizootics and sporadic epidemics , essentially in East Africa . Infection with this virus causes severe illness and abortion in sheep , goats , and cattle as well as other domestic animals . Humans can also be exposed through close contact with infectious tissues or by bites from infected mosquitoes , primarily of the Aedes and Culex genuses . Although the cycle of RVFV infection in savannah regions is well documented , its distribution in forest areas in central Africa has been poorly investigated . To evaluate current circulation of RVFV among livestock and humans living in the Central African Republic ( CAR ) , blood samples were collected from sheep , cattle , and goats and from people at risk , such as stock breeders and workers in slaughterhouses and livestock markets . The samples were tested for anti-RVFV immunoglobulin M ( IgM ) and immunoglobulin G ( IgG ) antibodies . We also sequenced the complete genomes of two local strains , one isolated in 1969 from mosquitoes and one isolated in 1985 from humans living in forested areas . The 1271 animals sampled comprised 727 cattle , 325 sheep , and 219 goats at three sites . The overall seroprevalence of anti-RVFV IgM antibodies was 1 . 9% and that of IgG antibodies was 8 . 6% . IgM antibodies were found only during the rainy season , but the frequency of IgG antibodies did not differ significantly by season . No evidence of recent RVFV infection was found in 335 people considered at risk; however , 16 . 7% had evidence of past infection . Comparison of the nucleotide sequences of the strains isolated in the CAR with those isolated in other African countries showed that they belonged to the East/Central African cluster . This study confirms current circulation of RVFV in CAR . Further studies are needed to determine the potential vectors involved and the virus reservoirs .
Rift Valley fever ( RVF ) is a viral zoonosis that affects mainly animals but is also found in humans . It is caused by an RNA virus of the Phlebovirus genus ( Bunyaviridae family ) , the genome consisting of three RNA segments: large , medium , and small [1 , 2] . RVFV is transmitted mainly by infected mosquitoes of the Aedes and Culex genuses , but humans can be contaminated by direct contact with blood ( e . g . aerosols , absorption ) or tissues ( e . g . placenta of stillborns from infected animals ) [3 , 4] . The virus was first identified in 1930 during an epidemic that caused deaths and sudden abortions among sheep on the shores of Lake Naivasha in the Great Rift Valley in Kenya [5 , 6] . Since then , the virus has spread to most African countries . The disease occurs in endemic and epidemic forms along the east and south coasts of Africa , in West Africa , in Madagascar [7 , 8] and as far north as Egypt , with a recent outbreak in the Arabian Peninsula [9 , 10] . Severe episodes of RVF have been reported among humans and animals in southern Africa [11 , 12 , 13 , 14] . The animals most frequently infected are sheep and cattle , followed by goats , with heavy economic losses due to abortions and high mortality rates among juvenile animals [15 , 16] . RVFV antibodies have been detected in many wild animal species , including ungulates in Kenya [17 , 18] , bats in Guinea [19] , and small vertebrates in Senegal and South Africa [20 , 21]; however , their role in maintenance of the virus in the ecosystem during inter-epidemic periods and their contribution to amplifying outbreaks remain unknown . RVFV was first isolated in the Central African Republic ( CAR ) in 1969 from a pool of Mansonia africana mosquitoes [22] . It was identified as the causative virus of RVF in 1983 [23 , 24] . RVFV-specific antibodies have since been detected in humans , and 15 strains of RVFV have been isolated from humans and sylvatic mosquitoes in CAR [25 , 26] , although no RVF outbreak has been reported . Current circulation of RVFV in the CAR is unknown , after a gap of two decades without surveillance; however , as animal breeding plays a large part in the economy of the CAR , an epidemic involving humans and animals is possible . We undertook a study to assess the current circulation of RVFV in livestock and humans in the CAR . We also sequenced the genomes of local strains isolated in 1969 from wild mosquitoes and in 1985 from humans in a forested area of the country in order to determine the genetic diversity of these strains which will serve as reference for the future .
The national ethical and scientific committees in charge of validating study design in the CAR approved the study design ( No . 9/UB/FACSS/CSCVPRE/13 ) . The study was described orally before blood samples were collected from human participants , and participants were included only if they gave written consent; for participants aged ≤ 18 years , a parent or guardian provided written informed consent . The informed consent form included a clause permitting use of the participants’ biological specimens for future research . No endangered or sheltered animal species were used in the survey . Verbal consent for testing their animals was acquired from farmers after the objectives of the study had been explained . Once permission was obtained for blood sample collection , an experienced veterinarian bled the animals gently . Samples were taken during the dry season in the first year and at the same sites during the rainy season in the second year . Blood was collected from sheep , cattle , and goats at three localities: the livestock market situated 13 km north of Bangui for cattle , the Ngawi market in Bangui commercial centre , and Ndangala village located 30 km south of Bangui for sheep and goats ( Fig 1 ) . The sex and age of each animal were noted . All animals under 3 years of age were considered juveniles and those over 3 years as adults . Blood samples were also collected from people at risk , such as stock breeders and people working in slaughterhouses and livestock markets . From both animals and humans , venous blood samples were collected in 5-mL Vacutainer tubes ( Becton Dickinson , Franklin Lakes , New Jersey , USA ) , which were placed in a cooler , transported to the laboratory , and centrifuged at 2–8°C for 10 min at 2000 rpm . Each serum sample was separated on collection into two aliquots and stored at –20°C until analysis . Each person who agreed to participate in this study completed an anonymous questionnaire that included demographics . The mosquitoes were collected in sylvan environments in 1969 , identified , and grouped into pools of up to 30 individuals per species per site; samples were stored at –20°C for a maximum of 4 days in the field , transported to the Institut Pasteur in Bangui , and stored at –80°C until virus isolation . The virus was isolated and amplified by four serial passages in suckling mice brain , as described by Saluzzo and colleagues [27] . The brain suspensions were then lyophilized and stored in sealed glass vials at room temperature until use . Serum samples were analysed for the presence of immunoglobulin M ( IgM ) and immunoglobulin G ( IgG ) antibodies to RVF with a SPU-02 RVF IgM and IgG Biological Diagnostic Supplies Ltd ( ELISA ) kit according to the manufacturer’s instructions . Briefly , plates were coated with a recombinant nucleocapsid RVFV antigen diluted 1:1000 in sodium bicarbonate buffer ( pH = 9 . 6 ) , covered with plate seals and incubated at 4°C overnight . Unbound antigen was removed by washing three times for 15 s each with PBS-T . Plates were then blocked with 10% skimmed milk in PBS ( PBS-SM ) at 37°C for 1 h and then washed . Test sera were added in duplicate at a dilution of 1:400 in 2% PBS-SM and incubated for 1 h at 37°C . The plates were washed once more , and HRP-conjugated anti-human IgG antibody , diluted 1:25 000 in 2% PBS-SM , was added to each well and incubated for 1 h at 37°C . After a final wash , chromagenic detection of HRP and absorbance measurement were performed as described previously . Negative and positive control sera were included for each plate . Sera samples were considered positive if their calculated optical density ( OD ) was ≥ 0 . 29 ( net OD serum/net mean OD positive control ) . RVFV was isolated from three samples: two strains isolated in 1969 from mosquitoes ( ArB1986 and HB74P59 ) and one from human serum in 1985 ( HB1752 ) . It was amplified by inoculation into the brains of newborn mice in a laboratory of biosafety level 3 . A brain suspension was prepared , lyophilized , and stored in glass vials at room temperature , and viral RNA was extracted with a QIAmp viral RNA Mini kit ( Qiagen , Valencia , California , USA ) according to the manufacturer’s instructions [28] . The extracted RNA was treated with Turbo DNase ( Life Technologies Inc . , Carlbad , California , USA ) and then retro-transcribed into cDNA with a SuperScript III First Strand Synthesis kit in the presence of random hexamers . The cDNA generated was amplified with Phi29 enzyme as described previously [29] . A fixed amount of amplified DNA was sequenced in an Illumina Hi-seq 2000 sequencer . An average of 30 × 106 single reads with 100 bases was obtained for each sample [28] . The quality of reads was assessed by FastQC , and the sequences were selected according to their quality . All reads corresponding to the mouse genome sequence were filtered by mapping with Bowtie 2 . 0 software on a Mus musculus Mn10 sequence . The viral reads corresponding to the RVFV genome were selected by a similarity approach with BLASTN search tools [30] . All selected viral reads were assembled with Ray software , with k = 25 , to obtain the full-length viral genomes [28] . In order to validate our approach for obtaining the complete sequence of RVFV by high-throughput sequencing , the RNA was extracted from three viral strains ( HB74P59 , HB1752 , and ArB1986 ) , and fragments of the small , medium , and large segments were amplified and sequenced . The sequence obtained from HB74P59 was compared with those obtained previously by Bird et al . in 2007 [31]; no difference was found in three segments obtained by high-throughput sequencing and classical Sanger sequencing . Moreover , no difference was found in the three segments of strain ArB1986 isolated from an Aedes palpalis mosquito and that of a strain isolated in the same city , Loko-Zinga , in 1969 but from another arthropod , Mansonia africana . The distribution of serological results for RVFV was analysed by species , season , and human age and sex and presented as proportions . The effect of each variables on the RVFV positivity was examined using the chi-squared test . P values < 0 . 05 were considered statistically significant . The variables were then put in a Logistic regression model in stepwise manner . The likelihood ratio test was used to compare the model with and without the variable . In case there was no evidence that the variable fitted , it was dropped ( parsimonious model ) . Statistical analyses were performed with STATA software version 11 .
A total of 1271 animals were sampled , comprising 727 cattle , 325 sheep , and 219 goats ( Table 1 ) . The overall seroprevalence in animals was 1 . 9% for anti-RVFV IgM antibodies and 8 . 6% for IgG antibodies . The seroprevalence varied significantly by ruminant . The IgM antibody titre , which indicates recent circulation of RVFV , was 4 . 3% in sheep , 1 . 4% in goats , and 1 . 1% in cattle ( P < 0 . 0001 ) , whereas the IgG seroprevalence was 12 . 9% in sheep , 7 . 8% in cattle , and 5 . 0% in goats ( Table 2 ) . However , the multivariate analysis showed that IgM and IgG seropositivity rates in sheep were higher than other species ( OR = 1 . 8 , 95% CI = 1 . 1–3 . 0 ) ( Table 1 ) . No significant difference in seropositivity was found between male and female animals ( Table 3 ) . The IgG seropositivity rate was 9 . 6% in adults and 5 . 1% in juveniles ( P < 0 . 02 ) , but no significant difference in positivity for anti-RVFV IgM antibodies was found between juveniles and adults ( Table 3 ) . The IgM and IgG seropositivity rates varied significantly according to the origin of the sample . No animals positive for IgG antibodies were found in Ndangala , whereas most of those positive for IgM antibodies were originated from this site , and cattle market are less likely to be IgM positive ( OR = 0 . 1 , 95% CI = 0 . 0–0 . 5 ) ( Tables 1 and 3 ) . All animals with positive IgM were found in the rainy season , and IgG seropositivity was more pronounced during dry season ( Table 3 ) . None of the livestock owners reported cases of abortion or death in the months before sampling that would indicate RVFV infection in their herds . Blood samples were collected from 335 people who were regularly in contact with blood from the animals . The mean age ( ±SD ) was 36 . 3 years ( ±18 . 1 ) , and the sex ratio ( M/F ) was 6 . 0/1 ( 287/48 ) . No evidence of recent RVFV infection ( absence of IgM ) was found in human samples; however , 16 . 7% had evidence of past infection ( IgG alone ) . Of these , 7 . 7% were stock breeders , 6 . 6% were butchers , 1 . 5% were slaughterhouse workers , and 0 . 9% were veterinarians ( Table 4 ) . A higher positivity rate was observed among people over 25 years of age ( P = 0 . 04 ) , and 17 . 8% ( 51/287 ) of males and 10 . 4% ( 5/48 ) of females were positive for IgG ( P = 0 . 29 ) ( Table 4 ) . A third RVFV strain obtained by high-throughput sequencing was isolated in Bangui at the end of December 1984 ( HB1752 ) . Genomic analysis of three segments showed that it was identical to a strain isolated in Bangui 3 months earlier; however , RVFV strains isolated from vectors in 1969 and from human cases in the CAR several decades later had different nucleic sequences , even though they belonged to the same cluster ( Fig 2 ) . The complete genome sequence was made available to GenBank ( HB1752 strain small , medium , and large accession numbers KJ782452 , KJ782453 , and KJ782454 respectively; ArB1986 strain small , medium , and large accession numbers KJ782455 , KJ782456 , and KJ782457 , respectively ) .
This study , the first in the CAR since RVFV was isolated in 1985 , shows that the virus continues to circulate in central Africa . The overall prevalence in animals in this study was lower than that reported in Comoros in 2009 ( 39% in sheep and 33 . 5% in goats ) [32] , in Madagascar in 2008 ( 24 . 7% in small ruminants ) [33] , and in Mozambique ( 35 . 8% in sheep and 21 . 2% in goats ) [34] . The same ELISA kits were used in all these studies , suggesting that the differences are due to climatic factors , entomological parameters , agro-ecological conditions , or sampling strategies . The higher prevalence in sheep is consistent with previous work , indicating that this species is preferentially infected with RVFV [15] . Most infected animals , especially sheep , were found in Ndangala , a rural forested area south of Bangui that has more rainfall than the rest of the country , with lower temperatures and constant humidity in the rainy season , during which time there is little husbandry . The high prevalence observed at this site , with the presence of IgM antibodies , suggests endemic virus circulation , which would be maintained by a sylvatic cycle involving wild animals and mosquitoes , as suggested by Olive et al . [35] . In a previous study in a forested area of the CAR , RVFV was isolated from wild mosquitoes , including Ae . palpalis [26] . In the rainy season , there are many potential breeding sites , which increases the density of vectors and subsequently increases transmission of arboviral diseases such as RVFV . IgM , which indicates recent infection , was present only during the rainy season , but IgG was also significantly associated with the rainy season . These finding are consistent with those in Mauritania and Senegal that indicate that the risk factors for RVF are linked to heavy rainfall and the presence of large temporary masses of surface water [36 , 37] . In a previous study , seropositivity for RVFV was associated with increased numbers of mosquito vectors [38] . Although we did not conduct entomological surveys , recent entomological surveillance for yellow fever identified several species of Aedes mosquito , including Ae . cuminsii , Ae . circumluteolus , and Ae . palpalis [39] , which are known vectors of RVFV [40] . The seroprevalence of RVFV was higher in adult than in juvenile animals . Similar results were reported in Mauritania and Senegal [32 , 33] , supporting the hypothesis of endemic circulation of the virus , as older animals would have longer exposure than younger ones [34] . The presence of IgG in young animals ( < 3 years ) in this study suggests recent circulation of the virus . This result is compatible with the IgM titres in each species , with high titres in samples taken from sheep in Ndangala ( Table 2 ) . The absence of IgG in animals from this region indicates that introduction of the virus south of Bangui is recent . Recent introduction of the virus associated with environmental modifications such as deforestation , population displacement due to the socio-political crisis , and introduction of a new vector competent for RVFV [41] could increase the risk for emergence of an epidemic . As no epidemic of RVF or obvious clinical signs of the disease ( such as abortions ) was observed , the infections were minor or sub-clinical [34] . In a study in Madagascar , circulation of RVFV during the dry season did not result in clinical cases [33] . Viral activity may be maintained in mosquitoes near rivers that do not dry up during the dry season , resulting in a low level of transmission among domestic animals . These observations and the recent epidemics in East Africa illustrate the risk for introduction of pathogenic strains of RVFV to CAR from countries such as South Sudan , which shares a long border with CAR and has had many epidemics and epizootics of RVF . The absence of anti-RVFV IgM antibodies among people regularly exposed to animals would appear to indicate that contact with the virus is uncommon and the public health risk is low . Nevertheless , the presence of IgG among breeders and butchers , who are in contact with the blood of these animals , should alert the authorities to strengthen surveillance of circulation of this virus . Studies to isolate the virus in the vector ( mosquitoes ) and longitudinal studies in sheep , goats , and cattle should be conducted to detect clinical cases , particularly among sheep and herders in Ndangala village , where evidence of current circulation of the virus was found . As most of the inhabitants of the rural areas in which the small ruminants were sampled are farmers who share the same environment as their animals and may have the same exposure to mosquitoes , a study should also be carried out to determine the prevalence of RVFV antibodies and to establish whether RVF occurs regularly in these zones and is thus a neglected cause of morbidity and mortality . Our study was limited to Bangui and neighbouring areas because animals are brought to the capital from all regions of the country . As we were unable to isolate recent strains of the virus , we could not establish the precise geographical origin of the viral strains currently circulating in the CAR . Nevertheless , on the basis of genomic data for old strains , the same strain may be circulating relatively freely in several vectors in a defined geographical area over a long period . Furthermore , although IgG responses persist for several years in continually exposed people , we were unable to obtain a second sample at an interval of 2 weeks and therefore could not demonstrate seroconversion , which would support recent infection . The low IgG titres and high IgM titres found in the Ndangala region suggest recent introduction of the virus into this forested area , probably associated with illegal movement of sheep and goats from the Democratic Republic of Congo . In view of the highly precarious situation in CAR , a large-scale molecular study will not be possible in the short term; furthermore , the socio-political upheaval in the country may change socioeconomic and environmental conditions and the time of infection before the study is conducted . Comparison of the sequences of strains isolated from vectors and humans in the CAR with those isolated in other African counties show that they belong to the East/Central African cluster , confirming RVFV strain exchanges among geographical areas . Propagation of RVFV from East Africa to other regions was noted in Saudi Arabia and Yemen in 2000–2001 [42] and in Chad in 2001 [43] during previous RVF outbreaks .
The results of this first study conducted in both humans and animals in the CAR are important for public health . They shows a high prevalence of RVFV in an area where neither epidemics nor clinical cases of RVF have been reported previously . These results are also important because , in the forested area south of Bangui where there is little husbandry , there is nevertheless low virus noise , which might suggest the presence of a reservoir that has come nearer to human habitats . Unexpectedly , we found low IgM titres in regions of previous intensive animal husbandry , because animal density in these areas has fallen sharply due to the migration of herders in response to the continuing instability in the CAR . Other studies are required to elucidate and measure the environmental risk factors for infection with RVFV in order to predict epidemics , and entomological studies should be performed to identify all the potential vector species to better understand the ecological and climate factors that favor the distribution of RVFV .
|
Rift valley fever virus ( RVFV ) is an arthropod-borne virus that causes serious illness in both animals and humans . RVFV is transmitted by direct contact with infectious tissues or by the bites of infected mosquito species of the Aedes and Culex genuses . Its distribution in tropical forests in central Africa is poorly documented . We assessed the current circulation of RVFV among livestock and humans in the Central African Republic ( CAR ) by detecting anti-RVFV immunoglobulin M ( IgM ) and immunoglobulin G ( IgG ) antibodies in sheep , cattle and goats and in people living in Bangui who were considered at risk . We also sequenced the complete genomes of two local strains , one isolated in 1969 from mosquitoes and one isolated in 1985 from humans living in forested areas . Sheep were the most frequently infected ruminants . IgM antibodies were found only during the rainy season; the frequency of IgG antibodies did not differ according to season . No evidence of recent RVFV infection was found in humans at risk; however , 16 . 7% had evidence of past infection . Phylogenetic analysis showed a perfect match of CAR strains with the East/Central African cluster . Our results confirm current circulation of RVFV in CAR . Further studies should be conducted to determine the vectors involved and the virus reservoirs .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion",
"Conclusion"
] |
[
"invertebrates",
"livestock",
"medicine",
"and",
"health",
"sciences",
"rift",
"valley",
"fever",
"virus",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"ruminants",
"pathogens",
"immunology",
"rna",
"extraction",
"microbiology",
"vertebrates",
"animals",
"mammals",
"viruses",
"rna",
"viruses",
"antibodies",
"insect",
"vectors",
"extraction",
"techniques",
"bunyaviruses",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"sheep",
"proteins",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"disease",
"vectors",
"insects",
"agriculture",
"goats",
"arthropoda",
"biochemistry",
"mosquitoes",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"cattle",
"amniotes",
"bovines",
"organisms"
] |
2016
|
Rift Valley Fever Virus Circulating among Ruminants, Mosquitoes and Humans in the Central African Republic
|
The Hedgehog ( Hh ) signaling pathway plays a key role in cell fate specification , proliferation , and survival during mammalian development . Cells require a small organelle , the primary cilium , to respond properly to Hh signals and the key regulators of Hh signal transduction exhibit dynamic localization to this organelle when the pathway is activated . Here , we investigate the role of Cell Cycle Related kinase ( CCRK ) in regulation of cilium-dependent Hh signaling in the mouse . Mice mutant for Ccrk exhibit a variety of developmental defects indicative of inappropriate regulation of this pathway . Cell biological , biochemical and genetic analyses indicate that CCRK is required to control the Hedgehog pathway at the level or downstream of Smoothened and upstream of the Gli transcription factors , Gli2 and Gli3 . In vitro experiments indicate that Ccrk mutant cells show a greater deficit in response to signaling over long time periods than over short ones . Similar to Chlamydomonas mutants lacking the CCRK homolog , LF2 , mouse Ccrk mutant cells show defective regulation of ciliary length and morphology . Ccrk mutant cells exhibit defects in intraflagellar transport ( the transport mechanism used to assemble cilia ) , as well as slowed kinetics of ciliary enrichment of key Hh pathway regulators . Collectively , the data suggest that CCRK positively regulates the kinetics by which ciliary proteins such as Smoothened and Gli2 are imported into the cilium , and that the efficiency of ciliary recruitment allows for potent responses to Hedgehog signaling over long time periods .
In the mammalian embryo , the Hedgehog ( Hh ) signaling pathway controls cell proliferation , cell survival , and tissue patterning ( cell fate specification and differentiation ) in most tissues such as the developing nervous system , skeleton , skin , and internal organs ( reviewed in [1] ) . The role of Hh signaling in tissue patterning has been most extensively studied in the context of the spinal neural tube . In this system , the Sonic Hedgehog ( Shh ) ligand functions as a morphogen [2]; cells experiencing the strong , intermediate , low , or absent signaling ( in a ventral-to-dorsal order ) adopt the floor plate , motor neuron , V0-V2 interneuron , or dorsal interneuron fates , respectively . The strength of signaling is determined by the amount of ligand cells experience , as well as the duration of that exposure [3] . At the surface of signal responding cells , Hh ligands bind to a complex including the transmembrane receptor Patched ( Ptch1 ) . In the absence of the Hh ligand , Patched inhibits the seven-pass transmembrane signal transducer , Smoothened ( Smo ) . Binding of Hh ligand to Patched relieves its inhibition on Smo , thereby activating it . In turn , active Smo regulates the activity of the Gli family transcription factors , through a process that is not fully understood . Mammals have three Gli genes ( Gli1- 3 ) [4] . Gli1 , whose expression is strongly induced by Hh signaling , encodes a transcriptional activator of Hh targets but it is not required for embryonic Hh signal transduction . Gli2 and Gli3 , however , are both necessary to control embryonic Hh signaling , functioning as activators or repressors of target genes . In the absence of Hh ligands , Gli2 and Gli3 are phosphorylated by a series of kinases , leading to proteasome-dependent processing into shorter repressor forms [5] . Unprocessed ( full-length ) forms are kept in a functionally inactive state due to their physical association with Suppressor of Fused ( Sufu , [6] ) . Once Smo is activated by signaling , the phosphorylation pattern of Gli proteins is altered [7] ) . These changes prevent proteolytic processing ( thus repressor formation ) and they render the Gli proteins active such that they can enter the nucleus and activate the expression of a series of genes , including Gli1 and Ptch1 . In mice , much of this signal transduction occurs in the primary cilium ( reviewed in [8] ) . This relationship appears to be generally conserved among vertebrates including humans [9 , 10] . The primary cilium , present on nearly every cell type in the body , is a microtubule-based appendage surrounded by an extension of the plasma membrane and anchored by the mother centriole [11] . It is assembled primarily through a mechanism termed Intraflagellar Transport ( IFT , reviewed in [11] ) . IFT particles associate with ciliary cargo and microtubule motors , forming trains that are transported in an anterograde direction towards microtubule plus-ends ( at the distal end of the cilium ) by Kinesin II . IFT machinery and cargo to be exported from the cilium are loaded onto IFT particles that are returned to the cell body ( retrograde transport towards microtubule minus-ends ) through cytoplasmic Dynein 1b . A large number of mutations have been identified in mice that block the formation of cilia or alter their assembly and all of these perturbations lead to changes ( increase or decrease ) in the activity of Hh signaling in the embryo [12–16] . Key regulatory components of the Hh pathway show dynamic changes in their localization to cilia , depending on the presence or absence of signals [17 , 18] . In the absence of signals , Patched1 localizes to the cilium , whereas Smo shows little or no localization . Only a small amount of Gli2 , Gli3 , and Sufu localize to the cilium ( where they concentrate at its tip ) . In the presence of signals , ciliary localization of Patched is lost , while Smo , Gli2 , Gli3 , and Sufu become enriched in the structure [17 , 18] . A number of experiments indicate that this change in ciliary localization is important for the factors to execute signal transduction [19 , 20] . Analysis of mouse mutants has shown that proper proteolytic processing of full-length Gli proteins to generate repressor forms is dependent on primary cilia [21] . In parallel , the ability of full-length Gli proteins to undergo differential phosphorylation , dissociate from Sufu , and activate Hh target genes also requires primary cilia [22] . Because of this , defects in ciliogenesis result in ligand-independent up-regulation of Hh targets , down-regulation or absence of ligand-dependent Hh responses , or a combination of the two effects . Studies of a mutation that perturbs ciliogenesis and Hedgehog signaling in the mouse , broadminded ( bromi ) , implicated Cell Cycle-Related Kinase ( CCRK , also known as Cdk20 ) in these processes [23] . CCRK shares homology with members of the Cyclin-dependent Kinase ( CDK ) family and it was originally identified in vertebrates as a Cdk2-activating protein [24] . However , subsequent work showed that CCRK has only a minor role in cell cycle progression [25] , leaving the role of mammalian CCRK unclear . CCRK homologs have been identified and studied in several species ( Chlamydomonas reinhardtii , C . elegans , Leishmania mexicana , and Danio rerio ) and , in all cases , disruption of the gene leads to defects in ciliary length and/or structure [23 , 26–28] . The homolog in the algae Chlamydomonas is called Long Flagella 2 ( LF2 ) . The originally identified mutations in lf2 cause flagella to be significantly longer than normal [26] . However , these mutations were subsequently found to be hypomorphic . In contrast , cells harboring a null allele , lf2-6 , generate flagella with a broad range of lengths ( either too long , too short , or flagella of different lengths on the same cell , [29] ) . Partial knockdown of Ccrk expression in cultured mammalian cells results in ciliary lengthening [30] , as in lf2 hypomorphic Chlamydomonas mutants [26] , but the effect of complete loss of CCRK on Hedgehog signaling and ciliogenesis in mammals has not been previously determined . Here , we find that the role of CCRK in mice is complex , as it positively and negatively controls both ciliary length and the activity of the Hh pathway . The results indicate that CCRK positively regulates import of ciliary cargo including Hh signaling components . They also suggest that efficient flux of Hh signaling components into , and out of , the cilium , is limiting for long-term , but not initial , Hh responses .
We generated a mutation in the mouse Ccrk gene through gene targeting by removing the first two exons of the Ccrk genomic locus ( see Materials and Methods ) . This allele , CcrkKO , should be null , as it lacks promoter sequences , transcriptional and translational start sites , as well as the ATP binding site required for kinase activity ( Fig 1A ) . Western blotting of CcrkKO homozygous mutant embryonic extracts indicated that CcrkKO represents a null allele ( Fig 1B ) . CcrkKO/KO ( hereafter referred to as Ccrk-/- ) mutant embryos exhibited an embryonic phenotype closely resembling that of bromi mutant embryos ( Figs 1C and S1 ) , whereas Ccrk+/- mice showed no apparent phenotype . Ccrk-/- embryos survived until late gestation ( embryonic day 17 . 5 , E17 . 5 ) . Like broadminded mutants , Ccrk-/- mutant embryos exhibited a constellation of defects including exencephaly , mild preaxial polydactyly/limb skeletal defects , reduction of the retinal pigmented epithelium , and cleft palate ( Figs 1C , 1D and S1 ) . We did not detect a noticeable difference in size between homozygotes and their wild-type littermates . In addition to failing to close the neural tube , Ccrk-/- mutants lacked the midline furrow in the midbrain , suggesting that the most ventral , Shh-dependent , floor plate cells were not specified in such mutants . At neural patterning stages ( E10 . 5 ) we found uniform low-level expression of Ccrk throughout the embryo and , as expected due to the mutation induced , Ccrk-/- mutant embryos failed to show detectable expression ( S2 Fig ) . The phenotypic features we observed have been described in other mouse mutants with defective Sonic Hedgehog signaling [31–33] and they are consistent with a role for CCRK in this pathway . We investigated the role of CCRK in Shh-dependent neural patterning through analysis of the Ccrk mutant spinal neural tube ( Figs 2 and S3 ) . Shh signaling specifies a number of cell fates in the neural tube in a concentration and time-dependent manner [3] . The most ventral cell types , the floor plate and p3 interneuron progenitors flanking the floor plate , require the highest/most prolonged extent of Shh signaling . We found that the floor plate was absent in the Ccrk mutant neural tube , even though Shh expression in the notochord was maintained ( Figs 2J and S4 ) . P3 progenitors , marked by Nkx2 . 2 expression , were reduced in number and located ectopically in the ventral midline of the mutants ( Fig 2I ) . The reduction in Nkx2 . 2+ cells was accompanied by ventral expansion of the Pax6-expressing domain in Ccrk mutants ( Figs 2F and S3H ) , as predicted by the established role of Nkx2 . 2 in repressing Pax6 expression . Motor neuron progenitors ( pMN ) and their post-mitotic descendants , marked by Olig2 , HB9 , and Isl1/2 expression , require intermediate levels of Shh signaling while high levels of signaling inhibit their specification . In Ccrk mutants , the motor neuron progenitor domain expanded ventrally across the midline ( Figs 2G , 2H and S3F ) , suggesting that mutant cells exhibit intermediate responses to Shh even when they are positioned ventrally near the source of Shh ligand . Quantitation of the data ( S1 Table ) indicates that such changes are statistically significant . Analysis of patterning over developmental time ( controlled for somite stage ) revealed that the dorsalized phenotype was evident at the earliest stages of fate specification ( 10–13 somite stage ) , although by E11 . 5 ( 45–47 somite stage ) the phenotype had partially recovered ( S4 Fig ) . We also observed dorsal expansion of the Olig2 , HB9 , and Isl1/2 expression domains ( motor neurons and their progenitors ) in the Ccrk mutant neural tube ( Figs 2G , 2H and S3F ) in comparison to controls ( Figs 2B , 2C and S3E ) , which was statistically significant ( S1 Table ) . The mutant phenotype suggests that Ccrk mutant neural progenitors in the midline fail to execute the strongest responses to Shh , whereas mutant cells in ventrolateral regions , dorsal to the normal motor neuron domain , exhibit higher than normal Hh pathway activity . As Shh expression in the notochord appeared normal in Ccrk mutants , we hypothesized that CCRK is required for proper responses to the highest level of Shh signals , rather than production of Shh ligands . To address this , we performed a genetic experiment . Rab23 is an antagonist of the Shh pathway . In Rab23 mutants , spinal neural progenitors adopt ventral cell fates , such as p3 and floor plate independently of Shh ligand or the pathway activator Smoothened [32 , 34] . If CCRK acts in the pathway downstream of Shh ligand and of Smoothened , then the Ccrk mutant patterning phenotype would be genetically epistatic to that of Rab23 mutants . Indeed , E10 . 5 Rab23-/-Ccrk-/- double mutants exhibited loss of floor plate and reduction of p3 progenitor cell numbers . These features were indistinguishable from those in Ccrk single mutants , despite the loss of inhibition from Rab23 ( S5 Fig , data quantitated in S2 Table ) . These data indicate that CCRK is important for regulating Hh pathway activity at a step downstream of Shh , Smoothened , and Rab23 . The slight dorsal expansion of the motor neuron domain in Ccrk mutant suggested that CCRK could also be required for complete repression of Hh pathway activity in the absence of signaling . Smoothened ( Smo-/- ) mutants are unable to respond to Hh ligands [35] . We reasoned that if loss of CCRK leads to mild , ligand-independent derepression of the Hh pathway , some Shh-dependent cell fates would be rescued in Ccrk-/-Smo-/- double mutants . Indeed , we observed restoration of some Shh-dependent cell types , Olig2+ and Nkx6 . 1+ progenitors , in Ccrk-/-Smo-/- double mutants , which were never seen in Smo-/- single mutants ( S6O , S6P , S6K and S6I Fig; data quantitated in S3 Table ) . These data support an additional role for CCRK in complete repression of Hh pathway activity in the absence of Smoothened-dependent signaling . Loss of Gli2 results in a neural patterning phenotype similar to Ccrk mutants ( loss of floor plate and reduction of p3 progenitors ) [33] . If the Ccrk mutant neural patterning phenotype results primarily from loss of Gli2 function , then loss of Gli2 in Ccrk mutants should have little , if any , effect on their phenotype . However , analysis of Ccrk-/-Gli2-/- double mutants revealed a much more dramatic dorsalization of neural tube progenitor identity than seen in either single mutant . ( S7K and S7I Fig ) . Quantitation and statistical analysis of these data are presented in S4 Table . Although Gli3 provides both positive and negative regulation of Hedgehog responses , its repressor function has a relatively dominant role in Hedgehog signaling [36] . In the spinal neural tube , the activator function of Gli3 is dispensable ( due to redundancy with Gli2 ) , whereas its repressor function is uniquely required . If Ccrk mutants retain at least some Gli2 function , then loss of the Gli3 repressor in this background should boost Hedgehog responses to provide phenotypic rescue . Indeed , Ccrk-/-Gli3-/- double mutants showed phenotypic rescue in comparison with Ccrk mutants ( S8I and S8J Fig , data quantified and analyzed in S5 Table ) . Given that Gli2-/-Gli3-/- mutants show the opposite phenotype ( loss of floor plate and p3 cell fates , [37] ) , we suggest that loss of CCRK partially impairs the functions of both Gli2 and Gli3 activators and that Ccrk mutants retain at least some Gli3 repressor function . To complement our genetic analysis , we investigated how the regulation of Gli2 and Gli3 is affected at the protein level by disruption of Ccrk . We performed western blotting for Gli2 and Gli3 using extracts from E9 . 5 wild-type and Ccrk mutant embryos ( Fig 3A ) . Wild-type embryos showed a single full-length species of Gli2 ( 190 kDa ) , whereas in Ccrk mutant extracts we observed an additional , faster migrating species ( Fig 3A ) . This pattern likely reflects a shift in posttranslational modification of Gli2 , such as decreased phosphorylation , leading to diminished Gli2 activator function . We investigated this possibility and found that Gli2 mobility was increased in wild-type samples treated with lambda phosphatase , whereas Gli2 in Ccrk mutant samples showed increased mobility similar to phosphatase treated wild-type samples ( lanes 2 and 4 in Fig 3B ) even without phosphatase treatment . This suggests that Gli2 in mutant cells occurs in a hypo-or non-phosphorylated state . As expected , inclusion of the phosphatase inhibitor Na3VO4 reversed the effect of phosphatase on Gli2 in wild-type samples ( Fig 3B , lane 3 ) . Curiously , Na3VO4 treatment caused Gli2 migration to be retarded in Ccrk mutant samples ( Fig 4B , lane 6 ) . These results were consistently observed in three separate experiments . Our interpretation of the results is that , during incubation of extracts Gli2 is phosphorylated and dephosphorylated by endogenous kinases and phosphatases ( when lambda phosphatase and Na3VO4 were omitted ) . It appears that the balance of these activities favors phosphorylation in wild-type samples and dephosphorylation in Ccrk mutant samples . We also observed profound effects of the Ccrk mutation on the abundance of Gli3 forms . In wild-type embryos , Gli3 is found in two forms: a full-length form ( 185 kDa ) that can be rendered into the activator state ( Gli3Act ) and a proteolytically generated 85 kDa repressor form ( Gli3Rep ) [5] . Shh signaling suppresses formation of Gli3Rep in favor of the full-length form that can function as a transcriptional activator . We found that the full-length form of Gli3 was significantly increased ( ~4-fold ) in Ccrk mutant embryos ( Fig 3C ) . Despite this increase , our genetic data indicate that the Gli3 activator function is impaired in Ccrk mutants given that Ccrk/Gli2 and Gli2/Gli3 double mutants show a similar , strongly dorsalized , neural patterning phenotype ( S7 Fig , [37] ) . The increase in full-length Gli3 levels in the mutants may reflect increased stability of non-functional full-length Gli3 ( see Discussion ) . We also observed a modest ( ~40% ) decrease in the levels of the Gli3 repressor form in Ccrk mutants , which may reflect inefficient proteolytic processing of Gli3 and may explain the weak ligand-independent activity of the Hedgehog pathway in Ccrk mutants indicated by the partial rescue phenotype of Smo/Ccrk mutants ( S6 Fig ) . Collectively , the data indicate that CCRK is required for proper posttranslational modifications of Gli2 and Gli3 , necessary for their functions as transcriptional regulators . To investigate the defect in Hedgehog signaling in Ccrk mutants in more detail , we generated primary mouse embryonic fibroblasts ( pMEFs ) from Ccrk mutant and control embryos . We compared Hh pathway activity between mutant and wild-type cells in unstimulated cells , as well as in cells stimulated with Smoothened agonist ( SAG ) . When exposed to varying concentration of SAG for 24 hours , the direct Hh target gene Gli1 ( monitored by qPCR ) was expressed at a reduced level ( 30–40% ) compared to wild-type cells ( Fig 4A ) . This phenotype was confirmed using wild-type and mutant cells transfected with a Gli binding site/luciferase reporter ( Fig 4B ) . We next investigated immortalized wild-type and Ccrk mutant MEF cell lines . We validated this model by stably transfecting the mutant cell lines with constructs overexpressing either wild-type CCRK or a version with a K33R mutation that is predicted to disrupt kinase activity [29] . The immortalized Ccrk mutant cells showed a deficiency in Hh pathway activity similar to primary Ccrk mutant cells ( Fig 4C and 4D ) . Mutant cells harboring the wild-type construct showed responses that were slightly higher than controls , whereas mutant cells harboring the K33R mutant version showed no evidence of rescue . These data confirmed the role of CCRK in the phenotype and indicated that the kinase activity of CCRK is required for its ability to potentiate Hh pathway activity . Next , we investigated the defect in Hh pathway activity as a function of time of SAG exposure . We stimulated cells with 200 nM SAG for periods ranging from 2 to 24 hours . Interestingly , we found normal or slightly elevated levels of induction of Gli1 expression in Ccrk mutant cells up to 12 hours of SAG exposure in comparison to control cells ( Fig 5 ) . However , after about 12 hours of exposure , wild-type cells exhibited continuously increased levels of expression , whereas the increase was modest for Ccrk mutant cells . A similar phenomenon was observed when cells were stimulated with lower amounts of SAG ( 5 or 20 nM , S9A Fig ) . Ccrk mutant cells showed a qualitatively similar pattern of responses with respect to Ptch1 expression , although expression by mutant cells was slightly reduced at times ≤12 hours with larger differences in expression seen at 14–24 hours ( S9B Fig ) . These results suggest that CCRK function is more important for later periods of signaling than for short-term responses . Because wild-type cells show a continuous increase in Gli1 expression with increased duration of SAG exposure , it is possible that the activity at 24 hours represents prior pathway activity combined with recent pathway activity . If this were the case , the defective responses in Ccrk mutant cells at 24 hours exposure could be due to a loss in their memory of prior activity or to a failure to mount strong responses beyond 8 hours of exposure . To investigate this , we conducted a hysteresis experiment to determine to what extent the cells retain activity from prior signaling events once the inducer is removed for extended periods ( S10 Fig ) . When wild-type and mutant cells were exposed to SAG for 8 hours , they mounted similar responses . When cells were treated with SAG for 8 hours , followed by SAG washout ( plus inclusion of the Smo inhibitor Cyclopamine ) for 24 , 48 , or 72 more hours , wild-type cells showed levels of Gli1 expression similar to , or somewhat higher than , those observed immediately after 8 hours of SAG exposure , indicating that prior exposure to SAG maintains Gli1 expression for an extended period . The Ccrk mutant cells also retained wild-type levels of Gli1 expression 24 hours after removal of inducer , although their responses were somewhat lower than wild-type at 48 and 72 hours . Collectively , these data argue that the failure of Ccrk mutant cells to mount strong responses to exposure times beyond 8 hours lies in their failure of long-term signaling rather than a loss of memory of prior signaling events . Because CCRK homologs in other organisms function in ciliogenesis and because ciliary defects were observed in mutants for the CCRK-binding protein BROMI , we investigated the effects of CCRK disruption on cilia in the mouse . Cilia in the Ccrk mutant embryonic neuroepithelium were generated at roughly normal frequencies but they adopted a shortened , swollen morphology ( Figs 6A and S11 ) . Staining with the ciliary marker Arl13b showed a small ring-like pattern ( Fig 6A ) . Both of these ciliary phenotypes resemble those found in bromi mouse mutants [23] . We next analyzed cilia in Ccrk mutant and control MEFs . Ccrk mutant MEFs generated cilia , albeit with notable differences from wild-type . First , we found that , while both Arl13b and the IFT protein IFT88 localized to Ccrk mutant cilia , they showed an accumulation at the distal end of the cilia in comparison to wild-type cilia ( Fig 6B ) . This phenotype suggests an imbalance between the processes of anterograde and retrograde IFT and is similar to the distal swelling seen in Chlamydomonas flagella null for the Ccrk homolog lf2 [29] . Second , we measured the length of wild-type and Ccrk mutant cilia under steady-state ( serum starved ) conditions ( Fig 6C ) . Although the average cilia lengths were generally similar ( 2 . 6 and 3 . 5 μm , for control and mutant cilia , respectively ) , there was a broader distribution of cilia lengths in the mutant cells than in the control cells ( std . dev . of 0 . 78 vs . 1 . 9 for control and mutant cilia , respectively ) . This finding is reminiscent of the length phenotype of Chlamydomonas lf2 null flagella , which can be much longer or shorter than those of wild-type [29] . We also interrogated the role of CCRK in a cell type that generates longer cilia , IMCD3 , engineered to express a fluorescent reporter , IFT88::YFP [38] , allowing us to monitor intraflagellar transport . We used CRISPR/Cas9-mediated mutagenesis to disrupt Ccrk in IMCD IFT88::YFP cells ( Fig 7A ) . Ccrk mutant IMCD cells generated cilia but , similar to neuroepithelial cells and MEFs , the mutant cilia showed swelling at the distal ends , as determined by scanning and transmission electron microscopy ( Fig 7B ) . This phenotype was accompanied by the accumulation of IFT88::YFP at the distal ends of cilia ( Fig 7C ) . Stable transfection of the Ccrk mutant IMCD cells with a construct expressing wild-type CCRK resulted in complementation/phenotypic rescue . We next monitored the movement of IFT88::YFP particles in live wild-type and Ccrk mutant IMCD cells using TIRF microscopy to generate kymographs from which we determined rates of anterograde and retrograde IFT ( Fig 7C ) . The rates of anterograde transport were not statistically different between wild-type and mutant cells . Retrograde IFT occurred at a slightly slower rate in Ccrk mutant cells compared to controls ( 0 . 62 and 0 . 51 μm/sec , in wild-type and mutant cells , respectively , Fig 7D ) . Although it was difficult to determine IFT frequencies due to movie-to-movie variability , IFT events occurred at normal or somewhat reduced frequencies in the mutant cells ( Fig 7E ) . These data indicate that IFT was only modestly affected in Ccrk mutant cells , a result similar to findings from Chlamydomonas lf2 mutants [39] . While this defect may have some contribution to the Ccrk mutant ciliary length and morphology phenotype , other deficits , such as transport of the ciliary cargo per se , may play a more important role . We determined whether the localization of Hedgehog signaling components to Ccrk mutant cilia was affected in embryonic fibroblasts . We assayed the localization of SuFu , Smo , Gli2 and Gli3 under unstimulated conditions or after 24 hours of stimulation with the Smoothened agonist . Each of these proteins shows enrichment in cilia upon Hh pathway stimulation [17 , 18] . We observed similar patterns of localization for these proteins under the two steady-state conditions in wild-type and Ccrk mutant cells ( Fig 8A ) . However , Gli2 localized to wild-type cilia more frequently than to mutant cilia under unstimulated conditions . We investigated the temporal change in localization patterns by assaying the frequency and amount of ciliary Gli2 localization as a function of time of SAG-mediated pathway induction . Interestingly , the frequency of wild-type cilia with Gli2 localization reached a plateau around 30 min of induction , whereas Ccrk mutant cilia did not show uniform localization until 4 hours of induction ( Fig 8B ) . Similarly , the kinetics of increase in ciliary Gli2 fluorescence upon induction were much slower in Ccrk mutant cells than in wild-type cells . Analysis of Smoothened ciliary localization revealed a similar delay in recruitment in Ccrk mutant cells . These data suggest that , although the rates by which IFT complexes move are relatively unaffected by loss of CCRK , the efficiency of ciliary recruitment and transport of Hedgehog regulatory proteins , and perhaps other cargo , is controlled by CCRK . These findings suggest that CCRK regulates ciliary length , morphology , and function in Hedgehog signaling by promoting the efficiency of ciliary cargo recruitment and transport .
Ccrk mutant cilia show two overt phenotypes: swelling at their distal ends and defective length control . The distal swelling is accompanied by accumulation of the ciliary proteins Arl13b and IFT88 at this end . Such a phenotype is consistent with a defect in retrograde IFT or , possibly , an imbalance in the efficacy of retrograde versus anterograde IFT [14 , 15] . The slightly reduced retrograde IFT rates and frequencies in Ccrk mutant cilia may be responsible for this phenotype . Alternatively , a defect in the IFT turnover process at distal ends of cilia , which involves disassembly/unloading of anterograde IFT trains and assembly/loading of retrograde IFT trains , could lead to such a phenotype . CCRK could regulate the activity of retrograde IFT machinery per se or it could control the ciliary import of proteins that are rate-limiting for retrograde transport . The role of CCRK in the control of ciliary length is less clear , as the means by which cells sense and control their lengths are not well understood . Cilia and flagella initially undergo a lengthening phase , then slow their growth to reach a steady-state length . Many models have been proposed to explain how cells sense their ciliary length , such as the “time-of-flight” or ion current models [40] . Regardless of the sensing mechanism , the cell must execute some form of length control in response to this information . We suggest that differential loading of cargo onto IFT particles provides this control [39 , 41] . According to this hypothesis , during a growth phase , IFT particles are efficiently loaded with cargo needed to assemble cilia at their tips . However , as the cilium begins to reach a steady-state length , cargo loading is significantly down-regulated , leading to reduction in the amount of ciliary cargo being delivered . If cargo , such as tubulin , were rate-limiting for ciliary assembly , this would cause slowing and then stopping of ciliary growth . Thus , ciliary cargo loading may be improperly regulated in Ccrk mutant cells and the variation in ciliary length may be explained by such a defect . In Chlamydomonas IFT trains loaded with tubulin are transported frequently during the growing phase , but this frequency is strongly attenuated when the cilium is at steady state . In lf2 mutants with regenerating flagella , the frequency of transport is greatly reduced compared to controls . In these mutant flagella , at steady state , the frequency is higher than in normal flagella . Reduced transport during growth delivers fewer tubulin subunits to the cilium , resulting in its shortening in some cells . However , in other lf2/Ccrk mutant cells , there may be a delay in reaching steady state , as cargo loading and transport is not appropriately down-regulated , causing those cilia to grow longer than normal . Disruption of Ccrk in the embryo results in phenotypes indicative of defects in Hedgehog signaling , similar to other mouse mutants with elevated or decreased activity of the pathway [31–33] . In Ccrk mutants , cell fates in the most ventral neural tube , that require strong Shh signaling , are either absent ( floor plate ) or reduced in number and mispositioned ( p3 cells ) , while cell fates requiring intermediate levels ( pMN cells ) are concomitantly expanded into ventral territories . This suggests that , while Ccrk mutants are capable of Shh signaling , the strongest responses are defective , with cells executing intermediate responses instead . In addition , the Ccrk mutant pMN domain also expands dorsally , suggesting that cells that would normally adopt fates controlled by low levels of signaling behave as if they experience intermediate levels . The observation that cells requiring weak activity of the pathway are restored in Smo mutants when Ccrk is mutated suggests that CCRK represses the Hh pathway in the absence of signals . Thus , CCRK both potentiates and represses the activity of the Hh pathway , depending on context . The Ccrk/Rab23 epistasis data , along with the SAG in vitro data , indicate that CCRK acts in the response to Shh ( rather than in ligand production ) at the level of Smo or downstream of it . Hh signaling in the mouse embryo is mediated by the combined functions of Gli2 and Gli3 activator/repressor forms . The epistasis and biochemical analyses suggest that all of these activities are partially impaired in Ccrk mutants . Despite the similarity between the Ccrk and Gli2 mutant patterning phenotypes , the Ccrk mutant phenotype is not caused by loss of Gli2 activity alone , as disruption of Gli2 in Ccrk mutants resulted in exacerbation of the phenotype ( stronger dorsalization ) and because disruption of Gli3 in Ccrk mutants resulted in phenotypic rescue , whereas disruption of Gli2 and Gli3 results in a strongly dorsalized patterning phenotype [37] . It is unlikely that the Ccrk mutant phenotype is due to loss of Gli3Act function alone , as Gli3 activator is not required for floor plate specification in the presence of Gli2 . Thus , the most parsimonious explanation for these data is that both Gli2Act and Gli3Act functions are impaired in Ccrk mutants and this has the combined effect of diminishing strong Shh responses . Low-level Smo-independent Hh pathway activity in Ccrk mutants might be due to reduced Gli3Rep function . Indeed , the mutants exhibited reduced amounts of the 85 kDa Gli3Rep form , consistent with diminished proteolytic processing of the Gli3 . However , the residual Gli3Rep in these mutants is likely to be functional since loss of Gli3 can rescue the Ccrk mutant phenotype . The state of Gli2 and Gli3 proteins in Ccrk mutant embryos supports the arguments presented above . The full-length ( FL ) forms of both Gli2 and Gli3 , representing GliAct or GliAct precursor forms , are affected in the mutants . FL-Gli2 changes its electrophoretic migration , consistent with inefficient post-translational modification ( e . g . , phosphorylation ) . The change could represent incomplete PKA phosphorylation at the Pc-g sites in the N-terminus of Gli2 , which potentiate Gli2Act by promoting its dissociation from Sufu [7] . FL-Gli3 shows dramatic accumulation in Ccrk mutants . Although this could , in principle , cause hyperactivation of Shh responses in Ccrk mutants , it may also be a secondary effect of reduced Gli3Act function . This is because release of full-length Gli proteins from their inhibitor , Suppressor of Fused ( Sufu ) [6] , leads to rapid degradation of such proteins , whereas failure to dissociate from Sufu leads to FL-Gli stabilization but loss of its transcriptional functions [42] . Collectively , the genetic and biochemical data suggest that the functions of Gli2Act and Gli3Act forms are impaired , possibly due to their inability to efficiently dissociate from Sufu . In addition , inefficient processing of FL-Gli3 results in diminished Gli3Rep function , which , in turn , increases inappropriate , ligand-independent activation of the Shh pathway . While the ciliary and Hh signaling phenotypes of Ccrk mutants share some similarities with those of other mouse mutants with mutations in ciliary regulators , they differ in important ways from each . Loss of IFT B proteins ( such as Ift172 ) , like loss of Ccrk , results in both diminished high level Hh responses and weak ectopic Hh pathway activity . However , IFT B mutants typically fail to generate cilia ( unlike Ccrk mutants ) and the intermediate/high level Hh responses are more significantly impaired in IFT B mutants than Ccrk mutants [13] . Loss of IFT A proteins ( such as Ift122 ) or cytoplasmic dynein ( Dync2h1 ) result in accumulation of IFT complexes in the distal tips of cilia , similar to Ccrk mutants . However , loss of IFT A proteins only causes increased Hh pathway activity , Dync2h1 mutants only show diminished pathway activity , whereas Ccrk mutants show both diminished and ectopic pathway activity [14 , 15] . Although loss of the ciliary kinesin Kif7 and loss of CCRK each cause deregulation of ciliary length , their effects on Hh pathway activity and neural patterning are largely opposite from one another [43 , 44] . Thus , it appears that the roles of CCRK in ciliary assembly and Hh pathway control are distinct from those of previously studied regulators , although CCRK and such regulators are likely to be functionally connected . We suggest that CCRK indirectly controls the function and biochemical state of Hh signaling components by promoting their import into the cilium . This is supported by our analyses of Smo and Gli2 ciliary localization . In normal cells during the first several hours of SAG exposure , the levels of Smo and Gli2 increase in the cilium over time indicating the rate of ciliary import of these proteins exceeds the rate of their ciliary export during the first 1–2 hours of signaling . In contrast , the rates of Gli2 and Smo ciliary accumulation were substantially delayed in Ccrk mutant cells . This observation may be explained by a reduced rate of ciliary import of Gli2 and Smo . Alternatively , there may be an increase in the rate of ciliary export of these factors in Ccrk mutant cells . While we cannot definitively distinguish between these possibilities at present , we favor the first one . The distal accumulation of Gli proteins in retrograde IFT mutant cilia [15] , coupled with direct interactions between Gli proteins and Kinesin II subunits [45] , suggest Gli proteins are transported by IFT . Because retrograde IFT rates are modestly impaired in Ccrk mutant cilia and the frequency of IFT events is somewhat reduced in such cilia , an increase in Gli2 ciliary export is unlikely . The accumulation of IFT88 at the distal ends of Ccrk cilia is also consistent with impaired retrograde IFT . Although we were only able to assay the change in cargo accumulation during the first few hours after stimulation ( as the levels of proteins reach a steady state thereafter ) , we hypothesize that Ccrk mutant cilia continuously exhibit reduced flux of cargo transport . A reduction in the rate of ciliary transport of Hh pathway components could result in diminished Hh pathway activity . Signaling downstream of Smo results in modification of Gli proteins and dissociation from their inhibitor Sufu , events which are dependent on their transport within cilia [20 , 21] . Thus , diminished influx of pathway activators ( Smoothened and Gli2 ) could result in the generation of fewer activated Gli transcription factor molecules per unit of time . This would cause a cumulative defect in the transcriptional output of the pathway . Examination of Gli1 induction in wild-type cells over time suggests there are at least two phases in the response . During the first phase , between 2 and 12 hours , the increase in target mRNA level is mostly linear with a shallow slope with respect to induction time . This could be explained by the generation of a fixed level production of target gene message per unit of time coupled with perdurance of previously made transcripts . During the second phase , between 12 and 24 hours of exposure , the magnitude of the response increases much more rapidly . The change of slopes is related to the state of the network during each of these periods . The change in the rate of target mRNA increase between the first and second phases may be explained by recruitment of a signal amplification mechanism to further enhance responses during the second phase . While we do not know the mechanistic basis of this amplification , it may rely on the production of more Gli2 which is transcriptionally up-regulated by Shh signaling in fibroblasts [46] . Thus , the time of onset of this amplification would be determined by the time it takes from the initial increase in Gli2 transcription to the generation of active Gli2 protein . Once active Gli2 protein levels accumulate , they may mediate a more robust response to SAG per unit of time . In Ccrk mutant cells , the rate of increase in Gli1 and Ptch1 expression with SAG exposure remains slow and continuous from 2 hours until well beyond 24 hours of induction , suggesting that the sensitization mechanism normally acting during the second phase is defective . The hysteresis experiments indicate that the dampened responses of mutant cells at late time points are caused by a reduction in the generation of activated Gli proteins per unit of time , rather than loss of cellular memory of prior signaling events ( due to , e . g . , shortened half-lives of activated Gli proteins ) . We suggest a model ( S12 Fig ) in which CCRK increases the rate of import of Hh pathway components during pathway stimulation . However , the decreased rate of cargo import in Ccrk mutant cells does not impact the ability of such cells to mount appropriate responses during the first phase of signaling . In contrast , the reduced import rate is limiting with respect to mounting an amplified response during the second phase . If the amplification mechanism occurs through increasing Gli2 protein levels , it is possible that the additional Gli2 critically depends on its efficient transport into and out of the cilium . While our model is testable , it remains unclear how the import of ciliary cargo is regulated and what the role of CCRK in this mechanism may be . Future identification of CCRK substrates should be helpful in this regard .
All experiments on mice were conducted according to institutional and NIH animal welfare regulations . Mice were euthanized according to AVMA approved guidelines ( CO2 inhalation ) . Animal experiments were approved by the IACUC committee at the University of Georgia ( Approval #A2012 12-009-A2 ) . A bacterial artificial chromosome clone containing the Ccrk gene generated from mouse AB2 . 2 embryonic stem ( ES ) cell genomic DNA was used for targeting construct design . The clone was modified by two homologous recombination steps in E . coli using a recombineering method ( Red/ET kit , GeneBridges ) . A LoxP site was inserted 1020 bp upstream of the first exon of Ccrk , and a LoxP-PGK::neoR-LoxP cassette was inserted 430 bp downstream of the end of exon 2 ( within intron 2 ) of Ccrk . A 12 . 5 kb Xba1 fragment ( harboring a 6 . 85 kb 5′ arm of homology and a 2 . 55 3′ arm of homology ) was subcloned into pBluescript . The XbaI fragment was then isolated from the vector backbone and used for homologous recombination in ES cells . AB2 . 2 mouse ES cells were grown on feeder cells , electroporated with the targeting construct , and selected with G418 following standard protocols [47] . Targeted clones were identified by PCR using a forward primer within the neoR sequence and a reverse primer positioned within the genomic sequence 245 bp downstream of the end of the 3′ arm of homology . Targeting was verified by PCR using primers at the 5′ and 3’ ends of homology and euploidy confirmed [47] . Injections of targeted ES cells resulted in several chimeric animals that transmitted the allele , Ccrkfl , through the germline . Ccrkfl/+ males were crossed with Tg ( EIIa-Cre ) /+ deleter females ( Jackson labs ) to generate the null allele of Ccrk . The progeny was screened for Cre-mediated excision between the 5’-most and 3’-most LoxP resulting in the CcrkKO mutant allele , harboring a deletion of exons 1 and 2 and carrying a single LoxP site using primer pairs for the wild-type ( GCAGGAGCCTATGCTGGATCCCTGT and AACGATCTCGCCAGTCTGTGCAGG ) and deleted ( CCTTCCCACGTTAGTGTAGGTTCTTCTC and GGAGGGTGACCACACATGAAAGTCT ) alleles . This allele was confirmed to be protein null by western blotting ( Fig 1B ) . Subsequent genetic experiments were performed using this null allele . A LacZ gene-trap allele of mouse Ccrk was generated from an embryonic stem cell line harboring a LacZ reporter inserted between exons 2 and 3 of Ccrk ( referred to here as Ccrktm1aLacz , also called Cdk20tm1a ( KOMP ) Wtsi ) . This line was obtained from the Knockout Mouse Project , Wellcome Trust Sanger Institute . The Ccrktm1aLacz mouse line was used to characterize the Ccrk expression pattern ( S2 Fig ) . Mice homozygous for this mutation showed a phenotype very similar to the CcrkKO allele described above . Control and Ccrk null MEFs were exposed to varying concentrations of Smoothened agonist ( SAG , Cayman Chemical ) , or to fixed concentrations of SAG for varying time periods , followed by harvesting of RNA , with vehicle ( DMSO ) serving as a negative control . All experiments were performed in triplicate . In hysteresis experiments , MEFs were induced with 200 nM SAG for 8 hours . The SAG was then washed out and the Smoothened inhibitor Cyclopamine ( 10 μM , Calbiochem ) , was added to counteract any signaling that could result from remaining SAG . In other samples , cells were continually exposed to SAG for up to 72 h . The relative , normalized expression of Gli1 was determined using qPCR . RNA was extracted from MEFs using the E . Z . N . A . Total RNA Kit I ( Omega R6834-00 ) . cDNA was synthesized using qScript ( Quanta ) . RT-qPCR was performed using Gli1 ( IDT Mm . PT . 58 . 11933824 ) and Ptch1 ( IDT Mm . PT . 58 . 5305068 ) , and normalized to Actb ( IDT Mm . PT . 58 . 33540333 ) levels using an Applied Biosystems 7500 Real Time PCR System . Normalized values were averaged among triplicates and expressed with standard error . Control and Ccrk mutant MEFs were grown to confluency on glass coverslips and serum starved in 1% FBS for 24 hours to induce ciliogenesis . The cells were processed for staining using standard protocols [54] with the following antibodies: rabbit polyclonal against mouse Smoothened ( kindly provided by Dr . Kathryn Anderson , Sloan Kettering Institute ) , rabbit polyclonal against Arl13b ( Kindly provided by Dr . Tamara Caspary , Emory University ) , mouse monoclonal against acetylated α-tubulin ( Sigma-Aldrich ) , and a mouse monoclonal against γ-tubulin ( Sigma-Aldrich ) . Cells were stained with appropriate secondary antibodies together with DAPI . Images were taken on Zeiss Axioplan 2 microscope with a 100x oil objective and analyzed using AxioVision4 . 6 and FIJI Software . MEFs were exposed to 200 nM SAG for time periods ranging from 30 min to 24 hours and then fixed . Experiments were performed in triplicate . Cells were immunostained as described above to visualize the localization of Gli2 and Smoothened to the cilium after different lengths of exposure to SAG . Fluorescent intensity was determined using ImageJ software . Background fluorescence was used to set thresholds for measuring frequency of cilia showing localization . Values were obtained for 50–200 cilia per condition/coverslip . Ciliary length was determined for control and Ccrk null MEFs through immunolabeling to visualize steady-state cilia length . The axonemes of cilia were labeled with Arl13b and acetylated α-tubulin , whereas the centrosome was labeled with γ-tubulin . Cilia in the focal plane were measured from the base of the cilium where it intersected the centrosome to the tip of the Arl13b staining . Length measurements were performed using Zeiss 4 . 0 and ImageJ Software . E9 . 5–11 . 5 embryos were dissected and prepared for cryosectioning using standard protocols [40] . Cryosections were stained with primary antibodies and fluorescent secondary antibodies ( Jackson Immunologicals ) , followed by counterstaining with DAPI . Primary antibodies were: mouse anti-Shh 5E1 , mouse anti-Nkx2 . 2 , mouse anti-HB9 , mouse anti-Isl1/2 , mouse anti-FoxA2 , mouse anti-Nkx6 . 1 , mouse anti-Pax7 ( all from DSHB at 1:10 ) , and sheep anti-Chx10 ( 1:600 , Exalpha ) , and rabbit anti-Pax6 ( 1:300 , Covance ) . Wild-type and Ccrk homozygous null IMCD3 cells were grown to confluency , followed by fixation in 2% glutaraldehyde/0 . 01% tannic acid and post-fixation in 1% O4Os . Samples were embedded in EPON resin for Transmission Electron Microscopy ( TEM ) or gold-coated ( Structure Probe , Inc ) for Scanning Electron Microscopy ( SEM ) . Sections for TEM were cut using a RMC MT-X microtome ( RMC Products ) and placed on copper grids and visualized using a JEOL JEM-1210 Transmission Electron Microscope ( JEOL , Inc . ) . SEM samples were visualized using a Zeiss 1450EP SEM microscope ( Carl Zeiss MicroImaging ) . Live imaging was performed using a Total Internal Reflection Fluorescence ( TIRF ) microscope [52] at 60X magnification with videos taken at 10 frames per second . Cells were plated on inverted Transwell® membranes ( Corning ) and grown for 7 days as described [53] . The videos were analyzed by generating kymographs using ImageJ . Anterograde and retrograde IFT speeds and frequencies were calculated by taking measurements of the IFT88::YFP tracks seen on the kymographs [52] . In situ hybridization was performed on frozen sections from E9 . 5 embryos according to methods previously described [54] . Whole-mount in situ hybridization was performed according to methods previously described [51] . Riboprobes were synthesized using a kit from Roche according to the manufacturer’s instructions from the following plasmid templates: Shh [55] , Gli1 [56] , Olig2 [31] , and Pax6 [57] . Analysis of Ccrk expression was performed using sense and antisense riboprobes synthesized from a 507 bp PCR-amplified fragment of the 3’ UTR region of mouse Ccrk ( from positions 1169–1676 in mRNA sequence ID: NM_053180 . 2 ) subcloned into pGEM-T Easy ( Promega ) . The Ccrk plasmid was digested with NcoI and transcribed using the Sp6 promoter to generate sense probes or digested with SacI and transcribed using the T7 promoter to generate antisense probe . Cartilage and bone from E16 . 5 Ccrk null and wild-type control embryos were stained with Alcian Blue and Alizarin Red according to standard protocols [47] . Whole-mount β-galactosidase staining of e10 . 5 Ccrktm1aLacZ/+ and control embryos was performed using a standard protocol [51] . Whole embryo protein lysates for immunoblotting were obtained by homogenizing E10 . 5 embryos in modified RIPA buffer with protease inhibitor cocktail ( Roche ) . Protein concentration was determined by the BCA Protein Assay ( Pierce ) . Equal amounts of protein ( 10 μg ) were loaded for electrophoresis . After electrophoresis , proteins were transferred to PVDF membranes ( Millipore ) . Membranes were washed with 1x TBS containing 0 . 2% Tween 20 ( TBST ) and blocked for 1 hour in TBST containing 5% skim milk . Membranes were then incubated overnight at 4°C with a mouse monoclonal anti-Gli3 ( kindly provided by Dr . Susie Scales , Genentech , CA ) , a guinea pig polyclonal antibody against mouse Gli2 ( described in [58] ) , a rabbit polyclonal antibody against CCRK ( kindly provided by Dr . Robert Fisher , Mount Sinai Hospital , NY , NY ) or mouse anti-β actin ( Cell Signaling Technology ) . After washing , membranes were incubated with HRP-labeled secondary antibodies . Signals were detected with a chemiluminescent reagent ( Millipore ) following the manufacturer’s instructions . Signals were quantitated using ImageJ . Whole embryo protein lysates for in vitro phosphatase treatment were prepared from e9 . 5 embryos by clarifying in Triton lysis buffer ( 1% triton X-100 , 50mM HEPES , 150mM NaCl , 2mM DTT ) with protease inhibitor cocktail ( Roche ) for 15 min at 17 , 000 × g at 4°C . Thirty micrograms of extracts plus lambda phosphatase reaction buffer supplied with manufacturer ( NEB , MA ) in 50 ul total reaction volume were incubated at 30°C for 30 min with 400 unit lambda phosphatase . After phosphatase treatment , the reactions were stopped and run on SDS-PAGE , followed by western blotting analysis with a guinea pig polyclonal anti-Gli2 antibody . Wild-type and CcrkKO/KO pMEFs were cultured as described above . Cells were transfected using Lipofectamine 2000 reagent ( Invitrogen ) with the 8xGliBS-Luc [59] , containing eight copies of Gli-binding elements fused to the firely luciferase gene , as well as a Renilla luciferase construct for normalization . Cells were serum starved in OptiMEM ( GIBCO ) for 24 h and treated for an additional 24 h period with varying concentrations of SAG or DMSO ( control ) in serum-free media . Gli reporter activity was assayed using the Dual-Luciferase® Reporter Assay System ( Promega ) . Firefly luciferase activity was normalized to Renilla luciferase activity . Experiments were performed in triplicate . Quantitation of cells positive for markers in the neural tube was determined by counting positive cells per section from two sections per embryo with an average obtained . The positions of dorsal and ventral limits of expression domains were determined by measuring the total ventral-dorsal length of the neural tube in sections and expressing the positions of these expression domain limits as fractions of the total ventral-to-dorsal length . Values were analyzed across 3 embryos per genotype per stage to obtain overall averages , standard deviations , and statistical significance . Statistical analysis was conducted using either Student’s t-tests or Chi-square tests ( for frequencies of Gli2 and Smo ciliary localization ) , computed using StatPlus software ( Analystsoft ) .
|
The importance of cilia in development and disease has become broadly appreciated in recent years due in part to their roles in signal transduction . Despite this attention , crucial aspects of ciliary assembly and function , such as the mechanisms controlling ciliary assembly and the signal transduction events occurring in cilia , remain unclear . Cilia play a central role in sensing and transducing Hedgehog signals in the context of mammalian embryogenesis and in a variety of cancers . Here , we investigate the functions of Cell Cycle Related Kinase ( CCRK ) , which plays an evolutionarily conserved function in the assembly of cilia and flagella . We find that mouse CCRK positively and negatively regulates ciliary length . We show that CCRK controls multiple aspects of Hedgehog signaling in vivo and in vitro by regulating the processing and activities of the Gli transcription factors . Our data suggest that CCRK controls Hedgehog signaling by promoting the efficient ciliary import of core signaling components .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"pathogens",
"enzymology",
"phosphatases",
"developmental",
"biology",
"regulator",
"genes",
"mathematics",
"statistics",
"(mathematics)",
"gene",
"types",
"embryos",
"cellular",
"structures",
"and",
"organelles",
"embryology",
"proteins",
"pathogen",
"motility",
"hedgehog",
"signaling",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"phenotypes",
"cilia",
"virulence",
"factors",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"cell",
"signaling",
"statistical",
"data",
"flagella"
] |
2017
|
Cell Cycle-Related Kinase (CCRK) regulates ciliogenesis and Hedgehog signaling in mice
|
A large number of highly pathogenic bacteria utilize secretion systems to translocate effector proteins into host cells . Using these effectors , the bacteria subvert host cell processes during infection . Legionella pneumophila translocates effectors via the Icm/Dot type-IV secretion system and to date , approximately 100 effectors have been identified by various experimental and computational techniques . Effector identification is a critical first step towards the understanding of the pathogenesis system in L . pneumophila as well as in other bacterial pathogens . Here , we formulate the task of effector identification as a classification problem: each L . pneumophila open reading frame ( ORF ) was classified as either effector or not . We computationally defined a set of features that best distinguish effectors from non-effectors . These features cover a wide range of characteristics including taxonomical dispersion , regulatory data , genomic organization , similarity to eukaryotic proteomes and more . Machine learning algorithms utilizing these features were then applied to classify all the ORFs within the L . pneumophila genome . Using this approach we were able to predict and experimentally validate 40 new effectors , reaching a success rate of above 90% . Increasing the number of validated effectors to around 140 , we were able to gain novel insights into their characteristics . Effectors were found to have low G+C content , supporting the hypothesis that a large number of effectors originate via horizontal gene transfer , probably from their protozoan host . In addition , effectors were found to cluster in specific genomic regions . Finally , we were able to provide a novel description of the C-terminal translocation signal required for effector translocation by the Icm/Dot secretion system . To conclude , we have discovered 40 novel L . pneumophila effectors , predicted over a hundred additional highly probable effectors , and shown the applicability of machine learning algorithms for the identification and characterization of bacterial pathogenesis determinants .
A large number of bacterial pathogens utilize secretion systems for pathogenesis . In these systems , a multi-protein complex is used to translocate a repertoire of proteins , termed effectors , into host cells during infection . These effector proteins were found to be critical for the pathogenicity of numerous pathogens , such as Salmonella enterica , Yersinia pestis , Pseudomonas syringae ( utilizing a type-III secretion system ) [1]–[3] , Legionella pneumophila , Coxiella burnetii Helicobacter pylori , Bordetella pertussis , Agrobacterium tumefaciens , Bartonella henselae ( utilizing a type-IV secretion system ) [4]–[6] , Vibrio cholerae , Mycobacterium tuberculosis , and Pseudomonas aeruginosa ( utilizing other secretion systems ) [7] , [8] , making them prime targets for research of bacterial virulence systems . L . pneumophila is an intracellular γ-proteobacteria , which is the causative agent of Legionnaires' disease: a severe pneumonia-like disease in which the bacteria infect and replicate in human alveolar macrophages [9] . L . pneumophila also infects a wide range of protozoan hosts , which serve as their environmental reservoir [10] , [11] . After internalization by their natural protozoan hosts or by alveolar macrophages , the bacteria are confined to a phagosome and utilize the Icm/Dot type IVb secretion system to subvert host cellular processes [12] , [13] . A large number of L . pneumophila encoded proteins were found to be translocated into the host cell in an Icm/Dot dependent manner [6] , [14]–[16] . Some of the translocated substrates were shown to manipulate host cellular activities , and it is believed that the vast majority of translocated proteins have a functional role during infection . The first L . pneumophila effector , RalF , was identified based on sequence homology to an eukaryotic Guanine Exchange Factor ( GEF ) domain [17] . Since then , a total of 105 genes were identified as effectors using various approaches such as sequence homology to eukaryotic domains and markers for horizontal gene transfer [17]–[21] , interactions with Icm/Dot components [22] , transfer of proteins between bacteria [23] , genetic assays in yeast [24]–[26] , similar regulatory elements [27]–[29] , and a predicted secretion signal [30] . Importantly , the cellular function of most effectors is still unknown , and for most of the effectors examined , their knockout failed to reveal an intracellular growth phenotype [20] , [22] , [26] , [29] , [31] . Moreover , in some cases , knocking-out a family of paralogous effector-encoding genes simultaneously produced no significant growth defect [23] , [32] . These observations suggest the existence of redundancy in terms of effector functionality , or alternatively , some effectors may function only in specific hosts , showing no intracellular growth defect when absent during infection of other hosts . The L . pneumophila Philadelphia-1 genome harbors 3 , 005 open reading frames ( ORFs ) [31] . Identifying ORFs that encode for effector proteins is critical for the understanding of the cellular processes involved in L . pneumophila pathogenesis . In this study , we developed a novel machine learning approach for the identification of effectors . To the best of our knowledge , this is the first attempt to present the task of effector identification as a computational classification problem . Our approach aims to extract features that distinguish effectors from non-effectors . These features are based , in part , on a systematic review of known characteristics of effectors and in part , on the discovery of novel features . These extracted features were then used to train a variety of machine learning algorithms , which produced a list of predicted effectors sorted by their likelihood . We followed up our predictions with experimental validations , using the CyaA reporter system , which led to the discovery of 40 novel L . pneumophila effectors .
We conducted three phases of learning and validation . In each learning phase , we included all the validated effectors known at that time . Specifically , in the second and third phases we added effectors validated in previous phases as well as validated effectors published during the course of this study . Furthermore , the features were updated to reflect the increase in our understanding of effector characteristics and to maximize the information extracted from validated effectors . The effectors discovered in each learning and validation phase are described below . Thus far , analyses of attributes characterizing effectors were based on a limited set of a few dozens of validated effectors [20] , [23] , [28]–[30] . The availability of 145 validated effectors motivated us to perform an in-depth analysis of effector characteristics . Interesting observations were obtained regarding three such characteristics – distribution of effectors in the genome , G+C content of effector encoding genes , and the C-terminal secretion signal .
In this study we have identified 40 new effectors , bringing the total of known effectors to 145 . The high rate of correct predictions suggests that effectors can indeed be clearly distinguished from the remaining ORFs according to the features described in this work . According to available expression array data [38] , these newly discovered effectors are expressed during intracellular growth in amoeba . The expression of 18 effectors was elevated post infection , the expression of 7 effectors was decreased , and the remaining 15 did not change substantially ( less than 1 . 5 fold change ) . The functional role of these effectors during infection has yet to be determined . Regarding the evolutionary origin of effectors , the low G+C content and spatial clustering ( Figure 3 ) support the hypothesis that effectors are often transferred via HGT . These results are in agreement with a recent publication showing that two L . pneumophila effectors were most likely acquired from Protozoa [39] , and with an additional evolutionary study , in which HGT from an amoeba to Legionella was demonstrated [40] . Regarding our 40 newly discovered effectors , the homology to eukaryotes was found to be restricted to specific domains , thus , it is currently impossible to pinpoint the exact evolutionary origins for these genes . The 145 currently validated effectors make L . pneumophila the organism with the highest number of validated effectors . This can provide a lower bound on the percentage of L . pneumophila ORFs that encode for effector proteins – 5% of the total number of ORFs . However , assuming that a large fraction of our predicted effectors are genuine ones , the estimate becomes close to 10% ( about 300 effectors ) , which constitutes an exceedingly large pool of effectors relative to any other known pathogenesis related secretion systems . An important result of our study is the list of additional 126 predicted effectors . Support for the validity of these putative effectors comes from a recent study concerning yeast growth defect of L . pneumophila ORFs [25] . In this study , three new effectors were validated . One of them ( ceg19 ) was independently validated in our study , and an additional ORF ( ceg9 ) is included in our list of putative effectors ( Table S2 ) . In the same study , 12 additional ORFs that showed yeast growth defects and were not experimentally tested for translocation were included in our list of putative effectors . Since effectors were shown to often confer a yeast growth defect phenotype [41] , [42] , this provides additional support for the validity of these putative effectors . Notably , out of our list of 126 predicted effectors , an additional effector ( legK1 ) was also recently validated , see Table S2 [43] . Two previous studies characterized the secretion signal located at the effector C-terminus [30] , [37] . We utilized the high number of validated effectors to statistically analyze the abundance of amino-acid groups at the C-terminus of effectors versus non-effectors and suggested a detailed description of this signal ( Figure 4 ) . We further computed the secretion signal among the 126 predicted effectors . The resulting signal is essentially similar to the one inferred from the list of validated effectors ( Figure S1 ) . This similarity supports both the validity of the putative effectors as well as the biological significance of the secretion signal . It should be noted that the previously suggested secretion signals [30] , [44] were not retained by the classification algorithms in the final learning phase and hence , detecting the secretion signal defined in this work among the putative effectors cannot be attributed to the inclusion of these features when training the classifier . Our machine learning approach is general and thus can be applied to other pathogens . However , the applicability of this approach to other bacteria requires a set of validated effectors , adjustment of the features in order to optimally discriminate these effectors from non-effectors , and an experimental system for prediction validation . Nevertheless , we anticipate that the overall scheme of effector identification will be useful for many pathogenesis systems , when their effector research reaches a proper stage . The pathogenesis system most resembling the one of L . pneumophila is that of C . burnetii , an obligate intracellular pathogen and a potential bioterrorism agent . These bacteria utilize an Icm/Dot type-IVb secretion system and translocate effector proteins in a mechanism similar to that of L . pneumophila , as indicated by the ability of C . burnetii effectors to translocate via the L . pneumophila translocation system [21] . Most of our features , the learning algorithms , and the experimental validation experiments used in this study to identify effectors are applicable with relatively minor changes to the study of C . burnetii pathogenesis , while for more distantly related pathogens ( e . g . , those using type-IVa secretion systems ) further adjustments are required . To summarize , in this work we have developed a combined computational-experimental approach to identify and validate pathogenesis determinants on a genomic scale . We have increased the number of validated effectors by more than 37% and suggested over a hundred putative ones . We have developed a general machine learning scheme for the prediction of effectors , which can be updated when new information becomes available . Finally , this work suggests that our approach is applicable in the identification and characterization of effectors in other bacterial pathogenesis systems .
The following genomes were acquired from the RefSeq database at NCBI ( http://www . ncbi . nlm . nih . gov/RefSeq/ ) : L . pneumophila Philadelphia-1 ( NC_002942 ) ; L . pneumophila Lens ( NC_006366 and NC_006369 ) ; L . pneumophila Corby ( NC_009494 ) ; L . pneumophila Paris ( NC_006365 and NC_006368 ) ; Escherichia coli strain K-12 DH10B ( NC_010473 ) , and Pseudomonas fluorescens Pf-5 ( NC_004129 ) . Two datasets of all human proteins were downloaded from ftp://ftp . ncbi . nih . gov/genomes/H_sapiens/protein/ . The first includes all the human annotated proteins and the second is comprised of ab initio protein predictions of “Gnomon” , an NCBI eukaryotic gene prediction tool ( http://www . ncbi . nlm . nih . gov/genome/guide/gnomon . shtml ) . The genome of Dictyostelium discoideum was downloaded from DictyBase ( http://dictybase . org/ ) . The T . thermophila genome was acquired from the TIGR database ( http://www . tigr . org/tdb/e2k1/ttg/ ) . For each of the three learning phases ( see Results ) , we constructed a dataset of known effectors and a dataset of non-effectors . The size of each non-effector dataset was five folds larger than its corresponding effector dataset . For each learning phase the effector dataset included all effectors known at that time ( published effectors and effectors we validated at previous learning phases ) . Since a dataset of experimentally validated non-effectors is unavailable , we searched for genes that are present in both L . pneumophila and E . coli , under the premise that such genes are most likely not related to the pathogenicity of L . pneumophila and are thus expected to be non-effectors . Specifically , BLAST-P similarity scores were computed for each L . pneumophila protein against the proteins of E . coli . An L . pneumophila protein was defined as a non-effector if it has a hit from E . coli with an E-value lower than 10−20 and sequence similarity higher than 50% . Additionally , the proteins that constitute the Icm/Dot secretion system were included in the non-effector datasets . A full list of all the effectors and non-effectors used in each phase is given in Table S1 . Each ORF in the L . pneumophila Philadelphia-1 genome was described using a vector of features . The features used for each learning phase are detailed in Table S2 and are summarized in Table 1 . Machine learning algorithms were performed using the WEKA package [47] . The following classification algorithms were tested: Naïve Bayes , Bayesian networks , SVM ( SMO ) , Neural networks ( Multilayer perceptron ) , and a Voting algorithm that is based on these four algorithms . Feature selection was performed using a “Wrapper” to find the best performing features for each one of the algorithms , using hill-climbing search algorithms . The classifiers were trained on datasets in which the ratio of effectors to non-effectors was 1∶5 . Specifically , as more effectors were included in the second and third phases , the number of non-effectors was increased accordingly to maintain this ratio . The classification performance was evaluated on the train datasets for each classifier separately . Classification performance for each classifier was evaluated using 10-fold cross validation , i . e . , 90% of the training data were randomly chosen and used to tune the parameters of each classifier , and the remaining 10% were used to evaluate the classifier performance [47] . The performance score is measured in terms of AUC , which accounts for both the fraction of true positives ( correctly classified effectors ) and false positives ( ORFs erroneously classified as effectors ) . Since the performance depends on the division of the training data , the procedure is repeated 10 times , so that each 10% is used once to evaluate performance . Classifier accuracy is defined as the average over these 10 repeats . The classifier with the highest average AUC was used at each learning phase to predict effectors . When we evaluated the classification performance of each feature group separately , a dataset in which the ratio of 1∶1 between effectors and non-effectors was used . This was done to avoid artificial high performance stemming from the excess of non-effectors in the training data . The computer code used to implement the machine learning scheme described here is available in http://www . tau . ac . il/~talp/LegionellaMachineLearning . The L . pneumophila strains used in this study were L . pneumophila JR32 , a streptomycin-resistant , restriction-negative mutant of L . pneumophila Philadelphia-1 , which is a wild-type strain in terms of intracellular growth [48] and GS3011 an icmT deletion mutant [49] . The E . coli strain used was MC1022 [50] . Bacterial media , plates , and antibiotic concentrations were used as described previously [51] . The plasmid pMMB-cyaA-C [28] was used for the cloning of all the cyaA fusions constructed . All the genes examined were amplified by PCR using a pair of primers containing suitable restriction sites at the 5′ end . The PCR products were subsequently digested with the relevant enzymes , and cloned into the pMMB-cyaA-C vector to generate plasmids . Table S5 includes the pair of primers , the enzymes used for digestion , and the generated plasmids . The generated plasmids were sequenced to verify that no mutations were introduced during the PCR . Furthermore , the formation of a fusion protein with a proper size was validated by Western analysis using the CyaA antibody 3D1 ( Santa Cruz Biotechnology , Inc . ) . Others and we have utilized the CyaA translocation assay to validate effector proteins [17] , [27] , [28] , [30] , [33] . Specifically , differentiated HL-60-derived human macrophages plated in 24-well tissue culture dishes at a concentration of 2 . 5×106 cells/well were used for the assay . Bacteria were grown on ABCYE ( ACES buffered charcoal yeast extract ) plates containing chloramphenicol ( Cm ) for 48 h . The bacteria were scraped off the plate , calibrated to OD600 of 0 . 2 , and 20 µl of these bacteria were spotted on an ABCYE plate containing Cm and 1 mM isopropyl-ß-D-thiogalactopyranoside ( IPTG ) and grown for 20 h . The bacteria were then scraped off the plate and calibrated in order to result with a multiplicity of infection ( MOI ) of 5 during infection . Cells were infected with bacteria harboring the appropriate plasmids and the plates were centrifuged at 180 g for 5 minutes followed by incubation at 37°C under CO2 ( 5% ) for 2 h . Cells were then washed twice with ice-cold PBS buffer ( 1 . 4 M NaCl , 27 mM KCl , 100 mM Na2HPO4 , 18 mM KH2PO4 ) and lysed with 200 µl of lysis buffer ( 50 mM HCl and 0 . 1% Triton X-100 ) at 4°C for 30 minutes . Lysed samples were boiled for 5 minutes , centrifuged for 10 minutes , and the supernatants were neutralized with NaOH . The levels of cAMP were determined using the cAMP Biotrak enzyme immunoassay ( EIA ) system ( GE-healthcare ) according to the manufacturer's instructions . The presence of the CyaA fusion proteins was detected by Western blot , using monoclonal antibody anti-CyaA 3D1 ( Santa Cruz Biotechnology , Inc . ) diluted 1∶500 and goat anti-mouse IgG conjugated to HRP ( Jackson Immunoresearch Laboratories , Inc . ) diluted 1∶10 , 000 .
|
Many pathogenic bacteria exert their function by translocating a set of proteins , termed effectors , into the cytoplasm of their host cell . These effectors subvert various host cell processes for the benefit of the bacteria . Our goal in this study was to identify novel effectors in a genomic scale , towards a better understanding of the molecular mechanisms of bacterial pathogenesis . We developed a computational approach for the detection of new effectors in the intracellular pathogen Legionella pneumophila , the causative agent of the Legionnaires' disease , a severe pneumonia-like disease . The novelty of our approach for detecting effectors is the combination of state-of-the-art machine learning classification algorithms with broad biological knowledge on effector biology in a genomic scale . Applying this method , we detected and experimentally validated dozens of new effectors . Notably , our computational predictions had an exceedingly high accuracy of over 90% . In analyzing these effectors we were able to obtain new insights into the molecular mechanism of the pathogenesis system . Our results suggest , for the first time , that over 10% of the Legionella genome is dedicated to pathogenesis . Finally , our approach is general and can be utilized to study effectors in many other human pathogens .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"infectious",
"diseases/bacterial",
"infections",
"computational",
"biology/systems",
"biology",
"computational",
"biology/genomics",
"infectious",
"diseases/respiratory",
"infections"
] |
2009
|
Genome-Scale Identification of Legionella pneumophila Effectors Using a Machine Learning Approach
|
Exonuclease-mediated RNA decay in plants is known to be involved primarily in endogenous RNA degradation , and several RNA decay components have been suggested to attenuate RNA silencing possibly through competing for RNA substrates . In this paper , we report that overexpression of key cytoplasmic 5’–3’ RNA decay pathway gene-encoded proteins ( 5’RDGs ) such as decapping protein 2 ( DCP2 ) and exoribonuclease 4 ( XRN4 ) in Nicotiana benthamiana fails to suppress sense transgene-induced post-transcriptional gene silencing ( S-PTGS ) . On the contrary , knock-down of these 5’RDGs attenuates S-PTGS and supresses the generation of small interfering RNAs ( siRNAs ) . We show that 5’RDGs degrade transgene transcripts via the RNA decay pathway when the S-PTGS pathway is disabled . Thus , RNA silencing and RNA decay degrade exogenous gene transcripts in a hierarchical and coordinated manner . Moreover , we present evidence that infection by turnip mosaic virus ( TuMV ) activates RNA decay and 5’RDGs also negatively regulate TuMV RNA accumulation . We reveal that RNA silencing and RNA decay can mediate degradation of TuMV RNA in the same way that they target transgene transcripts . Furthermore , we demonstrate that VPg and HC-Pro , the two known viral suppressors of RNA silencing ( VSRs ) of potyviruses , bind to DCP2 and XRN4 , respectively , and the interactions compromise their antiviral function . Taken together , our data highlight the overlapping function of the RNA silencing and RNA decay pathways in plants , as evidenced by their hierarchical and concerted actions against exogenous and viral RNA , and VSRs not only counteract RNA silencing but also subvert RNA decay to promote viral infection .
Among major infectious pathogens , viruses are the obligate intracellular agents that can infect all types of life forms from bacteria to plants and animals , and exclusively multiply in the host cells . The vast majority of known plant viruses are positive-sense single-stranded ( +ss ) RNA viruses that typically have a relatively small genome encoding no more than a dozen proteins . Since the genomic RNAs of +ss viruses are similar to the endogenous mRNAs , the regulation machinery of cellular RNA metabolism in the infected plants is unavoidably involved in viral infection . An essential RNA regulation mechanism is RNA silencing , which is an evolutionarily conserved and sequence-specific immunity response and triggered by double-stranded RNAs ( dsRNAs ) [1–3] . In plants , RNA silencing is recognized as a central antiviral pathway [2 , 3] . During viral infection , dsRNAs may originate from viral genome replication , via discrete intramolecular pairing within viral genomic RNA or de novo synthesis by the endogenous RNA-dependent RNA polymerases ( RdRps ) [4] . These dsRNAs are processed into 21- to 24- nucleotide ( nt ) small interfering RNA ( siRNA ) duplexes by the dsRNA-specific Dicer-like ( DCL ) proteins that contain RNase III activity . The resulting siRNAs duplexes are incorporated into the RNA-induced silencing complex ( RISC ) and guide the sequence-specific degradation of the target viral RNA by the Argonaute ( AGO ) protein that possesses RNaseH-like endonuclease activity [4 , 5] . As a consequence of the co-evolutionary arms race between plants and viruses , viruses have evolved to suppress or evade RNA silencing by encoding viral suppressors of RNA silencing ( VSRs ) . In the past twenty years , VSRs have been identified from almost all plant virus genera [5] . In addition to RNA silencing , RNA decay is another crucial pathway that regulates RNA turnover [6 , 7] . RNA decay or exonucleolytic RNA turnover is a 5’–3’ and 3’–5’ exoribonuclease-dependent , ubiquitous mechanism in eukaryotic cells by which mRNA molecules are enzymatically degraded [6] . RNA decay is essential for both mRNA quantity and quality control [6] . The degradation process is initiated by deadenylation to progressively remove the 3’ poly ( A ) tail followed by exosome complex-mediated 3’–5’ cleavage or decapping and subsequent exoribonuclease ( XRN ) -mediated 5’–3’ decay [6 , 7] . Deadenylation is catalysed by the conserved poly ( A ) -specific ribonuclease ( PARN ) as well as by the conserved carbon catabolite repressor 4 ( CCR4 ) complex [8–10] , and the removal of the 5’ cap structure is through concerted action of a set of conserved decapping proteins ( DCP ) [6 , 7] . In Arabidopsis thaliana , DCP1 , DCP2 , DCP5 , VARICOSE ( VCS ) and possibly DEA ( D/H ) -box RNA helicase 1 ( DHH1 ) constitute the decapping complex [11] . Deadenylation or decapping is a prerequisite for most RNA to be degraded by the 3’–5’ exoribonuclease exosome complex or by 5’–3’ XRN exoribonucleases , respectively [6 , 7] . Arabidopsis encodes three XRN proteins , AtXRN2 , AtXRN3 and AtXRN4 [11 , 12] . AtXRN4 is a predominantly cytoplasmic exoribonuclease that co-localizes with decapping proteins to form plant-processing bodies ( P-bodies ) for 5’–3’ RNA decay [11] , whereas functionally redundant XRN2 and XRN3 degrade transcripts within the nucleus [12] . Accumulated evidence suggests that RNA decay is also a major cellular antiviral mechanism [13] . Since viral RNAs are different for typical cellular mRNAs , the host RNA decay machinery can recognize and target these foreign RNAs for degradation . Indeed , knockout of key RNA decay pathway genes enhances viral infection and promotes the accumulation of viral RNA in infected cells [13] . A genome-wide screen of Saccharomyces cerevisiae single-gene deletion library and subsequent work revealed that XRN1p is involved in the degradation of tomato bushy stunt virus ( TBSV ) RNA in yeast [14 , 15] . Consistently , silencing of NbXRN4 in Nicotiana benthamiana , homologous to XRN1 in yeast , promotes viral RNA accumulation of tobacco mosaic virus ( TMV ) and TBSV [16 , 17] . The viral RNA level in Arabidopsis dcp2 mutant plants infected by a recombinant tobacco rattle virus ( TRV ) carrying the green fluorescence protein gene ( GFP ) is significantly higher than that in wild type plants [18] . In contrast , ectopic overexpression of AtXRN4 from Arabidopsis and OsXRN4 from Oryza sativa in N . benthamiana suppresses cucumber necrosis virus ( CNV ) and TMV infection , respectively [15 , 19] . Then , how do viruses cope with RNA decay ? In virus-infected metazoan cells , viruses manage to repress the key aspects of the host RNA decay pathway , which prevents the degradation of viral genomic RNA and promotes viral infection [13] . For instance , hepatitis C virus ( HCV ) targets the P-bodies and recruits P-body proteins for viral genome replication [20–22] . In the cases of picornaviruses such as poliovirus and human rhinovirus , viral proteases and/or the host cell proteasome appear to be involved in the degradation of the essential RNA decay proteins [23 , 24] . Despite recent progress in understanding the antiviral role of RNA silencing and RNA decay , many fundamental questions are yet to be unanswered . For instance , it is not clear if these two pathways are interlinked to operate against viral infection and whether they function simultaneously , sequentially or preferentially . The molecular mechanism underlying virus-mediated suppression of RNA decay remains poorly understood . This is particularly true in virus-plant interactions . In this study , we used turnip mosaic virus ( TuMV ) and N . benthamiana as a model system to investigate the possible involvement of several essential cytoplasmic 5’–3’ RNA decay pathway genes ( 5’RDGs ) in RNA silencing and RNA decay in virally infected plant cells . We found that knockdown of 5’RDGs but not overexpression of them suppresses PTGS induced by sense transcripts , and RNA silencing , compared to RNA decay , plays a predominant role in the degradation of foreign transcripts ( GFP or viral RNA ) . VSRs could hijack the key components from these pathways to suppress plant defense . These data highlight the overlapping function of the RNA silencing and RNA decay pathways in plants , as evidenced by their hierarchical and cooperative actions against foreign transcripts , and VSRs not only supress RNA silencing but also RNA decay to promote viral infection in plants .
We initiated this study by cloning and sequencing of four key 5’RDGs from N . benthamiana , including NbDCP1 , NbDCP2 , NbXRN4 and NbPARN . Sequence analysis revealed that the open reading frames ( ORFs ) of NbDCP1 , NbDCP2 , NbXRN4 and NbPARN contain 1113 nt ( GenBank accession number: KY402210 ) , 966 nt ( GenBank accession number: KY402211 ) , 2949 nt ( GenBank accession number: KY402212 ) , and 2056 nt ( GenBank accession number: KY402213 ) , respectively . The deduced amino acid sequences of NbDCP1 , NbDCP2 , NbXRN4 and NbPARN share 70 . 2% , 68 . 4% , 67 . 1% and 62% identity with their counterparts AtDCP1 , AtDCP2 , AtXRN4 and AtPARN from Arabidopsis , respectively . To document basic biology features of these 5’RDGs from N . benthamiana , we determined the expression pattern of these genes and subcellular localization of their encoded proteins . Quantitative real-time reverse-transcription PCR ( qRT-PCR ) was performed using total RNA isolated from different N . benthamiana tissues as template . All the four genes were constitutively expressed but their expression levels varied at different tissues ( Fig 1A ) . Overall , all the four genes showed the highest expression level in the root tissue , and the lowest in the stem tissue . To examine the subcellular localization of these four N . benthamiana 5’RDGs , their coding regions were firstly introduced into pDNOR221 vector by BP reaction ( Invitrogen ) , followed by recombination into in frame downstream of the coding sequence of yellow fluorescence protein ( YFP ) by LR reaction ( Invitrogen ) . The chimeric genes were transiently expressed in leaves of transgenic H2B-RFP N . benthamiana plants , and fluorescence was examined in agroinfiltrated transgenic leaves at 32 hours post infiltration ( hpi ) by confocal microscopy ( Fig 1B ) . YFP-NbDCP1 was present predominantly in the cytoplasm , forming the round granules , whereas the other three including YFP-NbDCP2 , YFP-NbXRN4 and YFP-NbPARN were evident in both the cytoplasm and nucleus , and some of YFP-NbDCP2 , YFP-NbXRN4 and YFP-NbPARN aggregated to form granules in the cytoplasm as well ( Fig 1B ) . Western blotting using GFP ( WB:GFP ) antibody ( Fig 1C ) detected the expected and specific band corresponding to the YFP-tagged NbDCP1 , NbDCP2 , NbXRN4 and NbPARN protein , confirming the presence of full-length recombinant proteins . To determine whether NbDCP1 , NbDCP2 and NbXRN4 interact with each other , we performed bimolecular fluorescence complementation ( BiFC ) assays [25] . The N- and C-YFP tagged proteins were co-expressed in leaves of transgenic H2B-RFP N . benthamiana plants and analyzed by live cell fluorescence microscopy at 32 hpi ( Fig 2A ) . Self as well as all mutual combinations of NbDCP1 , NbDCP2 , and XRN4 showed positive interactions . P3N-PIPO , which is a dedicated movement protein of TuMV [26] , was fused to N- or C-YFP to serve a negative control ( Fig 2A ) . The viral proteins P3N-PIPO and CI that interact to form a complex for viral cell-to-cell movement [27] were used as a positive control ( S1 Fig ) . Interestingly , the NbDCP2 self-interacting complex was present in the cytoplasm as well in the nucleus . Except the NbDCP2 self-interacting complex and the NbDCP2-N-YFP and NbXRN4-C-YFP combination , both of which were evenly distributed in the cytoplasm , the remaining combinations all formed cytoplasmic foci , regardless of reciprocal fusion with an N- or C-YFP ( Fig 2A ) . We further conducted a co-localization assay to confirm whether theses interaction complexes are formed in the same protein complexes . C-terminal cyan fluorescent protein ( CFP ) fused NbDCP1 ( NbDCP1-CFP ) and N-terminal YFP fused NbDCP2 ( YFP-NbDCP2 ) or NbXRN4 ( YFP-NbXRN4 ) , or C-terminal CFP fused NbDCP2 ( NbDCP2-CFP ) and YFP-NbXRN4 were co-expressed in the leaves of H2B transgenic N . benthamiana plants , and confocal microscopy was performed at 32 hpi . We found that NbDCP1-CFP and YFP-NbDCP2 or YFP-NbXRN4 co-localized in the cytoplasm to form the bright granules ( Fig 2B ) . Such bright foci were also observed in the cytoplasm of cells co-expressing NbDCP2-CFP and YFP-NbXRN4 ( Fig 2B ) . Taken together , these data suggest that NbDCP1 , NbDCP2 and NbXRN4 can form protein complexes through protein-protein interactions . Several recent studies have reported that essential components of the cytoplasmic 5’–3’ RNA decay pathway including XRN4 and DCP2 can suppress sense RNA-induced PTGS ( S-PTGS ) , but not inverted repeat RNA-induced PTGS ( IR-PTGS ) in Arabidopsis , suggesting that the RNA silencing and 5’–3’ RNA decay pathways are interlinked , possibly by sharing the same RNA substrate [28 , 29] . Consistently , compromising nonsense-mediated decay , deadenylation or exosome activity enhances S-PTGS , which requires host RNA-dependent RNA polymerase 6 ( RDR6 ) and SUPPRESSOR OF GENE SILENCING 3 ( SGS3 ) for the transformation of single-stranded RNA into dsRNA to trigger PTGS [30 , 31] . To determine whether the N . benthamiana 5’RDGs also affect S-PTGS and IR-PTGS as those in Arabidopsis , Agrobacterium cultures expressing 35S-GFP , 35S-GF ( GF: a fragment of GFP that can induce S-PTGS , Fig 3A ) and N-Myc-tagged NbDCP1 , NbDCP2 , NbXRN4 or NbPARN were co-infiltrated into N . benthamiana leaves . Leaves co-infiltrated with 35S-GFP , 35S-GF and empty vector ( Vec ) or the vector for expression of TBSV P19 ( a well-known gene silencing suppressor ) serve controls . Agroinfiltration of N . benthamiana leaves with 35S-GFP , 35S-GF and Vec induced GFP RNA silencing , and led to reduced GFP fluorescence under UV light at 5 days post infiltration ( dpi ) ( Fig 3B ) . While the intensity of green fluorescence increased substantially in leaf patches co-expressing GFP and P19 , no obvious differences of fluorescence were observed among leaf patches co-expressing Myc-tagged NbDCP1 , NbDCP2 , NbXRN4 or NbPARN or co-infiltrated with the vector control ( Fig 3B ) . Therefore , transient expression of the four 5’RDGs was unable to suppress the sense RNA trigger ( 35S-GF ) -induced GFP silencing . On the contrary , expression of these 5’RDGs led to the generation of more GFP-derived siRNAs and the reduced levels of GFP mRNA and protein , enhancing sense GFP-induced RNA silencing ( Fig 3D and 3F ) . We also examined the possible effects of the four 5’RDGs on dsRNA-induced gene silencing . Similar to previous observations [28–30] , there was no obvious RNA silencing suppression or enhancement , when 35S-dsGF ( dsRNA of GF ) as the silencing trigger co-expressed with 35S-GFP and N-Myc-tagged NbDCP1 , NbDCP2 , NbXRN4 or NbPARN in N . benthamiana leaves ( Fig 3C , 3E and 3G ) . As a positive control , P19 suppressed S-PTGS and IR-PTGS , and intensified GFP fluorescence , which was confirmed by immunoblot and RNA blot analyses ( Fig 3 ) . Taken together these data demonstrate that the four N . benthamiana 5’RDGs are likely involved in GFP-induced S-PTGS , but not in IR-PTGS . To further determine whether NbDCP1 , NbDCP2 , NbXRN4 or NbPARN is involved in RNA silencing , three expression vectors including 35S-GFP , 35S-GF and a hairpin RNAi construct containing NbDCP1 ( dsNbDCP1 ) , NbDCP2 ( dsNbDCP2 ) , NbXRN4 ( dsNbXRN4 ) or NbPARN ( dsNbPARN ) sequences under the control of the cauliflower mosaic virus ( CaMV ) 35S promoter , were agroinfiltrated into leaves of N . benthamiana plants . Compared to the weak GFP fluorescence in N . benthamiana leaf patches infiltrated with 35S-GFP , 35S-GF and Vec , expression of dsNbDCP1 , dsNbDCP2 , dsNbXRN4 or dsNbPARN led to an increase in green fluorescence in co-infiltrated regions ( Fig 4A ) , indicating that silencing of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN suppressed sense GFP-induced RNA silencing . Accordingly , qRT-PCR , RNA blot and protein gel blot analyses revealed that suppression of GFP silencing by dsNbDCP1 , dsNbDCP2 , dsNbXRN4 or dsNbPARN was accompanied by the increased levels of GFP mRNA and protein , and the reduced levels of GFP-specific siRNAs in the infiltrated leaves ( Fig 4B and 4C ) . As a control , a hairpin RNAi construct of GUS failed to suppress S-PTGS ( S2B Fig ) , indicating that knock-down of NbDCP1 , NbDCP2 , NbXRN4 and NbPARN specially inhibits S-PTGS , and expression of unrelated dsRNA does not overload and inhibit the silencing machinery . When 35S-GFP and 35S-dsGF were co-expressed with dsNbDCP1 , dsNbDCP2 , dsNbXRN4 or dsNbPARN in N . benthamiana , no GFP fluorescence was detected in the co-infiltrated areas , similar to the Vec-infiltrated leaf patches ( Fig 4D ) . In contrast , strong GFP signals were evidenced in the P19 co-expressing patches . GFP-specific siRNA blots showed that in comparison with Vec , silencing of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN drastically supressed the production of secondary siRNAs ( ‘P’ siRNAs ) during S-PTGS , but showed no obvious changes in the accumulation of primary siRNAs ( ‘GF’ siRNAs ) during IR-PTGS ( Fig 4E and 4F ) . These data support that NbDCP1 , NbDCP2 , NbXRN4 or NbPARN participate in GFP-induced S-PTGS , but are not involved in IR-PTGS . To analyze the role of NbDCP1 , NbDCP2 , NbXRN4 and NbPARN in RNA decay in the absence of S-PTGS , we used dsRDR6 transgenic plants , in which sense GFP induced S-PTGS is compromised [32] . Three expression vectors including 35S-GFP , 35S-GF and one of the following vectors: NbDCP1 , NbDCP2 , NbXRN4 , NbPARN and an empty vector ( Vec ) were co-infiltrated into dsRDR6 transgenic N . benthamiana leaves . At 5 dpi , the leaf patch infiltrated with 35S-GFP , 35S-GF and Vec still showed that strong GFP fluorescence owing to the suppression of S-PTGS in dsRDR6 N . benthamiana leaves compared to the corresponding patch in wild type ( Wt ) N . benthamiana leaves ( Fig 5A ) . Expression of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN in dsRDR6 transgenic plants reduced GFP fluorescence in comparison with the Vec control ( Fig 5A ) . qRT-PCR , RNA and protein gel blot analyses revealed that the reduced fluorescence in leaf patches expressing NbDCP1 , NbDCP2 , NbXRN4 or NbPARN was accompanied by a reduction of both GFP mRNA and protein , but without an increase of GFP-specific siRNAs ( Fig 5B and 5C ) . These data suggest that in RDR6-deficient plants , NbDCP1 , NbDCP2 , NbXRN4 or NbPARN suppresses GFP expression through the RNA decay pathway . In addition , 35S-GFP and 35S-GF were also co-infiltrated with dsNbDCP1 , dsNbDCP2 , dsNbXRN4 or dsNbPARN in dsRDR6 leaf patches . Consistent with our data in Fig 4 , much stronger fluorescence and more GFP RNA accumulations were observed when dsNbDCP1 , dsNbDCP2 , dsNbXRN4 or dsNbPARN was co-expressed with 35S-GFP and 35S-dsGF compared to Vec at 7 dpi ( S3 Fig ) . Taken together , these data suggest that all four 5’RDGs promote RNA silencing to target GFP RNA in Wt N . benthamiana plants , and degrade GFP transcripts through RNA decay in the RDR6-deficient plants where S-PTGS is blocked . Thus , RNA silencing compared to RNA decay has the priority to degrade GFP RNA . To clarify the role of RNA decay in TuMV infection , we determined the expression levels of the four 5’RDGs in TuMV-infected local and systemic leaves of N . benthamiana plants at 3 and 10 dpi by using qRT-PCR ( Fig 6A and 6B ) . The expression levels of all the four 5’RDGs were significantly upregulated in both local and systemic leaves in response to TuMV infection ( Fig 6A and 6B ) . We further checked whether TuMV replication proteins or replication vesicles are associated with P-bodies . The decapping protein DCP1 is a well-established marker for P-bodies in plants [11 , 33] . We co-expressed NbDCP1-CFP and three TuMV-encoded replication-required proteins including 6K2 ( which induces the formation of viral replication vesicles for virus replication ) , 6K2-NIa-VPg ( containing the genome-linked viral protein VPg ) and NIb ( the only viral RNA-dependent RNA polymerase ) with YFP tagged to their respective C-terminus . No typical co-localization signals were observed between NbDCP1 and these viral proteins ( Fig 6C ) . Possible co-localization between 5’RDGs and viral replication proteins was also tested using two TuMV infectious clones TuMV-6K2-mCherry and TuMV-CFP-NIb [34 , 35] . The former contains an extra copy of mCherry-tagged 6K2 , and in the latter , NIb is tagged by an N-terminal CFP . No co-localization was found between NbDCP1 and the 6K2-labelled viral replication vesicles or CFP-tagged NIb ( Fig 6D ) . These suggest that TuMV infection upregulates the RNA decay pathway which is not apparently associated with the virus replication complex . To examine whether these N . benthamiana 5’RDGs affect TuMV RNA accumulation , TuMV-GFP was co-infiltrated into N . benthamiana leaves with NbDCP1 , NbDCP2 , NbXRN4 , NbPARN or an empty vector ( Vec ) . At 3 dpi , the leaf patches infiltrated with these N . benthamiana 5’RDGs showed weaker GFP fluorescence compared to the Vec control ( Fig 7A ) . As GFP originated from the recombinant virus , its fluorescent intensity could be considered to be an indicator of TuMV replication . qRT-PCR analyses confirmed the reduced viral RNA levels in leaf patches co-expressing NbDCP1 , NbDCP2 , NbXRN4 or NbPARN ( Fig 7C ) . Consistently , this reduction was accompanied with an increase of TuMV-siRNAs ( Fig 7F ) . We also checked if NbDCP1 , NbDCP2 , NbXRN4 or NbPARN affects TuMV infection in dsRDR6 transgenic N . benthamiana plants ( Fig 7B ) . Overexpression of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN in dsRDR6 transgenic plants inhibited TuMV infection too , evidenced by attenuated GFP fluorescence and reduced levels of viral RNAs ( Fig 7B and 7C ) . Interestingly , remarkable amounts of TuMV-derived siRNAs were detected in the Vec-infiltrated control sample of dsRDR6 transgenic plants , possibly due to primary silencing induced by viral dsRNA ( viral RNA replicative intermediates ) during robust viral replication ( Fig 7F ) . These data demonstrate that expression of 5’RDGs inhibits TuMV RNA accumulation in both Wt and dsRDR6 transgenic plants . To examine 5’RDGs-mediated antiviral defense without dsRNA-mediated primary silencing in Wt and dsRDR6 transgenic N . benthamiana , we used a replication-deficient TuMV infectious clone , TuMV-GFP-ΔGDD [36] . This clone allows the transcription of full-length viral RNAs in the plant cell and subsequent translation of viral proteins . As the highly conserved GDD motif in NIb is mutated , the clone loses the ability to biosynthesize RNA and to generate viral RNA replicative intermediates . In comparison with the Vec control , expression of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN in Wt N . benthamiana suppressed TuMV-GFP-ΔGDD RNA accumulation and remarkably boosted the level of TuMV-siRNAs ( Fig 7F ) . In dsRDR6 transgenic plants , overexpression of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN reduced the accumulation of TuMV-GFP-ΔGDD transcripts and supressed the generation of TuMV-GFP-ΔGDD transcripts-derived siRNAs ( Fig 7F ) . These data suggest that NbDCP1 , NbDCP2 , NbXRN4 or NbPARN may degrade TuMV RNA via the RNA decay pathway when PTGS is disrupted . To minimize the interference of input RNA , we also conducted this assay with a very low concentration of agrobacterial cultures ( OD600 = 0 . 05 ) harboring TuMV or TuMV-ΔGDD . As expected , in TuMV-infiltrated Wt or dsRDR6 transgenic plants , viral RNA accumulation was significantly higher than that in the leaves infiltrated with TuMV-ΔGDD ( S4A Fig ) . Overexpression of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN , compared to Vec , reduced TuMV or TuMV-ΔGDD RNA accumulations . Overexpression of these RNA decay components also resulted in an obvious decrease of TuMV-derived siRNAs ( S4B , Fig 7F ) . Due to infiltration with a very lower concentration of agrobacterial culture , no detective viral siRNAs were found in TuMV-ΔGDD-infiltrated leaves . These results further suggest that during viral replication , siRNAs mainly derive from viral RNA replicative intermediates , which are proportional to viral RNA accumulations , rather than from RDR6-dependent secondary siRNAs . Taken together , these data support that 5’RDGs act as plant defense against TuMV RNA accumulation . To further investigate the effect of silencing of 5’RDGs on TuMV infection , N . benthamiana leaf patches were agroinfiltrated with TuMV-GFP and one of the following vectors: an empty vector ( Vec ) as a control , NbDCP1 , dsNbDCP1 , dsNbDCP2 , dsNbXRN4 and dsNbPARN . At 4 dpi , knock-down of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN led to the increased levels of TuMV-GFP fluorescence , GFP proteins , and TuMV RNA , and a reduced level of TuMV siRNAs ( Fig 8A and 8B and S5 Fig ) . Thus , silencing of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN promoted TuMV RNA accumulations possibly through suppression of TuMV-derived siRNA production . To verify if this also holds true for TuMV systemic infection , a modified TRV VIGS vector carrying the partial sequence of GUS ( as a control ) , NbDCP1 , NbDCP2 , NbXRN4 or NbPARN was pre-inoculated into N . benthamiana to knock down NbDCP1 , NbDCP2 , NbXRN4 or NbPARN expression ( S6 Fig ) . Silenced plants were then inoculated with TuMV-GFP . At 6 dpi , GFP signals started to appear along the veins in newly developed leaves under UV lamp in TRV-GUS-treated plants ( Fig 8C ) . However , all NbDCP1 , NbDCP2 , NbXRN4 or NbPARN-silenced plants developed much stronger GFP signals in the corresponding leaves ( Fig 8C ) . Consistently , higher levels of viral RNAs and GFP proteins were found in the 5’RDGs-silenced plants ( Fig 8D and 8E ) . It was also evidenced that silencing of 5’RDGs inhibited TuMV-derived siRNAs ( Fig 8E ) . These data suggest that knock-down of 5’RDGs facilitates TuMV infection and suppresses viral siRNAs in N . benthamiana . To further examine if these 5’RDGs also inhbit TuMV infection in Arabidopsis , we obtained Arabidopsis knockdown mutants of DCP2 and PARN , and knockout mutants of DCP1 and XRN4 ( S7 Fig ) . In comparison with Wt Arabidopsis plants , all mutants showed enhanced susceptibility to TuMV infection by accumulation of higher levels of viral RNA ( S8 Fig ) . Clearly , 5’RDGs play an anti-TuMV role in both N . benthamiana and Arabidopsis species . To screen for possible protein-protein interactions between the four 5’RDGs and 11 TuMV proteins , we conducted yeast two-hybrid ( Y2H ) assays . All the tested proteins were fused with the GAL4 transcription activation domain ( AD ) and the GAL4 DNA binding domain ( BD ) . As summarized in Fig 9A and 9B , regardless of whether AD or BD fusions were used for the assay , positive interactions were only found between NbDCP2 and VPg , and between NbXRN4 and HC-Pro . The interaction between NbDCP2 and VPg was further confirmed by BiFC in transgenic N . benthamiana expressing H2B-RFP as a nuclear marker ( Fig 10A ) . The NbDCP2 and VPg interaction was detected in the nucleus ( Fig 10A and S10 Fig ) . Consistently , NbDCP2 interacted with NbDCP1 ( serving as a positive control ) and formed bright granules in the cytoplasm , and no interaction was found between NbDCP1 and VPg ( Fig 10A ) . The fact that NbDCP2 interacts with VPg in the nucleus , and with NbDCP1 in the cytoplasm to form the decapping complex that is required for 5’–3’ RNA decay [11 , 33] prompted us to investigate whether TuMV VPg negatively regulates the assembly of the NbDCP1/NbDCP2 decapping complex . To test this hypothesis , YN-NbDCP1 and YC-NbDCP2 or YN-NbDCP2 and YC-NbDCP1 were co-infiltrated with VPg into the H2B-RFP leaves . No interaction signals between NbDCP1 and NbDCP2 were detected when VPg was co-expressed ( Fig 10A ) , suggesting that VPg indeed disrupts the NbDCP1 and NbDCP2 interaction . We further observed the subcellular co-localization of VPg , NbDCP1 and NbDCP2 . N . benthamiana leaves were co-infiltrated to transiently co-express VPg-YFP and NbDCP1-CFP or NbDCP2-CFP . No co-localization was observed for VPg-YFP and NbDCP1-CFP , whereas VPg-YFP co-localized with NbDCP2-CFP in the nucleus ( Fig 10B , two top rows ) . Moreover , co-expression of VPg remarkably suppressed the formation of the NbDCP1/NbDCP2 granules in the cytoplasm , in comparison with the vector control ( Fig 10B , two bottom rows , and 10C ) . Initially , we speculated that the reduced NbDCP1/NbDCP2 interaction was owing to the reduced NbDCP2 protein level induced by VPg . To test this possibility , total proteins were extracted from the leaves co-infiltrated with NbDCP2-CFP and empty vector ( Vec ) or Myc-VPg , and Western blot analysis was performed to determine the accumulation of NbDCP2-CFP . Co-expression of Myc-VPg did not obviously affect the NbDCP2-CFP protein level ( Fig 10D ) . Subsequently , we considered the possibility that the VPg-NbDCP2 interaction might negatively regulate the distribution of NbDCP2 in the cytoplasm to inhibit the formation of NbDCP1/NbDCP2 cytoplasmic granules . To test this assumption , we extracted the cytoplasmic and nuclear proteins separately , and performed Western blot analysis to determine the accumulation of NbDCP2-CFP in the cytoplasm and nuclear . Indeed , when Myc-VPg was co-expressed , the amount of NbDCP2-CFP obviously decreased in the cytoplasm but remarkably increased in the nucleus ( Fig 10D ) . These data suggest that TuMV VPg interferes with the interaction between NbDCP1 and NbDCP2 by targeting NbDCP2 from the cytoplasm to nucleus . The interaction between HC-Pro and NbXRN4 was verified by BiFC assays in N . benthamiana . Expression vectors YN-HC-Pro , YC-HC-Pro , YN-NbXRN4 and YC-NbXRN4 were generated to express HC-Pro or NbXRN4 fusions with YN or YC , respectively . Pairwise expression of YN-HC-Pro and YC-NbXRN4 or YN-NbXRN4 and YC-HC-Pro by agroinfiltration resulted in strong YFP fluorescence signals in the cytoplasm at 32 hpi ( Fig 11A ) . No YFP fluorescence was detectable in the leaf sample co-expressing HC-Pro and NbDCP1 , which serves as a negative control ( Fig 11A ) . We also examined the subcellular localization of HC-Pro and NbXRN4 in N . benthamiana leaf epidermal cells co-expressing CFP-tagged HC-Pro ( HC-Pro-CFP ) and YFP-tagged NbXRN4 ( YFP-NbXRN4 ) . We found that HC-Pro-CFP localized in the cytoplasm and nucleus , and some formed granules in the cytoplasm ( Fig 11B , panel I ) . HC-Pro-CFP co-localized with NbXRN4-CFP to form a bright dot in the cytoplasm ( Fig 11B , panel III ) . These results confirm that TuMV HC-Pro interacts with NbXRN4 . To assess the biological relevance of the HC-Pro and XRN4 interaction , we used the model plant Arabidopsis . At first , we confirmed that TuMV HC-Pro indeed interacts with AtXRN4 , the ortholog of NbXRN4 in Arabidopsis by conducting Y2H assays in yeast ( Fig 11C ) and BiFC assays in N . benthamiana ( Fig 11D ) . Then , we obtained an Arabidopsis xrn4 mutant and generated transgenic Arabidopsis plants expressing HC-Pro and AtXRN4 . Interestingly , the phenotype of the xrn4 mutant mimicked the mild phenotype of HC-Pro transgenic Arabidopsis plants ( Fig 11E ) , which displayed serrated leaf edges . Since the P-body is the RNA decay site , and HC-Pro interacts with XRN4 in the cytoplasmic granules resembling the P-body ( Fig 11A and 11D ) , we speculated that expression of HC-Pro might suppress AtXRN4 function in RNA decay . To test this idea , three potential substrates of AtXRN4 in RNA decay including AtEBF1 , AtRAP2 . 4 and AtNMT [37] were analyzed . We found that the mRNA levels of these three genes indeed significantly increased in the xrn4 mutant Arabidopsis plants and decreased in the XRN4 overexpression transgenic plants ( Fig 11F ) . Overexpression of HC-Pro significantly enhanced the mRNA accumulation of AtEBF1 , AtRAP2 . 4 and AtNMT , which was antagonized by co-overexpression of AtXRN4 ( Fig 11F ) . Consistently , co-overexpression of AtXRN4 partially remedied the typical phenotype induced by overexpression of HC-Pro ( Fig 11E ) and the expression levels of AtEBF1 , AtRAP2 . 4 and AtNMT were similar in HC-Pro-expressing Wt ( Col-0 ) and xrn4 mutant plants ( Fig 11G ) . Moreover , in TuMV-infected plants , the level of these XRN4 substrates was elevated ( Fig 11H ) , possibly due to HC-Pro . Taken together , these data suggest that HC-Pro interacts with XRN4 and the interaction inhibits XRN4 activity .
In this study , we found that overexpression of any of four essential RNA decay components 5’RDGs including DCP1 , DCP2 , XRN4 and PARN failed to suppress GFP-induced S-PTGS and production of siRNAs , and instead enhanced S-PTGS in N . benthamiana ( Fig 3 ) . We also found that knock-down of any of the four 5’RDGs genes repressed GFP-induced S-PTGS and inhibited the generation of GFP-derived siRNAs ( Fig 4 ) . These data suggest that 5’RDGs seem to play an additive role to S-PTGS in N . benthamiana . Our data are consistent with a recent report that the poly ( A ) tail of mRNA blocks RDR6 from converting canonical mRNAs into substrates for gene silencing and AtRDR6 has an intrinsic preference for poly ( A ) -less mRNAs over polyadenylated mRNAs as templates in Arabidopsis [38] . However , several previous studies have concluded that both 5′–3′ and 3′–5′ cytoplasmic RNA decay pathways repress S-PTGS in Arabidopsis [28–31] . For example , it has been shown that impairing deadenylation and decapping enhance S-PTGS in Arabidopsis [30 , 39] , possibly through restriction of RNA substrates from entry into the PTGS pathway . However , how deadenylation and decapping blocks the RNA substrate to access to PTGS is yet to be understood . It has also been suggested that RNA decay may compete for the same RNA substrates with RDRs-dependent RNA silencing to supress S-PTGS [28–30] . The assumption was based on the experimental evidence that the deficiency of RNA decay ribonucleases such as XRN4 enhances S-PTGS in Arabidopsis [28–30] . AtXRN4 does not play a significant role in controlling the degradation of unstable transcripts in A . thaliana , and it degrades predominantly the 5’ uncapped mRNA intermediates as well 3’ mRNA intermediates resulting from miRNA and possibly siRNA-mediated cleavage [37 , 40] . Moreover , XRN4-mediated decay also preferentially targets some transcripts such as those encoding nucleic acid–binding proteins and chloroplast proteins [40] . We speculate that rather than being competitive for substrates in Arabidopsis , NbXRN4 in N . benthamiana may degrade RNAs incompletely , generating RNA fragments , which facilitate RDRs-dependent dsRNA synthesis . Clearly , the interplay between RNA decay and RNA silencing is very complicated . The finding from this study may represent another example that not all findings from Arabidopsis can be simply extrapolated to other plant species such as N . benthamiana . The molecular mechanism by which the RNA decay pathway functions differently in relation to S-PTGS in Arabidopsis and N . benthamiana needs further study . RDR6 is required for S-PTGS by conversion of single stranded RNA into dsRNA , and GFP-induced S-PTGS is compromised in RDR6-defective N . benthamiana plants [32] . In this study , we found that the 5’RDGs suppressed GFP expression in dsRDR6 plants , and this was not concomitant with an increment of GFP siRNAs ( Fig 5 ) . Therefore , cytoplasmic RNA decay pathways are involved in the deadenylated or/ and decapped GFP RNA degradation . It is well known that silencing of RDR6 enhances transgene expression such as GFP [32 , 41] . In this study , knock-down of 5’RDGs in RDR6-defective plants further improves GFP expression ( S3 Fig ) , suggesting that RNA silencing and RNA decay may coordinate against the over-accumulation of exogenous transcripts . Taken together these data demonstrate that 5’RDGs from N . benthamiana may process GFP RNA to facilitate them into RNA silencing for degradation , and may directly degrade GFP RNA via the RNA decay pathway when S-PTGS pathway is interrupted . Thus , RNA silencing compared to RNA decay seems to play a predominant role in the degradation of GFP RNA , while both of them constitute an important defense against exogenous RNA . P bodies are the RNA decay sites , which have been implicated in infection by a number of RNA viruses [20–22 , 42] . In this study , we found that TuMV infection significantly upregulated the four 5’RDGs in both the inoculated leaves , and systemically infected leaves of N . benthamiana plants ( Fig 6A and 6B ) . TuMV replication-related proteins or TuMV replication complex were not co-localized with P bodies ( Fig 6C and 6D ) , suggesting that these 5’RDGs probably have no access to the virus replication site . As briefly discussed in Introduction , the cytoplasmic XRN4 inhibits infections by several RNA viruses in N . benthamiana such as TMV , TBSV , CNV and rice stripe virus infection [15–17 , 19] . We thus checked if these 5’RDGs affect TuMV RNA accumulation . Consistent with these reports and its roles in GFP RNA accumulation , overexpression of the four 5”RDGs including XRN4 supressed TuMV RNA accumulation and silencing of XRN4 and the other three 5’RDGs genes promoted TuMV infection in N . benthamiana ( Fig 7 , Fig 8 and S4 Fig ) . In addition , we also found that the orthologs of these 5’RDGs in Arabidopsis had the similar antiviral function ( S8 Fig ) . Therefore , we conclude that RNA decay is an antiviral pathway to TuMV in both Arabidopsis and N . benthamiana . To understand if and how TuMV counteracts the RNA decay machinery for viral infection , we screened possible interactions between 5’RDGs and TuMV encoding proteins . We found that NbDCP2 interacted with VPg , and NbXRN4 interacted with HC-Pro ( Fig 9 ) . Moreover , we found that the VPg/NbDCP2 interaction disrupted the formation of the NbDCP1/NbDCP2 complex possibly through targeting the cytoplasmic NbDCP2 to the nucleus ( Fig 10 and S5 Fig ) . Since the DCP1/DCP2 complex is essential for RNA decay , the VPg/NbDCP2 interaction would affect RNA decay-mediated TuMV RNA degradation . It is worth to mention that VPg is the virus-encoded protein that is covalently linked to the 5' end of the viral genome . Therefore , targeting DCP2 to the nucleus may inhibit the interaction of DCP2 with the genome-linked VPg in the cytoplasm , which can further supresses viral RNA degradation by RNA decay . Transgenic Arabidopsis plants expressing P1/HC-Pro , a potyvirus-encoded silencing suppressor , causes severe developmental anomalies such as stunting and serrated leaf edges . Initially , defects in both siRNA and microRNA ( miRNA ) pathways were through to account for the phenotype [43–45] . Subsequent studies showed that ectopic DCL1 largely alleviates developmental anomalies in P1/HC-Pro plants but cannot correct the P1/HC-Pro–associated defects in small RNA pathways , suggesting it is the aberrant Dicer activity that is responsible for developmental defects in the P1/HC-Pro plants [46] . In this study , we found that HC-Pro interacted with NbXRN4 and AtXRN4 in yeast cells and in plant ( Fig 9 and Fig 11 ) . Arabidopsis xrn4 mutant plants displayed serrated leaf edges , a developmental defect similar to HC-Pro transgenic plants , and overexpression of AtXRN4 could restore developmental defects in the HC-Pro plants ( Fig 11 ) , suggesting that the HC-Pro/AtXRN4 interaction may also contribute to the developmental defects through disruption of AtXRN4 activity to host gene expression . This assumption was further supported by the observation that overexpression of HC-Pro increased the level of XRN4 substrates likely via inhibition of XRN4 activity , mimicking the xrn4 mutant , and by the finding that ectopic overexpression of AtXRN4 mitigated the HC-Pro-induced phenotype and decreased the level of XRN4 substrates ( Fig 11 ) . Taken these data together , HC-Pro , as a highly efficient VSR of potyviruses , not only suppresses the PTGS and miRNA pathways , but also interferes with the XRN4-mediated 5’–3’ RNA decay pathway . Based on the above discussion , we propose a model that in addition to RNA silencing , the RNA decay pathway is another important component of antiviral immunity and viral VSRs function to counteract RNA decay ( Fig 12 ) . In brief , TuMV genomic RNA is deadenylated , decapped ( interfering with or removing VPg ) and degraded by XRN4-mediated 5’–3’ decay or exosome-mediated 3’–5’ degradation . Deadenylated , decapped , or XRN4-partially degraded RNAs facilitate RDR6 to transform ssRNA into dsRNA that triggers the PTGS pathway . Typical potyviruses encode two known VSRs: HC-Pro and VPg . HC-Pro interacts with XRN4 to inhibit its slicing activity . Here , HC-Pro as a major virus-encoded suppressor of RNA silencing is not included for discussion , which has been discussed in several reviews [47–49] . VPg may supress RNA decay-mediated degradation of viral RNA through targeting NbDCP2 to the nucleus to disrupt formation of the cytoplasmic NbDCP1/NbDCP2 complex . In addition , VPg interacts with SGS3 and mediates the degradation of SGS3 and its intimidate partner RDR6 via ubiquitination and autophagy pathways to block RDR6-mediated anti-viral response [36] .
N . benthamiana plants were maintained in an insect-free growth chamber at 25 oC and 60% relative humidity under a 16 h light/8 h dark photoperiod . Transgenic dsRDR6 and H2B-RFP lines were kindly provided by David C . Baulcombe ( Cambridge University , UK ) , and Michael M . Goodin ( University of Kentucky , USA ) , respectively . Arabidopsis plants were grown under the similar conditions described above for N . benthamiana . The Arabidopsis mutants: dcp1 ( SALK_014408C ) , dcp2 ( SALK_00519 ) , xrn4 ( SALK_014209 ) and parn ( Salk_072627 ) were obtained from the Arabidopsis Biological Resource Center ( ABRC ) at the Ohio State University . All mutants were confirmed by PCR essentially as previously described [36] . AtXRN4 and HC-Pro transgenic Arabidopsis plants were generated by the floral-dip method [50] and screened by directly spraying of 50 mg/L Basta solutions and subsequently by genomic PCR and DNA sequencing . Western blot and qRT-PCR were used to further confirm transgene expression . 35S:HC-Pro-CFP-1/35S:Myc-AtXRN4 transgenic plants were obtained by genetic crosses between 35S:HC-Pro-CFP-1 and 35S:Myc-AtXRN4-1 Arabidopsis plants and genotyping was done by PCR , qRT-PCR and Western blot . Full-length coding sequences of Arabidopsis AtDCP1 , AtDCP2 , AtXRN4 or AtPARN and N . benthamiana NbDCP1 , NbDCP2 , NbXRN4 or NbPARN were retried from public domains and obtained by RT-PCR . The amplified fragments were subcloned into pDONR221 entry vector and confirmed by DNA sequencing , and then transferred into different gateway-compatible vectors , including pGADT7-DEST ( AD ) , pGBKT7-DEST ( BD ) [51] , pEarleyGate-101 ( 101 , C-terminal YFP ) , pEarleyGate-102 ( 102 , C-terminal CFP ) , pEarleyGate-104 ( 104 , N-terminal YFP ) [52] , pBA-FLAG-4myc-DC ( pBA , N-terminal FLAG-4×Myc ) [53] , pEarleygate201-YN [C-terminal YN ( YN for the N-terminal half of YFP ) ] or pEarleygate202-YC [C-terminal YC ( YC for the C-terminal half of YFP ) [51] , to generate corresponding expression vectors . Gateway-compatible vectors ( AD/BD/101/102/104/pBA/YN/YC ) containing TuMV P1/HC-Pro/P3 /P3N-PIPO/6K1/CI/6K2/VPg/Pro/NIb/CP were described previously [36 , 54 , 55] . An RNAi construct with an inverted repeat sequence of NbDCP1 separated by an Arabidopsis intron was produced by overlapping PCR . A fragment of the NbDCP1 sense sequence was amplified using primers A-NbDCP1-F and A-Intron+NbDCP1-R , and primers B-NbDCP1+Intron-F and B-Intron-BamHI-R . The overlapping products were cloned into pCHF3 between the SacI and BamHI sites to produce pCHF3-35S:NbDCP1-intron . The corresponding antisense NbDCP1 fragment was amplified using primers C-NbDCP1-F and C-NbDCP1-R and subsequently cloned into pCHF3-35S:NbDCP1-intron between the BamHI and SalI sites to produce the RNAi construct pCHF3-35S:NbDCP1 . The similar strategy was used to generate the RNAi constructs of GUS , NbDCP1 , NbCP2 , NbXRN4 and NbPARN . A partial fragment of NbDCP1 , NbDCP2 , NbXRN4 or NbPARN was cloned into pTRV2 vector [56] to construct a TRV-based recombinant VIGS vector containing NbDCP1 , NbDCP2 , NbXRN4 or NbPARN . The coding sequence of full-length ( 1–711 bp ) , or N-terminal ( 1–340 bp ) of mgfp5-ER was cloned into pCHF3 to generate pCHF3-35S-GFP or pCHF3-35S-GF . The antisense fragment of mgfp5-ER ( N-terminal , 1–340 bp ) was cloned into pCHF3-35S-GFP to generate pCHF3-35S-dsGF . The detailed primers and restriction endonuclease sites were given in S1 Table . pCHF3:P19 and PTGS suppression assays were described previously [35 , 41 , 57 , 58] . GenBank accession numbers for the genes analyzed in this study are as follows: NbDCP1 ( KY402210 ) , NbDCP2 ( KY402211 ) , NbXRN4 ( KY402212 ) , NbPARN ( KY402213 ) , AtDCP1 ( NM_113710 ) , AtDCP2 ( NM_131203375 ) , AtXRN4 ( AF286718 ) and AtPARN ( AB223028 ) . Agrobacterium-mediated transient expression assays and viral infections in Wt N . benthamiana and H2B or dsRDR6 transgenic N . benthamiana plants were performed essentially as described [58 , 59] . The negative control plants were infiltrated with Agrobacterium cultures harboring an empty vector . The PTGS suppression assay and TuMV infection assay were repeated at least three times . For the TRV-VIGS assay , Agrobacterium cultures harboring pTRV1 and one of the pTRV2 vectors including TRV2-GUS ( control ) , TRV2-NbDCP1 , TRV2-NbDCP2 , TRV2-NbXRN4 and TRV2-NbPARN were mixed at a 1:1 ratio before infiltration . Y2H , BiFC and subcellular localization experiments were performed essentially as described [27 , 55 , 59] . For RNA blot analysis , total RNA was extracted from the infiltrated leaf patch , or virus-infected leaves with Trizol . High molecular weight ( HMW ) and low molecular weight ( LMW ) RNA analyses were conducted using 20 and 50 μg total RNA , respectively . HMW RNA was transferred to Hybond N+ nylon membranes ( Amersham Pharmacia Biotech ) by capillary transfer . Membranes were hybridized at 45°C to specific probes labeled with digoxigenin ( Roche ) . For siRNA blotting , LMW RNAs were enriched from total RNA , transferred to a Hybond-N+ membrane and hybridized for detection of siRNAs as described previously [41 , 57 , 58] . Blotted membranes were visualized by chemiluminescence according to the manufacturer’s manual ( ECL; GE Healthcare ) . For qRT-PCR analysis , total RNA was isolated using the RNeasy Plant Mini Kit ( Qiagen ) and treated with DNase I ( Thermo Fisher Scientific ) following the manufacturer's instructions . cDNA synthesized from reverse transcription of RNA samples was used to determine the mRNA levels of target genes as well as for quantification of TuMV accumulation levels . The procedures of cDNA synthesis and qRT-PCR assays were described previously [35 , 55] . NbActin or AtActinII was used as an internal control for N . benthamiana and Arabidopsis , respectively . All primer information used in qRT-PCR was given in S1 Table . Total protein was extracted from infiltrated leaf patches or TuMV-infiltrated or systemically infected N . benthamiana leaves as described previously [60] . Immunoblotting was performed with rabbit GFP polyclonal antibodies ( ab6556 , Abcam ) or Myc polyclonal antibodies ( ab9106 , Abcam ) , followed by goat anti-rabbit ( ab6721 ) secondary antibody conjugated to horseradish peroxidase ( Abcam ) . Blotted membranes were washed thoroughly and visualized by chemiluminescence .
|
RNA silencing and RNA decay are two essential pathways that determine the fate of cellular RNA molecules . RNA decay has been suggested to compete with RNA silencing for the RNA substrates and thus to suppress RNA silencing in plants . In this report , we show that the deficiency of key cytoplasmic 5’–3’ RNA decay pathway gene-encoded proteins ( 5’RDGs ) rather than overexpression of them attenuates RNA silencing and facilitates viral RNA accumulation in turnip mosaic virus ( TuMV ) -infected Nicotiana benthamiana plants . 5’RDGs can inhibit the accumulation of exogenous gene transcripts or viral RNA via the RNA decay pathway when RNA silencing is compromised . Therefore , RNA silencing and RNA decay are important defense mechanisms that coordinate to function against exogenous gene transcripts or viral RNA . It is well known that viruses have evolved to encode viral suppressors of RNA silencing ( VSRs ) that counteract RNA silencing for an invading virus to establish infection . We discover that two known VSRs of TuMV interact with 5’RDGs and the interactions subvert RNA decay . Taken together , this study provides new insights into the concerted function of RNA silencing and RNA decay in plant immunity against exogenous gene transcripts and viral RNA , and reveals a novel function of VSRs .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"biotechnology",
"plant",
"anatomy",
"rna",
"interference",
"gene",
"regulation",
"brassica",
"viruses",
"plant",
"science",
"model",
"organisms",
"rna",
"viruses",
"genetically",
"modified",
"plants",
"experimental",
"organism",
"systems",
"epigenetics",
"molecular",
"biology",
"techniques",
"plants",
"cellular",
"structures",
"and",
"organelles",
"genetic",
"engineering",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"small",
"interfering",
"rnas",
"genetically",
"modified",
"organisms",
"genetic",
"interference",
"hyperexpression",
"techniques",
"gene",
"expression",
"leaves",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"cytoplasm",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"rna",
"eukaryota",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"plant",
"biotechnology",
"organisms"
] |
2018
|
RNA decay is an antiviral defense in plants that is counteracted by viral RNA silencing suppressors
|
The protein complex known as cohesin binds pericentric regions and other sites of eukaryotic genomes to mediate cohesion of sister chromatids . In budding yeast Saccharomyces cerevisiae , cohesin also binds silent chromatin , a repressive chromatin structure that functionally resembles heterochromatin of higher eukaryotes . We developed a protein-targeting assay to investigate the mechanistic basis for cohesion of silent chromatin domains . Individual silencing factors were tethered to sites where pairing of sister chromatids could be evaluated by fluorescence microscopy . We report that the evolutionarily conserved Sir2 histone deacetylase , an essential silent chromatin component , was both necessary and sufficient for cohesion . The cohesin genes were required , but the Sir2 deacetylase activity and other silencing factors were not . Binding of cohesin to silent chromatin was achieved with a small carboxyl terminal fragment of Sir2 . Taken together , these data define a unique role for Sir2 in cohesion of silent chromatin that is distinct from the enzyme's role as a histone deacetylase .
Proper segregation of chromosomes at mitosis and meiosis requires sister chromatid cohesion . The process ensures that newly replicated chromatids bi-orient on spindle microtubules such that a single copy of each chromosome transfers to progeny cells . Defects in the sister chromatid cohesion pathway lead to certain developmental diseases , and chromosome segregation defects like those seen in cancer [1]–[4] . Cohesion of sister chromatids is mediated by a protein complex known as cohesin [5] , [6] . The core complex consists of a hetero-dimer of SMC proteins Smc1 and Smc3 , as well as non-SMC proteins Scc3/Irr1 and Mcd1/Scc1/Rad21 ( hereafter referred to as Scc3 and Mcd1 , respectively ) . The subunits form a large protein ring with a striking central void . Thus , a prominently held view is that cohesin holds sister chromatids together by single complexes embracing both chromatids . Elegant protein-crosslinking studies showed that single cohesin rings can indeed hold together two partially purified minichromosomes [7] . Other data raises the possibility that cohesin might hold sister chromatids together by a different mechanism [8]–[10] . Cohesin binds discrete sites on chromosomal DNA . Most non-centromeric sites in budding and fission yeasts lie within the AT-rich regions between convergently transcribed genes [11]–[13] . Transcriptional elongation redistributes complexes from intragenic to intergenic regions , suggesting that cohesin enrichment is maintained dynamically . In contrast to the situation in these fungal systems , cohesin maps along the lengths of actively transcribed genes in Drosophila and to sites within transcribed genes in humans [14]–[16] . Thus , cohesin binding and transcription are not always mutually exclusive . Cohesin is also found within pericentric heterochromatin regions where transcription is suppressed but not extinguished . In fission yeast , the complex is retained at these locations by Swi6 , a homolog of heterochromatin protein HP1 , which interacts with cohesin subunit Psc3 ( Scc3 in budding yeast ) [17] , [18] . During meiosis , Swi6 also interacts with shugoshin , a protein that protects centromeric cohesin from being dismantled [19] . In heterochromatin mutants , cohesin does not bind pericentric domains and mitotic chromosomes fail to mount properly onto spindle microtubules . Budding yeast lacks Swi6 and pericentric heterochromatin but it does contain transcriptionally silenced domains that nevertheless bind cohesin . Using the HMR locus as one representative example , we found that silencing mutations selectively disrupted cohesin binding and correspondingly abolished cohesion of sister chromatid DNA bearing the locus [9] . A search to understand why cohesin accumulates at HMR served as the impetus for this study . Based on the chromatin-mediated mechanism of regional DNA inactivation , transcriptionally silenced domains in budding yeast are referred to as silent chromatin [20] . Like heterochromatin domains in other organisms , silent chromatin is packaged with histones that bear a distinct signature of post-translational modifications . Specifically , acetylation and methylation of lysines are absent . Silent chromatin domains associate with a complex of non-histone silencing factors known as the Sir proteins ( Sir2 , Sir3 and Sir4 ) . Sir2 is a member of the evolutionarily conserved class of NAD+-dependent protein deacetylases known as sirtuins . The enzyme creates and maintains histone deacetylation within silent chromatin . Sir3 and Sir4 associate with the suitably deacetylated histones . The complex of Sir proteins is first recruited to sites of action by cis-acting elements known as silencers , which bind ORC , as well as Abf1 and Rap1 in various combinations . Following recruitment , cycles of histone deacetylation and histone binding allow the Sir proteins to spread over kilobases . A tRNA gene acts as a barrier element on the right side of HMR that blocks silent chromatin from spreading further downstream [21] . The element also augments HMR with sufficient cohesin for cohesion [22] , probably through recruitment of the Scc2/4 cohesin loading complex [23] , [24] . We considered two competing hypotheses to account for retention of cohesin at HMR . The first , based on a simple recruitment model , posits that a silent chromatin component interacts directly with cohesin or some factor associated with the complex . A second hypothesis stems from the ability of silent chromatin to impede a broad-range of DNA-based events , such as DNA replication , repair and transcription [20] . If silent chromatin also suppresses an activity that mobilizes cohesin , the complex would accumulate at silenced loci . To distinguish between these possibilities , we developed assays to determine whether silencing or silent chromatin components were required for cohesion of HMR . Our studies show that Sir2 is sufficient for cohesion , even in the absence of silencing .
Our principle assay for cohesion at HMR utilizes a strain in which the locus is tagged with lac-GFP and flanked by target sites for a site-specific recombinase [9] . Inducible excision after arrest in M phase converts HMR loci on sister chromatids into a pair of extrachromosomal circles that produce one bright fluorescent focus if they are held together and two foci if they are not ( Figure 1A and 1B ) . To test whether silent chromatin components can mediate cohesion we tethered individual silencing factors directly to the DNA circles ( Figure 1C ) . To this end , the E silencer of HMR was replaced with a synthetic construct ( 6lexopssEB ) that includes binding sites for Rap1 , Abf1 and the bacterial protein lexA . The I silencer was deleted . These modifications were previously shown to eliminate silencing of the locus [25] . Individual silencing factors were then targeted to HMR-6lexopssEB as lexA-linked fusion proteins . Cell cycle arrest in M phase , recombinase induction and fluorescence microscopy were performed as described previously [9] . Tethering silent chromatin components to DNA often nucleates silent chromatin assembly and restores transcriptional repression [25] , [26] . In these situations , it would be impossible to determine whether the tethered protein , a co-recruited protein or the silenced state was responsible for cohesion . Therefore , tethered proteins were also examined under conditions that abolish silent chromatin assembly to evaluate their precise roles in cohesion . Pilot experiments showed that excised circles bearing the HMR-6lexopssEB construct colocalized infrequently [9] . When lexA was expressed , only 22% of the nuclei contained the single bright fluorescent spot ( Figure 1D ) . Strikingly , cohesion of the circles increased to 67% when lexA was fused to Sir2 ( designated lexA-Sir278–562 ) . Tethering Sir2 to DNA was essential . In a strain lacking lexA binding sites at HMR , the chimera failed to produce cohesion ( Figure S1 ) . LexA-Sir278–562 lacks the first 77 amino acids of Sir2 that are dispensable for transcriptional repression [27] . We confirmed that lexA-Sir278–562 nucleates silencing at HMR using a strain that contains lexA binding sites and a TRP1 reporter gene at the locus ( Figure 1E ) . Taken together , these initial findings demonstrate that tethered Sir2 confers both silencing and cohesion at HMR . The Sir2 polypeptide consists of a conserved catalytic core , as well as N and C terminal domains that help target the deacetylase to sites of action [28] . An allele lacking the N-terminal 198 amino acids confers little transcriptional repression , even when tethered to DNA [29] . To generate a lexA chimera with similar characteristics , we eliminated the entire N-terminal domain ( amino acids 1–242 ) . Surprisingly , this construct ( lexA-Sir2243–562 ) yielded a measurable degree of silencing in a strain with intact SIR genes ( for comparisons , see lexA-Sir278–562 and lexA alone in figure 2A ) . Deletion of either the SIR2 or SIR4 genes eliminated silencing by lexA-Sir2243–562 , indicating that 1 ) the chimera operates within the Sir pathway , and that 2 ) the chimera requires the endogenous full-length Sir2 for transcriptional repression . The reliance of lexA-Sir2243–562 on other SIR genes , including SIR2 , for silencing made the chimera an ideal candidate for further study . Figure 2B shows that lexA-Sir2243–562 produced cohesion in over 60% of the nuclei examined . Importantly , cohesion of the excised HMR circles persisted in strains that lacked SIR2 , SIR3 or SIR4 . We conclude that tethered Sir2 can mediate cohesion in the absence of transcriptional silencing and without the aid of endogenous Sir proteins . Sir3 was also examined directly with the targeted cohesion assay . When the protein was linked to lexA , HMR circles colocalized in over 60% of wild-type cells ( Figure 2C ) . The tethered protein also conferred transcriptional repression in the wild-type reporter strain ( Figure 2A ) . Both cohesion and silencing by Sir3-lexA were significantly diminished by deletion of Sir2 . Elimination of Sir4 , on the other hand , disrupted silencing but not cohesion . A simple explanation for the requirement of Sir2 but not Sir4 is that tethered Sir3 recruits Sir2 , which in turn mediates cohesion of the locus . We note that in the absence of Sir2 , Sir3-lexA yielded a slightly higher level of cohesion than lexA alone ( Sir3-lexA = 34% vs lexA = 26 . 3% ) . This difference is sufficiently small ( p = 0 . 03 ) that we cannot conclude equivocally whether Sir3 possesses a subtle intrinsic cohesion activity . Given the strong cohesion signals afforded by Sir2 , we focused our attention on this Sir protein for the remainder of the study . If Sir2 mediates cohesion at HMR then the protein ought to impart cohesion when tethered at other genomic positions . We explored this possibility by targeting the protein to the LYS2 locus . LYS2 is situated near the center of chromosome II , hundreds of kilobases away from silent chromatin domains at the chromosome ends [30] . The locus had previously been modified to contain lexA-binding sites , as well as lac repressor sites and recombinase sites for the DNA excision assay [31] . When lexA alone was expressed , LYS2 DNA circles colocalized in 37% of the cells ( Figure 3 ) . This value is higher than the baseline for HMR cohesion under similar conditions ( Figure 1D ) . The value is sufficiently low , however , to detect increases in cohesion due to tethered Sir2 fragments . Indeed , cohesion of LYS2 circles increased to 60% when lexA-Sir2243–562 was expressed . Pairing persisted in a strain lacking SIR3 indicating that cohesion was due to the tethered protein and not due to formation of silent chromatin at LYS2 . These findings indicate that Sir2 can impart cohesion at chromosomal locations other than HMR . We next asked whether the deacetylase activity of Sir2 was responsible for Sir2-mediated cohesion . To address this question , we introduced a well-characterized active site mutation ( H364Y ) into lexA-Sir2243–562 . Previous studies showed that this mutation abolishes Sir2 deacetylase activity , silencing and silent chromatin formation [32] , [33] . We found here that the mutated polypeptide conferred as much cohesion to HMR circles as the unaltered polypeptide ( Figure 4A ) . This experiment was performed in a sir2 null strain to eliminate contributions of the endogenous gene ( see figure 2A ) . Furthermore , acting on the remote possibility that tethered Sir2 mediates cohesion by recruiting one of the other yeast sirtuins ( Hst1-4 ) , we repeated the experiment in media supplemented with nicotinamide , a generic sirtuin inhibitor [28] . No decrease in cohesion was observed ( 64% vs . 61% with nicotinamide; p = 0 . 5 ) . Collectively , these results show that the enzymatic activity of Sir2 or other sirtuins is not required for cohesion by tethered Sir2 . To map the Sir2 domain responsible for cohesion we generated a set of truncation mutants . Figure 4A shows that all but one of the constructs yielded HMR cohesion levels significantly above background . The exception is a lexA chimera that bears just the conserved catalytic core of Sir2 ( residues 243–499 ) . All of the other cohesion-proficient chimeras share in common a small C-terminal domain of Sir2 spanning amino acids 525–547 . The picture that emerges is that Sir2 contains a discrete motif within the non-catalytic , C-terminal region of the protein that mediates cohesion . Hst1 is a yeast sirtuin that bears considerable amino acid similarity to Sir2 in the C-terminal region ( Figure 4B ) . The deacetylase represses middle sporulation genes in vegetative cells , as well as genes involved in NAD+ and thiamine biosynthesis [34]–[36] . Hst1 differs from Sir2 in that the protein acts locally to repress specific promoters rather than by forming an extended repressive domain [37] . We tethered the C-terminal domain ( amino acids 440–503 ) to HMR in the same Δsir2 strain used to evaluate the lexA-Sir2 constructs . Figure 4A shows that lexA-Hst1440–503 imparts a comparable degree of targeted cohesion . These results indicate that Sir2-mediated cohesion is not limited to just one member of the sirtuin family . Both Sir2 and its yeast paralog Hst2 can form homotrimers [38] , [39] . Thus , one explanation for DNA colocalization is that tethered Sir2 fragments on different DNAs associate with one another directly . To explore this possibility we performed a two-hybrid analysis using lexA-Sir2243–562 ( H364Y ) as a bait protein . The experiments utilized a HIS3 reporter strain that lacks the endogenous SIR2 gene . A weak positive interaction signal was obtained with a prey vector bearing full length Sir2 fused to the Gal4 activation domain ( Figure 5A ) . Importantly , no interaction was seen with a prey vector bearing the shorter Sir2243–562 fragment . Given that all of our critical experiments were performed with this fragment in strains lacking full length Sir2 , colocalization of HMR circles is not likely attributable to Sir2 self-association . Cohesin mediates cohesion of the native HMR locus [9] . We therefore anticipated that cohesin genes would also be required for cohesion by tethered Sir2 . To test this possibility we crossed temperature sensitive alleles of MCD1/SCC1 and SMC3 into our DNA circle-producing strain . The scc1-73 and smc3-42 mutants and a wild-type counterpart were arrested in mitosis at permissive temperature ( 24°C ) . After recombining the HMR locus , the cultures were divided: half was maintained at the permissive temperature while the other half was shifted to 37°C , the non-permissive temperature for these mutants ( see Figure 5 legend for details ) . In the wild-type strain , cohesion of the HMR circles by lexA-Sir2243–562 was unaffected by the temperature shift ( Figure 5B ) . By contrast , both mutant strains displayed a significant reduction in HMR cohesion at the non-permissive temperature . This data indicates that cohesin is responsible for cohesion of DNA circles bound by Sir2 . Chromatin immunoprecipitation ( ChIP ) was used previously to show that cohesin associates with HMR in a silencing-dependent manner [9] . We showed that Mcd1-TAP binding was lost when silent chromatin assembly was blocked by 1 ) deletion of SIR3 , or 2 ) inhibition of the Sir2 deacetylase ( see ChIPs of chromosomal templates in figures 5A and 7D of [9] ) . In the current study , a similar ChIP protocol was used to test whether lexA-Sir2243–562 retained cohesin at HMR-6lexopssEB . Unexpectedly , we could not obtain reproducible enrichment of the targeted locus . A variety of conditions and reagents were tested , and the procedure was validated with native silent chromatin ( see below ) . We suspect that the level of cohesin necessary for colocalization in the targeted cohesion assay falls below the detection limit of this ChIP experiment . We turned instead to a protein chimera approach we recently developed to study other aspects of transcriptionally silent chromatin [40] . In that study , Sir3 was fused to the Rpd3-family deacetylase Hos3 to show that the roles of Sir2 in silencing could be bypassed entirely . We demonstrated that the Sir3-Hos32–549 chimera 1 ) spread throughout HMR , 2 ) deacetylated histones across the locus and 3 ) required both silencers and Sir4 to mediate repression . Silencing could also be achieved by fusing Sir3 to a fragment of Sir2 that possessed enzyme activity but that lacked domains necessary for targeting . Here these Sir3 chimeras ( Sir3-Sir2243–562 and Sir3-Hos32–549 ) were used to investigate the role for Sir2 in binding cohesin at a silenced domain . An additional chimera ( Sir3-Hos32–549-Sir2499–562 ) was constructed to study the contribution of the 64 amino acid , cohesion-proficient fragment of Sir2 . Mating assays were used to evaluate the silencing potential of each chimera . In this assay , loss of HMR silencing in MATα cells creates a pseudo-diploid state that blocks mating and thus subsequent growth on SD indicator plates . Figure 6A confirms previous findings that Sir3-Sir2243–562 and Sir3-Hos32–549 mediate silencing of HMR in the absence of endogenous Sir2 . The figure shows that Sir3-Hos32–549-Sir2499–562 also conferred silencing of HMR , albeit at a reproducibly reduced level . This functional assay indicates that Sir3-Hos32–549-Sir2499–562 delivers the Sir2499–562 fragment to the site where cohesion and cohesin binding were to be tested . Cohesion by the Sir3 chimeras was evaluated in a sir2 null strain that produces GFP-tagged HMR circles with wild-type silencers ( Figure 6B ) . In the absence of a chimera , HMR cohesion occurred in 33% of the cells . When Sir2 or the Sir3-Sir2243–562 was expressed , HMR cohesion levels rose to 69% and 61% , respectively . By contrast , expression of Sir3-Hos32–549 did not increase cohesion above background levels . Remarkably , addition of Sir2499–562 to the Sir3- Hos32–549 chimera restored colocalization to the level obtained with Sir3- Sir2243–562 . This analysis indicates that the C-terminal fragment of Sir2 must be present within silenced chromatin for cohesion to occur . ChIP of TAP-tagged Mcd1 was used to measure the ability of the chimeras to position cohesin at the HMR a2 gene . A cohesin-associated region of chromosome V ( designated 549 . 7 ) that is not influenced by the SIR genes was used as a point of comparison [22] . Reference SIR2 and Δsir2 strains in figure 6C confirmed earlier findings: binding of Mcd1-TAP at HMR is hindered when silent chromatin is disrupted by loss of a single Sir protein , in this case Sir2 . Expression of the Sir3-Sir2243–562 chimera restored cohesin binding at HMR to within 20% of the native level . By contrast , expression of the silencing-proficient Sir3-Hos32–549 chimera did not raise cohesin binding above the background sir2 null level . Importantly , the addition of 64 amino acids of Sir2 to the end of the Sir3-Hos32–549 chimera increased cohesin binding substantially . Taken together , these data indicate that Sir2 must be present within silent chromatin for cohesin to accumulate at silenced loci , and that a small C-terminal portion of Sir2 is sufficient for this activity . Sir2 associates with the cluster of tandemly repeated ribosomal RNA genes known as the rDNA array . In this context the protein suppresses recombination between the repeated elements and suppresses RNA polymerase II transcription within each element [41]–[43] . Sir2 has been implicated in cohesin binding at the rDNA [44] , [45] . It therefore seemed prudent to test whether rDNA-specific , protein partners of Sir2 modulate cohesion by the tethered protein . We first considered Net1 , which along with Sir2 and Cdc14 forms the RENT complex [46] , [47] . This protein is required for Sir2 binding at the rDNA and it has been found at HMR when over-expressed [46] , [48] . A 15 amino acid C-terminal truncation of Sir2 disrupts the Net1-Sir2 interaction , abolishing rDNA silencing but not silencing of telomeres or the HM loci [49] . Figure 4A shows that deleting these 15 residues ( lexA-Sir2243–547 ) did not interfere substantially with cohesion of HMR circles . We conclude that the RENT complex is not necessary for cohesion by tethered Sir2 . Transcriptional silencing by Sir2 at the rDNA recombinational enhancer requires a set of interacting proteins that includes Tof2 and a pair of bifunctional factors Csm1 and Lrs4 . During meiosis I , Csm1 and Lrs4 form the monopolin complex that orients sister chromatid pairs towards the same spindle pole [50] , [51] . Csm1 interacts with both Mcd1 and Smc1 prompting Huang and Moazed to hypothesize that these proteins link cohesin to the rDNA [52] . We tested whether these genes were required for cohesion by lexA-Sir2243–562 . Figure 7 shows that neither TOF2 , CSM1 nor LRS4 were required for colocalization of HMR circles . Collectively , the findings indicate that these rDNA silencing and stability proteins do not contribute to cohesion of HMR by tethered Sir2 .
Although our studies here focused on HMR we expect that the relationship between Sir2 and cohesion extends to other loci where Sir proteins assemble . Indeed , preliminary evidence indicates that the HML mating-type locus is also cohered in a silencing-dependent manner ( Campor and Gartenberg , unpublished results ) . Why do silent chromatin and cohesion converge ? Initial studies suggested a role in regulating transcriptional repression . Donze and Kamakaka first showed that silencing spread beyond HMR barrier elements in cohesin mutants [21] . Steve Bell and colleagues followed by showing that cohesin delayed establishment of silencing in cells that were walked step-wise through the cell cycle [53] . A parsimonious explanation for these observations is that cohesin impedes the conversion of active chromatin to silenced chromatin . Numerous studies in higher eukaryotes have further linked cohesin to gene regulatory phenomena ( see [54] for a review ) . Intriguingly , cohesin was recently shown to form loops between enhancers and promoters by interacting with a transcriptional coactivation complex known as mediator [55] . Similarly , cohesin forms loops between distant sites by binding the mammalian CTCF , a protein that associates with insulators as well as other gene regulatory elements [56]–[59] . In yeast , silent chromatin domains fold-back upon themselves and interact with one another over great distances [60] , [61] . Thus , one possibility is that cohesin facilitates long interactions to regulate silent chromatin domains . The rationale for Sir2 mediating cohesion might alternatively be related to its role in genome stabilization at the rDNA . Binding of the deacetylase is necessary for binding of cohesin , which in turn is thought to block unequal sister chromatid exchange by maintaining the register between rDNA elements of opposing sister chromatids [44] . Exactly how Sir2 retains cohesin at the locus in not entirely clear . In one model , the deacetylase modulates cohesin levels indirectly by silencing a conserved RNA polymerase II promoter element near the rDNA recombinational enhancer [45] . According to the model , transcription by RNA polymerase II displaces cohesin when Sir2 is absent . A competing model by Huang and Moazed suggests that direct interaction between cohesin and one of the components of the rDNA silencing pathway , Csm1 specifically , could account for recruitment of the complex [52] . Our work with tethered fragments of Sir2 suggests an even more direct possibility: the polypeptide , not its capacity to silence , mediates cohesin recruitment directly . We note that a direct recruitment model for Sir2 need not be mutually exclusive with models based on transcriptional inhibition , or with other factors that contribute to rDNA stability [62] . Acetylation and deacetylation of cohesin subunits plays a newly appreciated role in regulating cohesion during the cell cycle . Cohesion is established during S phase when the Eco1 protein acetyltransferase acetylates Smc3 [63]–[66] . This modification persists until cohesin complexes disassemble at anaphase onset . Deacetylation is a prerequisite for Smc3 to be re-used in the next cell cycle . Recently , the Rpd3-family member Hos1 was identified as the principle Smc3 deacetylase in yeast [67]–[69] . That residual deacetylation persists in the absence of Hos1 suggests that additional Smc3 deacetylation activities remain to be discovered [67] . Following DNA double strand breaks , Eco1 similarly acetylates Mcd1 to establish damage-induced cohesion [70] . Presumably , a parallel pathway exists for Mcd1 deacetylation . Whether Sir2 or a combination of sirtuins is involved in Mcd1 deacetylation or the residual deacetylation of Smc3 has not been determined . The catalytic activity of Sir2 accounts for all other known functions of the enzyme . By contrast , cohesion by tethered Sir2 fragments does not even require the conserved catalytic core . Instead , we found that a small domain at the carboxyl-terminus was responsible . We anticipate that this domain retains cohesin at silenced loci by interacting directly with a cohesin subunit or with other proteins involved in cohesin utilization . Conversely , such an interaction could be important in situations where Sir2 might be recruited to sites where cohesin binds . Significant homology exists between the C-terminal domain of Sir2 and Hst1 . That this Hst1 domain also mediates cohesion when tethered to DNA suggests that cohesion occurs at the numerous promoters where Hst1 binds to regulate gene expression [34]–[36] . The lack of a homologous C-terminal domain in mammalian sirtuins thwarts a simple extrapolation to a cohesion connection in higher eukaryotes . However , the characteristics of two mammalian sirtuins , SirT1 and SirT6 , warrant consideration ( see [71] , [72] , and references therein ) . Like Sir2 , these mammalian enzymes deacetylate histones ( and other protein targets ) to regulate gene expression . Additionally , SirT1 plays multiple roles in heterochromatic repression and SirT6 localizes to heterochromatin domains . Double strand breaks may represent sites of particular interest . Cohesin is recruited to these sites in yeast and in humans , as are Sir2 , Hst1 , SirT1 and SirT6 [73] , [74] . The mammalian enzymes have been shown to suppress genomic instability , in part , by modifying DNA repair factors ( see [75] for most recent example ) . Whether SirT1 and SirT6 also link sister chromatid cohesion to these chromosome-based events has not yet been tested .
Table S1 provides a complete list of strains used in this study . Cohesion of HMR by tethered proteins was measured with variants of strain CSW19 ( RS::6lexopssEB-a2a1-256lacop-TRP1-ΔhmrI::RS ( LEU2::GAL1-R ) 2::leu2-3 , 112 ADE2::HIS3p-lacGFP::ade2-1 ) . Recombinase target sites are designated as RS . Cohesion of the LYS2 gene was measured with variants of strain CSW91 ( RS::lys2-TRP1-4lexop-256lacop::RS ) . Cohesion of HMR by Sir3 chimeras was measured with native silencers in strain CSW84 ( RS::HMRE-a2a1-HMRI-TRP1-256lacop::RS Δsir2 ) . Silencing assays were performed with strains derived from GA-2050 ( Aeb4lexop-TRP1-HMRI ) and two hybrid assays were performed with strain JCY13 ( LYS2:: ( 4xlexop ) -HIS3 Δsir2 ) . ChIP assays were performed in variants of strain CSW116 ( MCD1-TAP Δsir2 ) . Complete ORF deletions were generated by PCR-mediated gene replacement using purified plasmids or extracted yeast DNA as PCR templates . All modifications were confirmed by combined gain and loss of diagnostic PCR products . Strains CSW18 and CSW19 are segregants of a cross between CSW10 and YCL49 . Strains CSW47 and CSW48 are segregants of crosses between CSW36 and either K5832 or CRC85 . Strain CSW91 is a segregant of a cross between CSW19 and GA-2627 . Plasmids pRS403-lexA-Sir278–562 and pRS403-lexA were integrated in single copy at the HIS3 locus of CSW19 to yield strains CSW36 and CSW37 , respectively . Tables S2–S3 provide detailed information about the plasmids used and how they were constructed . In addition to using traditional bacterial cloning techniques , plasmids were constructed in yeast by PCR-mediated plasmid gap repair ( P-MPGR ) or fragment-mediated plasmid gap repair ( F-MGPR ) using restriction digestion products . SIR2 truncations were generated by oligonucleotide-mediated plasmid gap repair ( O-MPGR ) . All modifications within the gene chimeras were confirmed by sequencing . Plasmid sequences are available upon request . Colocalization of excised DNA circles in M phase was measured as described in Chang et al . [9] , unless specified otherwise . To retain plasmids , selective media was used for pregrowth on dextrose and subsequent growth on raffinose overnight . When the cultures reached mid-log phase the following morning , an equal volume of YPA media plus raffinose was added . Twenty minutes later , nocodazole ( stock 1 mg/ml in DMSO , Cf = 10 µg/ml ) was added to initiate M phase arrest . After three hours , galactose ( Cf = 2% ) and benomyl ( stock 1 mg/ml , Cf = 10 µg/ml ) were added . Two hours later , cells were harvested , fixed and mounted on microscope slides with agar pads . Serial sections were obtained by fluorescence microscopy and GFP-foci/nucleus were counted manually . All measurements ( reported as the percentage of cells with single dots ) are based on at least three independent trials , which were pooled because they satisfied χ2 tests of homogeneity of proportions . Error bars represent the standard error of proportion . In each data panel , values were compared to an appropriate control by χ2 tests and judged as significant using a 95% confidence interval . To measure silencing of TRP1 inserted at HMR or two-hybrid interactions with the HIS3 reporter construct , plasmid-bearing strains were grown to saturation in selective media to retain plasmids and spotted in 10-fold serial dilutions . One set of selective plates was used to measure reporter gene expression and a second set was used as a loading control . Strain YFC9 ( MATα Δsir2 ) bearing Sir2-substitution plasmids was grown to saturation in selective medium , diluted 10-fold and then spotted on a lawn of mating tester strain K125 ( MATa ) on YPDA plates . After at least 5 hr at 30°C , the cells were replica plated to SD agar to measure mating and SC-trp as a loading control . Nocodazole was added to mid-log cultures that were either grown in YPDA overnight or that were sub-cultured in YPDA for 3 hours after overnight growth in selective media to retain plasmids . Three hours later , cross-linking and subsequent ChIP procedures were performed according to [22] using anti-TAP antibody ( Open Biosystems ) and Protein A beads ( Invitrogen ) . PCR reactions were run in multiplex using primer sets listed in Table S4 . Simultaneous amplification of a cohesin-free site ( 534 ) was included as an internal negative control of the immunoprecipitation reaction ( Figure S2 ) . Gels were stained with EtBr and destained in water before digital photography ( Alpha Innotech ) . All bands were found to be non-saturating and within the linear range . Reported values were calculated as ( a2/549 . 7 ) IP/ ( a2/549 . 7 ) In .
|
Replication of chromosomes in each cell cycle produces pairs of identical sister chromatids that are held together by a protein complex known as cohesin . At mitosis , cohesin is dismantled , permitting segregation of one full set of chromosomes to each daughter cell . Cohesin binds at discrete sites along chromatids , including domains that are commonly referred to as silent chromatin in budding yeast . Silent chromatin , like heterochromatin in higher eukaryotes , is a repressive structure that hinders a variety of DNA-based events . We discovered that a single silent chromatin constituent , Sir2 , was both necessary and sufficient for cohesion of silent chromatin domains . Sir2 is the founding member of the sirtuin family of NAD-dependent protein deacetylases that exist in most organisms . Substrate deacetylation by sirtuins has been linked to numerous pathways that promote health and survival in humans , including lifespan extension . Enrichment of cohesin at silent chromatin domains in yeast , however , is the first example of a role for Sir2 that does not explicitly require the protein deacetylase activity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology"
] |
2011
|
Targeted Sister Chromatid Cohesion by Sir2
|
RNA silencing is one of the main defense mechanisms employed by plants to fight viruses . In change , viruses have evolved silencing suppressor proteins to neutralize antiviral silencing . Since the endogenous and antiviral functions of RNA silencing pathway rely on common components , it was suggested that viral suppressors interfere with endogenous silencing pathway contributing to viral symptom development . In this work , we aimed to understand the effects of the tombusviral p19 suppressor on endogenous and antiviral silencing during genuine virus infection . We showed that ectopically expressed p19 sequesters endogenous small RNAs ( sRNAs ) in the absence , but not in the presence of virus infection . Our presented data question the generalized model in which the sequestration of endogenous sRNAs by the viral suppressor contributes to the viral symptom development . We further showed that p19 preferentially binds the perfectly paired ds-viral small interfering RNAs ( vsiRNAs ) but does not select based on their sequence or the type of the 5’ nucleotide . Finally , co-immunoprecipitation of sRNAs with AGO1 or AGO2 from virus-infected plants revealed that p19 specifically impairs vsiRNA loading into AGO1 but not AGO2 . Our findings , coupled with the fact that p19-expressing wild type Cymbidium ringspot virus ( CymRSV ) overcomes the Nicotiana benthamiana silencing based defense killing the host , suggest that AGO1 is the main effector of antiviral silencing in this host-virus combination .
Viruses are among the most important plant pathogens that cause huge economic losses in many agriculturally important crops worldwide . The invasion of the host by viruses deeply alters the physiology of the plants at cellular and tissue levels due to the interaction of the virus with the cellular pathways , which ultimately leads to viral symptom development . During evolution , plants have developed diverse strategies to combat virus infections . Amongst others , RNA silencing is one of the most important mechanisms that serve to fight against viruses [1 , 2] . RNA silencing is a conserved eukaryotic pathway involved in almost all cellular processes like development , stress responses and antiviral defense . RNA silencing relays on the 21–24 nt short interfering RNAs ( siRNAs ) or micro RNAs ( miRNAs ) the hallmark molecules of silencing [3] . The siRNAs and the miRNAs ( collectively named small RNAs , sRNAs ) are processed by RNase III-type ribonucleases , the DICER ( in plants Dicer-Like , DCL ) enzymes [4 , 5] in collaboration with their partner DOUBLE-STRANDED RNA BINDING ( DRB ) proteins [6–9] . sRNAs are 2’-O-methylated by HUA ENHANCER1 ( HEN1 ) at their 3’ protruding ends [10] , a reaction that serves to protect them against poly-uridylation and subsequent degradation [11] . sRNAs then associate with ARGONAUTE ( AGO ) proteins [12–14] the central effectors of RNA-induced silencing complex ( RISC ) [15 , 16] . Based on the sequence complementarity , sRNAs guide RISC to silence cognate RNAs through cleavage or translational repression ( post-transcriptional gene silencing , PTGS ) or induce chromatin/DNA modifications of the specific genomic locus ( transcriptional gene silencing , TGS ) [17–19] . In some specific cases , amplification of silencing occurs through double-stranded RNA synthesis by RNA-dependent RNA polymerases ( RDRs ) and secondary siRNA production [20–22] . sRNAs are non-cell autonomous , they can move within the plant to transmit gene silencing from cell-to-cell or systemically on long distance as mobile silencing signals [23–25] . Players of the antiviral silencing overlap with those of the endogenous silencing pathway [1 , 26] . Antiviral RNA silencing is triggered by the presence of viral dsRNA structures such as replication intermediates or intra—molecular fold—back structures of the invading virus . These dsRNA structures are processed by DCL4 or DCL2 , to produce viral short interfering RNAs ( vsiRNAs ) [2 , 3 , 27–32] . The vsiRNAs guide self-silencing of the parental viral genomic RNA as part of the antiviral response through the action of AGO effectors [1 , 13] . Among the AGOs implicated in antiviral defense , AGO1 and AGO2 were identified as the most important players . AGO1 was shown to have antiviral roles against a number of viruses in A . thaliana [33–36] , N . benthamiana [26 , 37 , 38] and in rice [39] . AGO2 was found to be important in the antiviral silencing response in A . thaliana [35 , 36 , 40–45] . In N . benthamiana , the important model organism for plant virology studies , AGO2 was proposed to protect against Potato virus X [46] and the suppressor-deficient Tomato bushy stunt virus ( TBSV ) [47 , 48] . However , recent observation suggested that AGO1 constitutes a solid layer of defense against tombusvirus infections [49] . As AGO1 is the negative regulator of AGO2 , it is believed that AGO2 represents a second layer of antiviral defense [40] . Viruses , to counteract host defense , have evolved viral suppressors of RNA silencing ( VSRs ) providing strong evidence for the antiviral nature of silencing [1 , 50 , 51] . Most viruses studied so far were found to encode at least one VSR . VSRs were shown to block silencing at multiple steps like initiation , effector complex assembly , silencing amplification but also through transcriptional control of endogenous factors , hormone signal modulation or interaction with protein-based immunity [51 , 52] . The absence or inactivation of VSRs leads to the recovery of plants from viral infections , demonstrating the effect of plant antiviral silencing response [53–55] . Although several VSRs have been identified in the past , our knowledge about the precise molecular basis of their action and their multifunctional roles have only been resolved in a few cases [1] . The p19 protein of tombusviruses is one of the best-known VSR . Crystallographic studies have shown that p19 tail-to-tail homodimer acts as a molecular caliper to size-select and sequester siRNA duplexes in a sequence-independent manner [56–58] preventing the loading of siRNAs into AGO effector proteins [59 , 60] . Based on p19 expressing transgenic plants it was proposed that during virus infection p19 efficiently prevents miRNAs loading into RISC deeply compromising the endogenous miRNA pathways of the plants [61–63] . In particular , it was reported that three distinct VSRs ( HcPro , p19 and P15 ) compromised the regulation of the miR167 target AUXIN RESPONSE FACTOR 8 ( ARF8 ) when constitutively expressed in transgenic plants [64] . It was also proposed that misregulation of miR167 is the major cause for the developmental aberrations exhibited by VSR transgenic plants and for the phenotypes induced during viral infections [64] . Contradictory to these , other data suggests that during genuine Cymbidium ringspot virus ( CymRSV ) infections miRNA sequestration by p19 is very poor and may depend on spatial and temporal co-expression of miRNA duplex and the VSR [65] . vsiRNAs but not miR159 were shown to be sequestered efficiently into p19-homodimer:siRNA nucleoprotein complex , whereas miR159 was efficiently incorporated into RISC complex [66 , 67] . These findings suggest that , in virus-infected plants , p19 potently affects vsiRNA-pathway and at the much lesser extent the miRNA one . Besides , independently of its siRNA binding capacity , p19 similarly to other VSRs can promote miR168 transcriptional induction that results in miR168-guided AGO1 down-regulation [66 , 68] . Thus , the interaction of p19 with endogenous silencing pathways and its contribution to viral symptom development is far from being fully uncovered . To better understand the impact of p19 on silencing and its role in viral symptom development we employed a synthetic p19-expressing transgenic N . benthamiana plant line ( p19syn ) or wild type plant as control in combination with its wild-type ( CymRSV ) or suppressor deficient virus ( Cym19stop ) infection: ( i ) wt virus infection of wt N . benthamiana ( p19 in “cis” ) , ( ii ) Cym19stop-infection in wt N . benthamiana ( virus present , no suppressor ) , ( iii ) Cym19stop-infection in p19syn plants ( p19 in “trans” ) and ( iv ) p19syn plants ( p19 only ) . In this way we were able to compare the impact of p19 on its own or in the genuine virus-infected background . We analyzed p19 ability to sequester vsiRNAs and plant endogenous sRNAs , with and without viral background . Consistently with our previous results , we have found that p19 can bind vsiRNAs when expressed either “in cis” ( from the CymRSV wild-type virus , in wild-type N . benthamiana ) or “in trans” ( transgenically expressed in p19syn plants infected with the suppressor-deficient Cym19stop virus ) . In line with our previous findings , p19 efficiently bound endogenous sRNA duplexes only in the absence of the virus infection , suggesting that p19 impact on endogenous pathways is restricted . Analyzing the siRNA pool immunoprecipitated by p19 through high-throughput sequencing , we found that p19 changes the bias of positive vsiRNAs towards a more equilibrated positive/negative strand ratio , suggesting a preference for perfect ds-vsiRNAs . We also showed that p19 prevents mainly AGO1 but not AGO2 loading with vsiRNAs . This finding suggests a key role of AGO1 opposed to AGO2 during the antiviral response .
To uncouple p19 effects elicited by virus infection on RNA silencing and host plant symptom development we prepared p19-expressing N . benthamiana plants ( Fig 1A–1C ) . To avoid the interference between the p19 transgene and the challenging virus ( p19-deficient , Cym19stop ) , we modified the p19 transgene introducing all possible silent nucleotide changes . In this way , we reduced the nucleotide sequence similarity between the transgene and the challenging virus to 68% while keeping the amino acid identity at 100% ( S1 Fig ) . These plants were named synthetic-p19 expressing plants ( p19syn ) . The p19syn plants showed strong phenotype characterized with elongated stem internodes and typical leaf distortions ( Fig 1A , 1C and 1D and S2A Fig ) suggesting that the expressed p19 protein retained it suppressor activity , thus potentially compromising the endogenous silencing pathways . Importantly , this phenotype was clearly different from that of virus-infected stunted dwarf plant ( S2A Fig ) . Transgenic line 1–57 was selected for further studies ( Fig 1A and 1B ) . First we tested the silencing suppressor activity of transgenically expressed p19 in a GFP transient assay ( see Materials and Methods ) . When GFP sense transgene was transiently expressed in wild-type plant leaves , spontaneously triggered silencing almost completely diminished GFP expression at four days post infiltration . In contrary when GFP was expressed in p19syn plants its expression was still strong as visualized under UV light ( Fig 1D ) . The lack of GFP silencing in p19syn plants confirmed the suppressor activity of the p19 transgene . Next we tested p19 suppressor activity in an authentic virus infection context: we challenged the p19syn plants by the infection with Cucumber mosaic virus + yellow satellite RNA ( CMV + Y-satRNA ) . CMV + Y-satRNA was reported to induce bright yellow symptoms on N . benthamiana through targeting the tobacco magnesium protoporphyrin chelatase subunit I ( ChlI ) gene involved in chlorophyll biosynthesis by Y-satRNA derived siRNA [69] . The CMV-Y-satRNA infected wt N . benthamiana plants developed the bright yellow symptoms while the infected p19syn plants failed to show the typical yellowing ( Fig 1E ) . All these confirmed that the transgenically expressed p19 works as a silencing suppressor in vivo . It is generally assumed that virus encoded suppressors strongly interfere with the endogenous silencing pathway and are central players in the development of viral symptoms [1 , 61–64] . However , this notion mostly comes from studies that used VSR-expressing transgenic plants without analyzing the effect of the VSRs in an authentic virus infection background . To reinvestigate this dogma we set up an experiment in which we could compare p19 effects ( vsiRNA and endogenous sRNA binding ) with or without its parental virus infection background . We compared the sRNA binding capability of p19 both in mock- and Cym19stop-inoculated p19syn plants . This setup allowed us to analyze the impact of p19 when provided “in trans” during virus infection . It is worth noting that the suppressor mutant Cym19stop virus in infected p19syn plant was able to invade whole leaves similarly to the CymRSV in wt plants ( S2B Fig ) . In contrast , in the absence of p19 , the Cym19stop virus is restricted to the veins [70] ( S2B Fig ) . Besides this , we also inoculated wt plants with CymRSV to study p19 activity “in cis“ . Based on previous studies [61 , 62] we expected p19 to bind ds-sRNAs of both plant and viral origin . p19 immunoprecipitations ( IP ) were prepared from pooled systemically-infected leaves of virus-infected plants and the corresponding mock-inoculated leaves of p19syn plants . cDNA libraries of sRNAs were generated using RNA samples isolated from inputs and p19 IPs . After quality control filtering and processing steps ( see Materials and Methods ) , sequences flanked by the 3’ and 5’ Solexa adaptors , and ranging in length from 16 to 28 nt , were aligned to the N . benthamiana and the CymRSV genomes , respectively [71 , 72] . Analysis of p19-bound sRNAs from mock-inoculated p19syn plants revealed that p19 binds efficiently endogenous sRNAs ( Fig 2A and 2B ) , including members of several miRNA families ( S3A Fig ) [73] . Surprisingly , the analysis of p19-bound sRNA libraries derived from both CymRSV-infected wt N . benthamiana ( “cis“-p19 ) and Cym19stop-infected p19syn plants ( “trans-p19“ ) have shown a different picture: p19 bound almost exclusively vsiRNAs but not endogenous sRNAs ( Fig 2A and 2C–2F ) . This suggests that the abundantly produced vsiRNAs may outcompete the plant sRNAs from p19 binding during virus infection . Specific enrichment of vsiRNAs versus endogenous miR159 , one of the most abundant miRNA , was quantified by Northern blot analysis . p19 had a much weaker affinity for miR159 during virus infection: the IP/input ratio of p19 bound miR159 was 1 . 2 mock-inoculated samples , while during virus infection ( Cym19stop-infected p19syn plants ) it dropped to 0 . 29 ( Fig 3A ) . We also quantified the percentage of enrichments in case of vsiRNAs and endogenous sRNAs within p19 IPs compared to inputs from our deep seq data ( Fig 2A ) . The input library of Cym19stop-infected p19syn plants contained 28% N . benthamiana reads while in the p19-IP they represented only 2% ( p19 specifically enriched vsiRNAs from 72% in the input to 98% in the p19-IP ) . Similarly , in the CymRSV-infected wt plant 12% N . benthamiana reads in the input sample has dropped to 1% ( p19 enriched the 88% vsiRNAs of the input to 99% in the p19-IP ) . We concluded therefore that p19 ability to sequester endogenous sRNAs is strongly decreased by the virus infection and p19 preference to vsiRNAs does not depend on the expressional origin of p19 protein ( viral vs transgenic expression ) . To better understand the biological relevance of vsiRNA-mediated endogenous sRNA binding and out-competition/release from p19 sequestration we analyzed the behavior of known miRNA-target mRNA pairs [73] . We compared RNAseq data obtained from mock-inoculated p19syn plant samples ( when p19 binds to miRNAs ) and from Cym19stop virus-infected p19syn plant samples ( when p19 binds preferentially vsiRNAs while miRNAs are outcompeted/released ) . In the absence of the virus , p19 efficiently bound miRNA duplexes ( S3A Fig ) and this correlated with elevation of most of the miRNA-target mRNAs as the consequence of miRNA duplex sequestration by p19 and inability to program miRISC for cleavage ( p19syn compared to wt N . benthamiana , S3B Fig ) . Upon Cym19stop virus infection however , the levels of most miRNA target RNAs were downregulated ( compared to mock-infected p19syn ) as the consequence of miRNA out-competition/release from p19 ( S3B Fig ) . We went further and specifically looked to accumulation of trans-acting RNAs derived from TAS3 precursor , the target of miR390 [74] in a Northern blot assay ( S3C Fig ) . In p19syn plants the level of miR390 was slightly elevated while the TAS3-derived D7 tasiRNA dropped below the detection level ( compared to wt N . benthamiana ) . This was likely the consequence of the inhibition of the cleavage of TAS3 transcripts by p19-captured miR390 . Indeed , miR390 is efficiently enriched in p19 IP ( p19syn mock-infection , S3A Fig ) . When p19syn plants were infected with Cym19stop virus , miR390 binding by p19 decreased ( S3A Fig ) , and consequently the activity of miR390 was restored that lead to D7 tasiRNA accumulation ( to a similar level as detected in wt N . benthamiana , S3C Fig ) . Altogether our findings support the hypothesis that during virus infection p19 preferentially binds vsiRNAs while endogenous sRNAs are outcompeted/released from binding . High-throughput sequencing analysis showed that CymRSV-derived vsiRNAs produced during infection have a strong bias towards positive strand polarity ( 95% positive , 5% negative polarity ) ( Fig 3B and S4 Fig ) . These data , which are in line with our and other previous observations , suggest that the majority of vsiRNAs are produced from fold-back structures of the positive strand of the viral RNAs [27 , 30 , 31 , 75] . Hot spots of vsiRNA generation were observed ( S4 Fig ) as earlier [31 , 75] . The polarity analysis of p19-immunoprecipitated vsiRNAs revealed a more equilibrated positive/negative strand ratio ( significant enrichment in negative strand derived vsiRNAs with 65% positive , 35% negative strands ) in the CymRSV-infected plants ( Fig 3B ) . In Cym19stop-infected p19syn plants 79% of vsiRNAs produced were positive-stranded ( 21% negative ) , while p19syn—immunoprecipitated the ratio changed to 62% positive , 38% negative ( Fig 3B ) . Based on these we conclude that p19 preferentially enriches positive-negative ds-vsiRNA pairs possessing perfect duplex structure . To formally test the impact of mismatches within the duplex sRNAs on the affinity of p19 we compared the affinity of p19 protein towards the miR171 duplex miRNA family ( Arabidopsis miR171a , miR171b , miR171c all containing mismatches ) and a perfect artificial siR171 duplex ( for structures see Fig 4A–4D ) using in vitro electro-mobility shift assay [57] . The presence of mismatches within the stem of sRNAs strongly reduced p19 binding affinity towards duplex sRNAs ( Fig 4A–4F ) . Consistent with our findings , it has been also shown previously that p19 preferentially binds to perfect sRNAs duplexes but not imperfect miRNAs duplexes [76] . We have also analyzed the 5’-nucleotide preference of p19 binding . No 5’-nucleotide sorting of vsiRNAs in the p19 complex was observed regardless of p19 expressional context ( from the virus or the transgene ) ( Fig 2C and 2E ) . The relative abundance of sRNAs possessing different 5’-nucleotides closely followed the ratio of the input samples . The distribution of p19 bound vsiRNAs along the viral genome was found to be similar to that in the input ( S4 Fig ) showing that there is no sequence preference in p19 vsiRNA binding . To better understand the p19 protein effects on vsiRNAs we analyzed the size distribution of these during infection . Upon CymRSV infection vsiRNAs produced are predominantly of 21nt and 22nt in length ( Fig 3C ) . In addition to these , the 20nt long vsiRNAs are still present although at much lower level . In Cym19stop virus-infected wt plants we observed a shift towards slightly longer forms: most of the vsiRNAs were 22nt long , the abundance of the 21nt and 20nt long forms being reduced ( Fig 3C ) . These results are in line with our previous findings [31] . miRNAs having enhanced electrophoretic mobility were also detected earlier in the presence of p19 [62 , 77] . To test if shortening is indeed an effect of p19 protein itself we analyzed parental virus- ( Cym19stop ) derived vsiRNA when p19 was provided “in trans“ . The length shift to 1- or 2-nucleotide shorter vsiRNA forms was confirmed ( Fig 3A–3C ) . Shortening of endogenous miRNAs was also observed in the absence of virus infection ( Fig 3D ) . Analysis of selected endogenous miRNAs , where the precise sequence and biogenesis/maturation are known , allowed us to establish that the truncation occurred at the 3’ end but not 5’ end . The truncation of miRNAs happened mainly in p19-sRNA complexes as was observed in p19-IP , however not all p19 bound miRNAs are truncated and the reason for this has not been clarified yet ( Fig 3D ) . Nuclease treatments on in vitro bound p19:siRNA complexes further confirmed that the p19 protein can protect the double-stranded stretch of sRNA duplexes ( S5 Fig ) . The exonuclease ( RNaseA ) -mediated digestion occurred in discrete 1- and 2-nucleotide steps while the dsRNA region ( 19nt length ) was protected by the p19 protein . Shortening of sequestered sRNAs , therefore , is not dependent on the virus infection , occurs on 3’ end , involves both vsiRNAs and miRNAs and is likely the direct consequence of p19-binding and exonuclease activity . Multiple AGOs were shown to have antiviral functions . In A . thaliana and N . benthamiana AGO1 and AGO2 were described as the most important effectors while others such as AGO5 , 7 and 10 to have minor roles during antiviral silencing [13 , 34 , 36 , 44] . The current model of the inhibitory effect of p19 suggests that sequestering vsiRNAs prevents AGO loading . The inhibitory effect of p19 protein on RNA silencing during infection was quite evident . In fact , wt virus infection showed strong viral symptoms that culminated in complete necrosis and collapse of the plants while the Cym19stop-infected wt plants recovered from viral infection [70] , the virus accumulation was restricted to the vascular tissues and a few cell layers around the veins ( [70] and S2 Fig ) . To get a better insight into the detailed mechanism of p19 actions we analyzed the AGO1- and AGO2-bound vsiRNAs in CymRSV- and Cym19stop-infected wt N . benthamiana by co-immunoprecipitations ( Fig 5 ) followed by deep sequencing analysis ( Fig 6 and S6–S10 Figs ) . Loading of siRNAs into a particular AGO is preferentially directed by their 5’-terminal nucleotide: AGO1 prefers sRNAs having 5’U while AGO2 preferentially binds 5’A sRNAs [78 , 79] . As expected , the AGO1 co-immunoprecipitated plant sRNAs possessed predominantly 5’U while AGO2 immunoprecipitated sRNAs mainly 5’A , with a relatively high amount of 5’U species ( Fig 6 ) . We have also found 5’U endogenous sRNA binding by AGO2 when we processed the raw data obtained from previous report [36] . We expected that p19 would drastically reduce the loading of vsiRNAs into AGO1 and AGO2 . Surprisingly , vsiRNA loading into AGO1 was compromised in the presence of p19 ( during CymRSV-infection compared to Cym19stop-infection ) : we observed relatively high “background” of vsiRNAs without 5’ sorting preference in AGO1 ( compare Fig 6D with 6F ) . Conversely , the amount of vsiRNAs and their 5’ sorting into AGO2 was very similar during CymRSV- and Cym19stop- infections ( Fig 6D–6F ) . This suggests that the presence of p19 preferentially impact vsiRNAs’ AGO1 but not AGO2 effector loading . In the same time ( the same sample set ) endogenous sRNAs were efficiently precipitated as 5'U by AGO1-IP proving that the IP worked correctly ( Fig 6C–6E ) . Note that the reads of endogenous sRNAs in CymRSV and Cym19stop are lower compared to mock-infected sample due to the high amount of vsiRNA presence ( that impacts the bias during deep sequencing ) . The AGO1 IP derived from CymRSV-infected plants contained a similar miRNA profile as the mock inoculated plants , in contrast to AGO2 IP in which the levels of analyzed miRNAs were reduced ( Fig 6C and 6D and S6 and S7 Figs ) . Importantly , this occurred only in wt CymRSV infection when a high level of p19 is expressed . Efficient incorporation of vsiRNAs into AGO2 but not AGO1 may cause the out-titration of AGO2-bound endogenous sRNA species ( during CymRSV infection ) . In contrast , during Cym19stop-infection AGO1-loading occurred as expected predominantly by 5’U-sorting of vsiRNA and endogenous sRNAs ( Fig 6E and 6F ) . The obtained results were confirmed with a second AGO1 and AGO2 IP that gave very similar results although had slightly higher background of contaminating 24nt species ( S9 Fig ) . These findings suggest that p19 protein itself compromises AGO1- but not AGO2-loading during viral infection . The specific impact of p19 on vsiRNA AGO1-loading found in the deep sequencing analysis was also confirmed by Northern blot analysis . vsiRNAs loading into AGO1 was less efficient than AGO2 in CymRSV-infected plants when the p19 was provided in “cis” ( Fig 5A ) or in “trans” when two independent p19syn lines were infected with Cym19stop virus ( Fig 5B ) . We also analyzed the distribution of AGO1- and AGO2-bound vsiRNA along the viral genome . This generally followed the biogenesis of vsiRNAs and we could not define any sequence preference or specific hotspots of AGO1- or AGO2-loading ( S8 Fig ) . The strong spikes of certain vsiRNAs may arise due to the sequencing bias , therefore , do not necessarily represent vsiRNAs preferred for binding [80] .
During virus infection , high amounts of vsiRNAs are produced . These vsiRNAs are efficiently sequestered by p19 suppressor inhibiting their incorporation into RISC [67] . The consequence of p19 vsiRNA binding is that the strong positive strand bias of vsiRNA biogenesis in the input sample ( 95:5 positive/negative ) is changed to a more equilibrated positive/negative stand ratio ( 65:35 positive/negative ) . This result suggests that there are qualitative structural differences between vsiRNAs and that p19 preferentially binds vsiRNAs derived from perfect dsRNA or highly structured RNA species . The preference of p19 towards even structured ds-vsiRNAs is in agreement with p19 crystal structure: p19 homodimer leans on the ds-sRNA backbone . If the backbone structure is distorted by mismatches , the sRNA could become less accessible to p19 sequestration . Indeed , p19 bound siR171a perfect duplex with higher affinity than natural miR171a , miR171b or miR171c duplexes ( Fig 4 ) . This may likely be one of the reasons why ds-vsiRNAs are preferred by p19 instead of mismatches containing endogenous miRNA duplexes during viral infection ( Figs 2 and 3A ) . The excess of vsiRNAs over endogenous miRNAs in virus-infected plant may also contribute to the preferential binding of vsiRNAs by p19 . Moreover , the difference in the biogenesis of miRNAs versus vsiRNAs could also be a further important factor in the mechanism of the sRNA sequestration by p19 . In addition to the previous findings , we have also shown that sRNA binding by p19 happens without 5’-end nucleotide selection including vsiRNAs or endogenous miRNAs . Previous studies [62 , 77] have reported the truncation of p19-bound sRNAs by 1 or 2 nucleotides . In the case of vsiRNAs , the site of truncation ( 5’ vs . 3’ end ) cannot be defined since the generated vsiRNAs started from almost every single nucleotide of viral genome ( S4 and S8 Figs ) . During Cym19stop-infection on wild-type plants truncation does not occur while in p19syn plants , which provide p19 in trans , it can be observed . Shortening also happens on miRNAs in p19syn plants without virus infection . In summary , the truncation is likely induced directly by p19-binding and is not due to the virus infection or restricted to a specific class of sRNAs . Why shortening does not happen in the absence of p19 , on the free vsiRNAs , which theoretically would be more accessible ? In cells endogenous free sRNA duplexes ( e . g . , miR168/miR168star ) [54] and free vsiRNAs in Cym19stop infection can be observed [32 , 67] . The stability of these sRNAs ( miRNAs and vsiRNAs ) is conferred by HEN1-mediated methylation [11] . The crystal structure of p19:siRNA complex shows that the last two single-stranded nucleotides at 3’ terminus of siRNAs are protruding from the complex [57 , 58] . Furthermore , the p19 bound vsiRNAs are not methylated at 3’ terminus [65] therefore may be sensitive to exonucleases . We propose therefore that p19 binding inhibits sRNAs methylation and as a consequence of this the protruding unprotected two nucleotides at the 3’-end of sRNAs are trimmed by cellular 3’-exonucleases . Whether the truncation of sRNAs is a simple byproduct of binding or has a definite biological importance remains to be seen . Trimming of sRNAs may inactivate and render them incompetent for AGO-loading . Contradictory to this , we find efficient binding of 19nt and 20nt vsiRNAs by AGO2 ( Fig 6D , S9D Fig ) . This observation supports the “catch and release” of vsiRNAs by p19 proposed earlier [81] . It has been long suggested that VSRs interfere with endogenous silencing pathways , and this may contribute to the viral symptom development [1 , 61–64] . Constitutive expression of p19 in N . benthamiana leads to the development of a strong phenotype that is quite different from symptoms observed during parental viral infection ( Fig 1 and S2 Fig ) . The strong phenotype of p19syn plants may arise , at least in part , due to the sequestration of endogenous sRNAs by p19 . Indeed , we could immunoprecipitate endogenous miRNAs with p19 from transgenic plants ( Fig 2B ) . Importantly , however , miRNA sequestration by p19 provided either in cis or in trans was drastically reduced when the virus was present ( Fig 2C–2F ) . Importantly , miRNA out-competition/release correlated with downregulation of miRNA targets ( S3B Fig ) and reestablishment of tasiRNA biogenesis ( S3C Fig ) . Out-competition/release of p19-bound endogenous sRNAs/miRNAs upon virus infection seems to be biologically relevant and could have an important role in moderating the virus impact on plant . This needs to be further investigated . In conclusion therefore , our findings deny the model in which miRNA binding by p19 is the key step for the development of virus-induced symptoms [1 , 61–64] . It is more likely that p19 has an indirect effect through the specific inhibition of antiviral plant response and the viral symptoms are the outcome of a complex virus-host interaction during the viral invasion of plant cells . What could be the criteria for vsiRNA selection by AGO-loading machinery ? We previously observed that in wt tombusvirus infection , p19 protein prevents vsiRNA loading to AGO/RISC complexes , however , even in the absence of p19 only a small fraction of the vsiRNAs is loaded into effector complexes ( Fig 5 ) the rest remaining in a free , probably double-stranded form [32 , 67] . This suggests that a big part of the abundantly produced vsiRNAs is AGO-incompetent , or there is no free AGO protein present to be loaded into . The structure of the ds-sRNA stem could be an important feature for vsiRNA selection into AGOs ( as we have shown for p19 ) . A similar analysis of sRNA duplexes as in the case of p19 cannot be done , since in p19 binding both strands of ds-sRNAs are retained , while in AGOs , after loading , one strand is eliminated . The other possibility of the inability of vsiRNAs to load into effector complexes could be the shortage of silencing proteins like DCL/DRB or AGOs during the assembly of these effectors . It was shown that specific regulatory mechanisms are induced by the virus to dampen silencing: translation of AGO1 protein is decreased by the suppressor-mediated miR168 over-accumulation [54 , 66] . However , the down-regulation of AGO1 protein was not observed in p19syn plants ( Fig 1B ) . The reason for this could be the relatively low level of p19 produced from transgene compared to virus infection ( Fig 5 ) . Another important observation is that vsiRNAs loading is selectively prevented mainly into AGO1 but not AGO2 in the presence of p19 in both CymRSV infected wt or Cym19stop infected p19syn plants ( Fig 5 ) . During virus infection , the decrease in the translation of AGO1 protein leads to accumulation of AGO2 , due to the absence of AGO1-miR403-mediated posttranscriptional down-regulation of AGO2 [54 , 66] . One possibility , therefore , is that the vsiRNAs will be loaded into the available AGO2 while AGO1-loading will be decreased . We could not observe a significant increase of AGO2 protein in the presence of p19 ( Fig 1B ) . The other possibility is that AGO1 and p19 compete for the same set of vsiRNAs while AGO2 requirement for vsiRNA features is different . p19 therefore , would selectively impact AGO1 but not AGO2-loading . Interestingly , we observed relatively high p19-depending “background” of vsiRNAs without 5’ sorting in AGO1 IP , unlikely to be AGO1-incorporated sRNAs . p19 could affect the connection between biogenesis/loading complexes DCLs/DRBs with AGO1 effector [7 , 9] . In line with this hypothesis , it was demonstrated that p19 can compromise the transfer of siRNA from DICER-R2D2 into RISC complex using Drosophila embryo extracts based in vitro system [59] . Regardless of the reason of how AGO1-loading is compromised by p19 , it seems that AGO2 is not enough to fight off the virus and help the plant to recover in the absence of AGO1-loading/activity . It has been suggested previously that AGO2 but not AGO1 plays role in the antiviral response against tombusvirus infections , including Tomato bushy stunt virus ( TBSV ) [47 , 48] . We have done TBSV-VIGS ( Virus Induced Gene Silencing ) experiment , using p19 inactivated virus vector ( TBSVp19stop ) , which carried Nb-PDS and Nb-AGO1 sequence ( S11 Fig ) . When NbAGO1 was silenced by VIGS the virus accumulated at higher level and plants have shown stronger phenotype ( S11 Fig ) . The obtained results further support the idea that AGO1 has a major role in antiviral response against tombusvirus infection . However , the additional role of other plant AGOs in antiviral response remained to be explored and it likely depends on specific features of the highly diverse plant viruses . The availability of CRISPR/Cas9 system for plant research will also help to clarify the specific roles of plant effectors in antiviral silencing response .
The synthetic CymRSV ORF5 ( p19 ) was essentially constructed following the previously described antivirus-induced transgene silencing strategy [82] . As the first step , we introduced , in CymRSV ORF5 , all possible silent point mutations by selecting those most compatible with the N . tabacum codon usage . The resulting nucleotide sequence was further modified to avoid the presence of cryptic splicing and polyadenylation signals using Net2gene splicing prediction ( http://www . cbs . dtu . dk/services/NetGene2/ ) and HCpolyA ( http://bioinfo4 . itb . cnr . it/~webgene/wwwHC_polya_ex . html ) software , respectively . The synthetic ORF5 ( S1A Fig ) was synthesized by Life Technologies and cloned in pJIT61 [83] between the CaMV 35S promoter and 35S terminator . The gene cassette was excised with KpnI and BglII and cloned in KpnI-BamHI of pBin19 ( pBinCymRSVp19syn ) N . benthamiana was transformed with the recombinant Agrobacterium tumefaciens strain C58C1 ( pGV2260 ) harboring the plasmid pBinCymRSVp19syn , and kanamycin-resistant plants were regenerated as previously described [83] . The primary transformants were checked for the presence of p19syn transgene by PCR and for the expression of the p19 protein by Western blotting with the anti-CymRSV-p19 antibody as previously described [56] . N . benthamiana plants were grown at 22°C . At six-leaves stage plants were infiltrated with A . tumefaciens C58C1 harboring the appropriate constructs in the pBIN61 plasmid . pBIN61-Cymp19 and pBIN61-GFP were grown on selective media overnight , resuspended in the infiltration buffer ( 10 mM MES , 0 . 15 mM acetosyringone , 10 mM MgCl2 ) kept on 25°C for 4h , and subsequently infiltrated into wild-type or p19syn plant leaves at OD600 = 0 . 4 . In vitro transcription of CymRSV , Cym19stop , TBSV-PDS-GFP and TBSV-PDS-AGO1-1 RNAs from linearized template plasmids and inoculation of RNA transcripts onto N . benthamiana plants were performed as described previously [84] . CMV Y-sat infection was performed as described earlier [69] . Total RNA was extracted from 100 mg of leaf tissue . The homogenized plant materials were resuspended in 600 μl of extraction buffer ( 0 . 1 M glycine-NaOH , pH 9 . 0 , 100 mM NaCl , 10 mM EDTA , 2% SDS ) and mixed with an equal volume of phenol . The aqueous phase was treated with equal volumes of phenol-chloroform and chloroform , precipitated with ethanol , and finally resuspended in sterile water . RNA gel blot analysis of higher molecular weight RNAs was performed as described previously [84] . RNA gel blot analysis of 21–24 nt RNAs was performed as follows . Approximately 5 μg of total RNA was separated by 15% PAGE with 8 . 6 M urea and 1xTris-borate-EDTA . RNA was electro-blotted onto Hybond-NX membranes and fixed by chemical crosslinking at 60°C for 1 hr [85] . Small RNA Northern blot hybridization and analysis were performed using complementary locked nucleic acid ( LNA ) oligonucleotides ( Exiqon , http://www . exiqon . com ) . Mock- or virus-infected systemical leaf tissues were homogenized in extraction buffer ( 150 mM Tris-HCl , pH 7 . 5 , 6 M urea , 2% SDS , and 5% μ-mercaptoethanol ) . Samples were boiled , and cell debris was removed by centrifugation at 18 , 000 x g at 4°C for 10 min . The supernatants were resolved on 12% SDS-PAGE , transferred to Hybond PVDF membranes ( GE Healthcare ) and subjected to Western blot analysis . For detection anti-p19 [70] , NbAGO1 [86] and NbAGO2 custom antibody were used . NbAGO2 antibody was generated by immunization of rabbits with the synthetic peptide ( CLEDPEGKDPPRDVF ) ( GenScript , http://www . genscript . com/ ) . The proteins were visualized by chemiluminescence ( ECL kit; GE Healthcare ) according to the manufacturer’s instructions . For immunoprecipitation , 1–5 grams of mock- , CymRSV- or Cym19stop-infected N . benthamiana leaves showing systemic symptoms ( or leaves at the same stage and positions from mock-inoculated plants ) were collected , ground in 1:3 ( w/v ) amount of immunoprecipitation buffer ( 40 mM HEPES/KOH 7 . 4 , 100 mM KOAc , 5 mM MgOAc , 5% glycerol , freshly added 4 mM DTT ) , and cleared by centrifugation ( twice at 15 , 000 x g for 5 min ) . Cleared lysates were kept on ice until immunoprecipitation with antibody-coated protein A-Sepharose ( GE Healthcare ) . Beads were washed before adding the antibodies ( described earlier ) . For mock immunoprecipitation preimmune serum was used . Antibodies coated beads were incubated with the relevant cleared lysates for 4h at 4°C . After immunoprecipitations the beads were washed five times with ice-cold immunoprecipitation buffer for 2 min each . Input extracts and eluates of immunoprecipitations were used for Western and Northern blot analysis . The library preparation was described previously [31] , shortly: RNA samples were purified by cutting the sRNA region from 8% denaturing polyacrylamide gels ( acrylamide:bisacrylamide ( 19:1 ) 1xTBE , 8 . 6 M urea ) . After gel electrophoresis , the gels were stained by SYBRGold ( Thermo Fisher Scientific ) . Bands at the small RNA range were cut out and crushed . Gel particles were shaken overnight in RNase free water at 4°C , followed by RNA isolation ( described above ) . TruSeq Small RNA Sample preparation kit ( Illumina ) was used for library preparation; we followed the manufacturer’s protocol . In the case of the AGO1- and AGO2- immunoprecipitation 9 libraries were pooled together . In the case of the p19 immunoprecipitations , 4–4 libraries were pooled together . The libraries were sequenced on Illumina HiScanSQ platform ( UD-GenomMed Medical Genomic Technologies Ltd . , Debrecen , Hungary ) that yielded approximately 100 M reads per lane ( 50bp , single end ) ( S10 Fig ) . RNA-seq library preparation was done according to the manufacturer protocol ( TruSeqStranded mRNA Library Prep Kit ) . The libraries were sequenced on Illumina HiScanSQ . 135 M 100 bp paired-end reads were produced per lane . 3 samples were pooled together on a lane . The libraries were submitted to GEO and can be accessed through series accession number GSE77070 . One hundred nanograms of synthetic sRNA ( 5’-UGAUUGAGCCGCGCCAAUAUC-3’ ) was 5’ end labeled by T4 polynucleotide kinase ( Fermentas ) with 32P isotope . After the reaction was stopped , 10 ng was saved for further process and the rest was mixed with 500 ng of unlabeled synthetic RNA with the sequence of 5’-UAUUGGCGCGGCUCAAUCAGA-3’ . The mixture was heated to 95°C for 2 minutes in a thermocycler and was cooled to 5°C ( 2°/2 min ) to gain 19 nt perfectly matched double strand with 2 nt protruding at the 3’ end . The sufficient amount of DNA loading dye was added to the mixture and to the saved labeled single stranded RNA . Both samples were run on a 8% acrylamide:bis-acrylamide 19:1 1xTBE gel . Gel was directly exposed . The dsRNA region was cut from the gel . The gel piece was shredded using a 0 . 5 ml tube with several holes in the bottom in a 2 ml tube via centrifugation . The shredded gel pieces were shaken overnight in 500 μl of 300 mM NaCl at 4°C . Gel pieces were filtered out by using Spin-X column ( Corning ) . 400 μl of dsRNA solution was precipitated by adding 20 μg glycogen ( Fermentas ) and 1 ml ethanol . The precipitated ds-RNA was resolved in IP buffer described before . Purified p19 described earlier was used for the assay [56] . A dilution series of p19 ( ~1 μg ) was made in 1x IP buffer . An equal amount of gel-purified ds-siRNA was added to each p19 dilution and incubated at room temperature for 10 min . Then 10 ng of RNase-A ( Sigma ) was added and incubated for 10 and 30 min at room temperature . After incubation the samples were mixed with DNA dye and ran on a 16% acrylamide: bis-acrylamide ( 19:1 ) 1xTBE gel . Decade marker ( Ambion ) and synthetic RNAs were used as size markers . Gels were dried and were directly exposed . For band shift assays wild-type p19 protein was purified from E . coli as described previously [56 , 67] . Custom RNAs used were ordered from Dharmacon , ( http://dharmacon . gelifesciences . com ) for sequence see Fig 4 . Labeling and annealing of si/miRNA duplexes was carried out as described previously [67] . Purified p19 protein and labeled si/miRNAs were incubated for 30 min at room temperature in band-shift buffer [67] . Complexes were resolved on 8% polyacrylamide 0 . 5xTBE gels . Gels were dried and exposed to a storage phosphor screen and bands quantified ( Molecular Dynamics Typhoon Phosphorimager , GE Heathcare ) . In situ hybridization was performed as previously described in [87] . Detection of viral RNA expression patterns were made by using nonradioactive in situ hybridization on histological sections of leaf tissues . Digoxigenin labeled antisense RNA probe was synthesized by in vitro transcription from the linearized CymRSV construct . After demultiplexing of the raw data we used the UEA workbench version 3 . 0 [88] for adapter removal . Quality control consisted of filtering out reads with less than 14 nt ( without the adapter sequences ) and reads showing low complexity . We used PatMaN v1 . 2 . 2 . [89] to align reads allowing 0 mismatches . Reads not matching either genome were removed . Reads passing quality control is referred as “total” in this article . In Fig 3B reads were normalized to 1 million viral reads . In S7 Fig reads were normalized to 1 million N . benthamiana genome matching reads . In all other cases reads were normalized to 1 million total reads . After demultiplexing we used FastQC 0 . 10 . 1 to check general attributes . Trim_galore 0 . 4 . 1 and FASTX Toolkit 0 . 0 . 13 were used to remove adaptor sequences , low quality bases , reads under 20 nt and unpaired reads . Bowtie2 [90] was used to align reads to Nbv5 [91] transcriptome database . Reads were counted for homologs of known miRNA targets . NCBI-blast+ 2 . 2 . 28 [92] was used to validate miRNA targets . Samtools 0 . 1 . 19-96b5f2294a was used during alignment evaluation . Read counts were normalised to 1 million total reads .
|
To better understand the specific effect of p19 viral suppressor of RNA silencing ( VSR ) on antiviral silencing and endogenous small RNA pathways , we generated a N . benthamiana plant ( p19syn ) capable of sustaining the ectopic expression of the Cymbidium ringspot virus ( CymRSV ) p19 upon infection with a suppressor-deficient CymRSV ( Cym19stop ) . By using wt and p19syn plants in combination with CymRSV and Cym19stop , we were able to analyze the effects of p19 provided “in trans” and “in cis” during the viral invasion of the plant . We have shown that p19 can efficiently sequester endogenous small RNAs ( sRNAs ) in mock-inoculated p19syn plants while it does not bind these sRNAs upon Cym19stop infection . Also , the presence of p19 in virus infection did not alter the expression of miRNAs significantly . These findings do not support the widely accepted assumption that viral symptoms are the direct consequence of the impact of VSRs on endogenous silencing pathways . We demonstrated that p19 preferentially sequesters positive:negative viral short interfering RNAs ( vsiRNAs ) pairs and that the binding by p19 is independent of vsiRNA sequence or the type of the 5’-end nucleotide . We have also found that 3’ truncation is induced on p19 bound sRNAs . Finally using AGO1- and AGO2- immunoprecipitation experiments we observed that p19 specifically compromises vsiRNAs’ loading into AGO1 but not AGO2 . Since antiviral silencing is strongly inhibited by p19 , this suggests that AGO1 is the main effector protein against CymRSV tombusvirus .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biotechnology",
"plant",
"anatomy",
"rna",
"interference",
"gene",
"regulation",
"viruses",
"micrornas",
"plant",
"science",
"rna",
"viruses",
"immunoprecipitation",
"genetically",
"modified",
"plants",
"epigenetics",
"plants",
"flowering",
"plants",
"genetic",
"engineering",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"genetically",
"modified",
"organisms",
"nicotiana",
"genetic",
"interference",
"gene",
"expression",
"leaves",
"precipitation",
"techniques",
"agriculture",
"biochemistry",
"rna",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna",
"agricultural",
"biotechnology",
"plant",
"biotechnology",
"organisms"
] |
2016
|
Distinct Effects of p19 RNA Silencing Suppressor on Small RNA Mediated Pathways in Plants
|
Fibromuscular dysplasia ( FMD ) is a nonatherosclerotic vascular disease leading to stenosis , dissection and aneurysm affecting mainly the renal and cerebrovascular arteries . FMD is often an underdiagnosed cause of hypertension and stroke , has higher prevalence in females ( ~80% ) but its pathophysiology is unclear . We analyzed ~26K common variants ( MAF>0 . 05 ) generated by exome-chip arrays in 249 FMD patients and 689 controls . We replicated 13 loci ( P<10−4 ) in 402 cases and 2 , 537 controls and confirmed an association between FMD and a variant in the phosphatase and actin regulator 1 gene ( PHACTR1 ) . Three additional case control cohorts including 512 cases and 669 replicated this result and overall reached the genomic level of significance ( OR = 1 . 39 , P = 7 . 4×10−10 , 1 , 154 cases and 3 , 895 controls ) . The top variant , rs9349379 , is intronic to PHACTR1 , a risk locus for coronary artery disease , migraine , and cervical artery dissection . The analyses of geometrical parameters of carotids from ~2 , 500 healthy volunteers indicate higher intima media thickness ( P = 1 . 97×10−4 ) and wall to lumen ratio ( P = 0 . 002 ) in rs9349379-A carriers , suggesting indices of carotid hypertrophy previously described in carotids of FMD patients . Immunohistochemistry detected PHACTR1 in endothelium and smooth muscle cells of FMD and normal human carotids . The expression of PHACTR1 by genotypes in primary human fibroblasts showed higher expression in rs9349379-A carriers ( N = 86 , P = 0 . 003 ) . Phactr1 knockdown in zebrafish resulted in dilated vessels indicating subtle impaired vascular development . We report the first susceptibility locus for FMD and provide evidence for a complex genetic pattern of inheritance and indices of shared pathophysiology between FMD and other cardiovascular and neurovascular diseases .
Fibromuscular dysplasia ( FMD ) is a non-atherosclerotic and non-inflammatory vascular disease leading to stenosis , aneurysm , dissection , and/or occlusion of medium-sized arteries , in particular the renal and extracranial cerebrovascular arteries [1–3] . FMD predisposes to hypertension , transient ischemic attack and stroke [2 , 3] . Intriguingly , 75% to 90% of FMD patients are women[3 , 4] and FMD is increasingly considered to be a silent and under-diagnosed condition [5] . The angiography-based classification of renal FMD distinguishes between patients with multifocal stenoses , including the "string-of-beads” FMD pattern , and unifocal ( or focal ) FMD with corresponding differences in sex ratio , median age of diagnosis and smoking status [6 , 7] . The pathogenesis of FMD is unknown and there are strong arguments in favor of a genetic origin . We have previously reported familiality of FMD in ~10% of patients [8] . The US FMD registry has also described a family history in first or second-degree relatives for FMD ( 7% ) and aneurysm ( 23% ) [3] . Nonetheless , the unknown status of most of family members compromised heritability estimates and made the assessment of the genetic mode of inheritance difficult . Recently , we investigated the coding genomes from 16 affected siblings and excluded the existence of a major mutated gene in familial FMD [9] . The complexity of the diagnosis based on computed tomography angiography ( CTA ) and/or magnetic resonance angiography ( MRA ) and a general lack of awareness among affected patients and clinicians result in an under-diagnosis bias . To advance our understanding of the etiology of FMD , we performed a genetic association study that identifies a first genetic susceptibility locus for FMD . In addition to the high prevalence of asymptomatic FMD ( ~3–6% ) [3 , 10] and the existence of environmental modifiers ( e . g female hormones , lifetime mechanical stress ) our study provides genetic and functional evidence supporting for the first time a complex genetic basis for FMD .
We carried-out a multi-stage genetic association study , including a discovery and four validation case control cohorts , to identify genetic determinants of FMD . All FMD patients and controls were of European ancestry and have similar overall clinical features ( Table 1 ) . First , we analyzed 25 , 606 common ( minor allele frequency ≥ 0 . 05 ) genetic variants in 249 FMD cases and 689 controls . Despite the small sample size , we also performed stratified analyses including only females ( 193 patients and 416 controls ) . Associations across chromosomes are summarized in Manhattan Plots ( S2 Fig ) . In the global analysis , no SNP achieved the adjusted significance threshold for multiple testing ( P = 1 . 95×10−6 , S2 Fig ) . Nevertheless , the strongest association signal located on chromosome 6 ( rs9349379 , effect allele frequency ( EAF ) in cases = 0 . 70 , odds ratio ( OR ) = 1 . 65 , P = 1 . 47×10−5 ) surpasses the adjusted threshold in the female only analysis ( OR = 1 . 99 , P = 8 . 16 × 10−7 , S2 Fig , S1 Table ) . In addition to rs9349379 , we selected for follow-up loci that showed suggestive association with FMD ( P<10−3 ) and were located in or near either biological candidate genes ( e . g extracellular matrix degradation ) or previous cardiovascular genome-wide association ( GWAS ) signals . We also prioritized four SNPs located within one megabase interval around rs9349379 , the top associated variant . The first follow-up study included 402 FMD patients from the ARCADIA registry , who had similar clinical characteristics to the patients of the discovery stage and were compared to 2 , 537 controls from PPS3 ( Table 1 ) . Of the 16 SNPs selected , 13 passed genotyping QC criteria in cases and controls . Three SNPs , all located in the PHACTR1 locus , showed replicated association with FMD ( rs9349379-A , P = 7 . 21 × 10−4 , rs9369640-C , P = 8 . 45 × 10−4 and rs1332844-C , P = 1 . 72 × 10−3 , S1 Table ) . Conditioned regression analyses for rs9349379 in the case control cohort including the discovery and first follow-up study samples indicated association signal redundancy for rs9369640 ( P = 0 . 21 ) and rs1332844 ( P = 0 . 24 ) with rs9349379 being the most statistically significant . Next , we investigated the association of rs9349379 with FMD in three additional and independent case-control cohorts from the USA . Overall , of the four follow-up studies , three showed a significant effect of rs9349379 on the risk of FMD; all studies showed consistent direction of effect of the FMD risk allele rs9349379-A being more prevalent in FMD cases ( Table 2 ) . The association meta-analysis included 1 , 154 FMD patients and 3 , 895 controls and indicated an overall OR of 1 . 39 for rs9349379 ( EAF = 0 . 69 in global cases sample ) , with a high level of significance ( P = 7 . 36×10−10 ) that is below the genomic threshold and no indices for heterogeneity between studies ( P = 0 . 574 , Table 2 ) . As for rare variants ( MAF < 5% ) , we observed 62 , 767 polymorphic variants in the discovery case control analyses that we analyzed using the SKAT-O gene-based association test . None of the 9 , 967 genes covered with at least 2 polymorphic variants showed significant association with FMD after Bonferroni correction ( P = 5 . 02 × 10−6 , strongest gene P = 1 . 42 × 10−4 ) , or biological candidacy among best scoring genes , which discourages follow-up genotyping or sequencing efforts in larger samples . Here we aimed to characterize the association of rs9349379 with artery thickness and stiffness using a non-invasive method in 2 , 458 healthy volunteers . Genetic association analyses indicate significant association between the FMD increasing risk allele rs9349379-A and greater intima-media thickness ( IMT ) ( βadd = 11 . 65 μm , Padd = 1 . 65×10−4 , Fig 1 ) and wall to lumen ratio ( WLR ) ( βadd = 0 . 004 , Padd = 0 . 002 , Fig 1 ) , and decreased circumferential wall stress at DBP ( βadd = -0 . 72 , Padd = 0 . 004 , S2 Table ) . Interestingly , despite their smaller number ( N = 975 ) compared to males ( N = 1 , 483 ) , females present more accentuated effect of the rs9349379-A allele on IMT ( βadd = 14 . 83 μm , Padd = 0 . 001 ) and WLR ( βadd = 0 . 006 , Padd = 0 . 003 ) . However , we detected no significant interaction with sex . rs9349379-A also associated with increased wall cross-sectional area ( βadd = 0 . 24 mm2 , Padd = 8 . 67×10−4 , S2 Table ) , peripheral SBP ( βadd = 0 . 70 mmHg , Padd = 0 . 009 , S2 Table ) and central pulse pressure ( βadd = 0 . 62 mmHg , Padd = 0 . 002 , Fig 1 ) . Variant rs9349379 resides in the fourth intron of the phosphatase actin regulator 1 gene ( PHACTR1 ) in a putative regulatory sequence . We assessed PHACTR1 expression using quantitative real time PCR in mRNAs from primary cultured human fibroblasts . PHACTR1 expression did not differ in fibroblasts from FMD patients compared to age and sex matched controls ( Fig 2 ) . However , stratifying by rs9349379 genotype indicated increased expression in individuals carrying the FMD risk allele rs9349379-A ( Padd = 0 . 003 , Fig 2 ) . At the protein level , using a specific antibody , we found that PHACTR1 is expressed in endothelial and medial smooth muscle cells of carotid arteries from both normal and FMD patients ( Fig 2 ) . Given the unclear role for PHACTR1 in vascular development and maintenance , we assessed the effect of PHACTR1 perturbation in the zebrafish . Morpholino injected embryos had undetectable phactr1 transcript levels at 72 hours post fertilization while overall morphology was unchanged ( S3 Fig ) . Compared to control injected zebrafish , phactr1 knockdown resulted in a marked disorganization of the developing hepatic portal vein and the segmental vessels in the developing trunk ( Fig 3 and S3 Fig ) . Further analysis of the diameter of three major peripheral vessels , the dorsal aorta , caudal artery and posterior cardinal vein , demonstrated a nearly 8% dilatation ( P<0 . 05 ) of the posterior cardinal vein in phactr1 suppressed embryos .
Our study describes for the first time the genetic association of rs9349379 , a common variant in PHACTR1 , with arterial fibromuscular dysplasia ( FMD ) . We demonstrate that this common variant increases by ~40% the risk of FMD in five independent case-control studies . These findings are based on genetic data from 1 , 154 FMD patients and 3 , 895 controls , the largest investigation conducted so far to elucidate the genetic basis of this intriguing vascular disease with unknown etiological origin and challenging clinical features . To date , the genetic investigation on FMD was limited to the screening of candidate genes involved in rare vascular and arterial syndromes , and/or the study of underpowered series of patients . [9 , 11 , 12] For long considered a rare disease , FMD has also been hypothesized to be under the genetic control of highly penetrant genetic defects . Our study provides the first evidence for a complex genetic pattern of inheritance for FMD , involving a common genetic allele ( frequency = 0 . 60 in the general population ) . The identification of a genetic susceptibility locus for FMD supports the concept that this disease is controlled by a large number of genetic determinants in strong interaction with various environmental factors , including female sex and mechanical stresses . We provide genetic evidence for increased IMT , narrowed artery lumen and no change in arterial stiffness , even after adjusting for blood pressure , in healthy volunteers carrying the rs9349379-A , the FMD risk allele . These genetic associations support the observation of carotid concentric hypertrophy in FMD risk carriers , especially females . This finding is consistent with previous arterial features reported in FMD patients when compared to age , sex and SBP matched controls . [13] Circumferential wall stress was decreased in rs9349379-A carriers , showing that hypertrophy may overcome the moderate increase in blood pressure . Of note , the effect size of the FMD risk allele on IMT corresponds to approximately five years of aging for the IMT , when reported to sex and age reference values for arterial geometry , [14] suggesting potentially an accelerated arterial aging in carriers of the rs9349379-A FMD risk allele . PHACTR1 was previously identified by several genome-wide association studies ( GWAS ) as a risk locus for cardiovascular and neurovascular diseases . PHACTR1 is a confirmed susceptibility locus for coronary artery disease ( CAD ) and myocardial infarction ( MI ) , [15 , 16] migraine[17 , 18] and more recently cervical artery dissection ( CeAD ) , a rare condition defined as a mural hematoma in a carotid or vertebral artery and a cause of stroke . [19] Of note , the association of rs9349379 is in the opposite direction for FMD , CeAD and migraine , with rs9349379-A at risk , when compared to CAD and MI with rs9349379[G] allele at risk . Migraine and CeAD share several clinical features with FMD . Migraine is also more prevalent in females than in males and it is reported by a third of FMD patients . [3] CeAD is an important risk factor for subarachnoid hemorrhage stroke , as is cerebrovascular FMD . [3] The investigation of a large series of CeAD patients indicated that cervical FMD is reported in 5 . 6% of the patients[20] and carotid dissection is the presenting manifestation in 12 . 1% of patients of the US FMD registry . [3] In contrast to migraine and CeAD , the implication of the same common variant in PHACTR1 in CAD/ MI and FMD is rather unexpected , though it involved a different allele at the same genetic variant . The clinical link between FMD and CAD/MI is less obvious , except for the high proportion of FMD in rare forms of MI involving spontaneous coronary artery dissection , that also present in young females without atherosclerotic risk factors . [21–23] By definition FMD does not involve atherosclerotic stenosis that is concomitant with CAD and MI through pathogenesis processes ( e . g dyslipidemia and inflammation ) . Future comprehensive genetic investigation by full GWAS for FMD will allow the assessment of the putative consistency and opposition of effects between risk loci for these cardiovascular and neurovascular diseases . The rs9349379 is intronic to PHACTR1 in a putative noncoding regulatory sequence . PHACTR1 encodes a phosphatase and actin regulator protein and its function is not fully elucidated . Here we describe significant correlation between rs9349379 genotypes and PHACTR1 expression in human fibroblasts from FMD patients and controls . This association is consistent with the expression quantitative trait loci ( eQTL ) data from several artery beds of the Genotype-Tissue Expression ( GTEx ) project and a recent study where increased expression is reported in human coronary arteries from donors carrying the rs9349379-A . [16] Using a genome-editing technique with CRISPR-Cas9 applied to human umbilical vein endothelial cells , the deletion of the sequence containing rs9349379 caused a 35% decrease in PHACTR1 expression and impaired the fixation of the myocyte enhancer factor-2 ( MEF2 ) , supporting a functional regulatory effect of this genetic variant in vitro . [16] However , other transcription factors may also bind to this site , as the knockdown of the expression of two members of the MEF2 family did not change PHACTR1 expression . [24] We note that ENCODE ChIP-Seq data derived from genomes extracted from two immortalized cell lines after stimulation by estradiol indicates several consistent peaks for ESR1 suggesting a putative regulation of PHACTR1 by this female hormone . This putative regulation that needs to be established experimentally is consistent with the high proportion ( ~75–90% ) of women among FMD patients . Molecular studies have also linked PHACTR1 to cell adhesion and migration in angiogenesis via vascular endothelial growth factor ( VEGF ) stimulation . [25 , 26] In contrast to other studies , [26 , 27] Beaudoin et al reported a lack of evidence for the induction of PHACTR1 expression in endothelial cells with pro-angiogenic ( VEGF ) , pro-inflammatory stimulations or shear stress . [16] Our immunohistochemistry staining of PHACTR1 indicates the presence of PHACTR1 both in endothelial and smooth muscle cells of normal and FMD carotid arteries . The genetic implication of PHACTR1 in FMD , the subtle impairment in the development of vasculature in zebrafish , in addition to the evidence from recent GWAS describing an increasing number of loci with genes involved in vessel wall biology in CAD and MI[28] all support PHACTR1 plays a key etiological role in vascular structure . Recent work on PHACTR1 regulation in atherosclerosis showed strong expression in human atherosclerotic plaque macrophages lipid-laden foam cells , adventitial lymphocytes and endothelial cells . [29] This group also describes the absence of PHACTR1 in SMCs of healthy and athesclerotic aortas , which is not supported by our immunostaining on medium-sized arteries using a different anti-body , though they have detected the expression of an intermediate transcript in SMCs . Further investigation is required to understand this discrepancy that might reflect PHACTR1 is much more abundantly expressed in macrophages from atherosclerotic plaques , where this population of cells is highly represented and biologically active , which is different from the FMD and healthy vessels without atherosclerosis . Thus , according to rs9349379 genotype , the manifestation of one specific vascular disease depends on multiple and particular environmental triggers ( e . g . female specific hormonal context , mechanical movements of medium size arteries in the case of FMD ) that are still to be specifically determined for each of these cardiovascular and neurovascular diseases . In summary , our study provides genetic and functional evidence to support PHACTR1 as a first susceptibility locus for FMD . Further functional exploration of this locus and more comprehensive genetic investigation through genome-wide association will provide additional predisposing loci to FMD and help understanding the etiological mechanisms of non-atherosclerotic arterial stenosis .
We obtained individual written informed consent from all participants included in France and the USA case control studies . French FMD patients ( RVDRC and ARCADIA ) from the European Hospital Georges Pompidou ( HEGP ) are part of the ARCADIA/PROFILE protocol that was approved by the Ile-De-France research ethics committee ( Comité de Protection des Personnes: CPP d’île de France ) on 03/04/2009 ( ID: 2009-A00288-49 ) . The ethics committee of the Paris-Cochin hospital approved the protocols of the SU . VI . MAX study ( CCPPRB No . 706 ) and PPS3 ( CPP No . 2007-A01386-47 ) . PPS3 is declared at ClinicalTrials . gov ( Identifier: NCT00741728 ) . Mayo Clinic case control study is part of the Vascular Diseases Biorepository study and was approved by the Mayo I Mayo Clinic Institutional Review Boards ( IRB # 08–008355 ) . UM/Cleveland case control study was approved by each institution IRB protocols: the University of Michigan IRB number HUM00044507 and the Cleveland Clinic IRB number 10–318 . The DEFINE-FMD study was approved by the Human Research Ethics Committee of the Icahn School of Medicine at Mount Sinai ( Study ID: HS#13-00575/GCO#13–1118 and is registered with ClinicalTrials . gov Identifier: NCT01967511 . We used a three-stage association design . First , we performed an exome-chip based genetic association in 249 French FMD patients ( RVDRC cohort ) and 689 controls from SU . VI . MAX [30] . Second , we followed-up 13 loci in an independent set of 402 French patients ( ARCADIA registry ) and 2 , 537 controls from PPS3 [31] . Three additional studies from the USA totaling 512 patients and 669 controls were used for further replication: Mayo Clinic cohort [32] , University of Michigan ( UM ) /Cleveland Clinic cohort and the DEFINE-FMD study . All participants are of European ancestry and presented similar clinical characteristics ( Table 1 ) and homogeneous diagnosis , exclusion and inclusion criteria . Genotypes were generated using the Illumina-HumanExome-12v1 array in RVDRC , SU . VI . MAX and PPS3 , [34] Illumina-Human-Omni-Express-Exome in the DEFINE-FMD Study participants , Illumina-Infinium-Human-CoreExome in Mayo Clinic cohorts and by individual genotyping in ARCADIA ( KASP technology ) and UM/Cleveland Clinic cohorts ( Taqman ) . We applied quality control ( QC ) filters to the discovery cases and controls as recommended[35] ( individual’s call rate < 97% , extreme heterozygosity , sex-discordant , duplicate or relatedness unsing PLINK ( version 1 . 07 ) . [36] Individuals with non-European ancestry were detected and excluded ( 54 cases and 9 controls ) using the EIGENSTRAT program[37] and visualized by principal components analysis ( PCA ) including HapMap phase 3 samples . Cases and controls displayed a comparable distribution after ancestry QC ( S1 Fig ) . From 240 , 748 successfully genotyped SNPs , we excluded monomorphic variants ( n = 129 , 890 ) , call rate < 99% , deviation from Hardy-Weinberg Equilibrium in cases and/or controls ( P < 10−5 ) and minor allele frequency ( MAF ) <0 . 05 in controls . The final analysis included 25 , 606 variants ( 25 , 138 autosomal and 468 X-linked ) in a sample of 249 FMD cases and 689 controls . Comparable individuals and markers QC was applied to follow-up cohorts using arrays ( PPS3 , Mayo Clinic case control cohorts , the DEFINE-FMD study ) or individual genotyping ( ARCADIA and UM/Cleveland cases control cohort ) . Carotid parameters were measured and calculated as previously reported [34] . Briefly , a 10 MHz 128 transducer linear array probe was positioned on the carotid area . Measurements were performed on a 4 cm segment of the right common carotid artery , 1 cm proximal to the bifurcation/sinus throughout the cardiac cycle for 6 seconds . A longitudinal section showing clear interfaces for blood/intima and media/adventitia was obtained . The system allows real-time radiofrequency signal analysis with operator-independent determination of external diameter ( Dext ) , internal diameter ( Dint ) , and intima-media thickness ( IMT ) on 128 lines throughout the cardiac cycle . Distension was measured on 14 lines at high-pulsed radiofrequency ( 600 Hz ) . The axial resolution was 34 μm for diameter , 17 μm for IMT , and 1 . 7 μm for distension . [38] Aortic blood pressure was estimated from the distension waveform according to van Bortel et al . [39] The distensibility coefficient , representing the elastic properties of the artery as a hollow structure , was calculated as dLCSA/ ( LCSA×central PP ) , where LCSA is the lumen cross-sectional area and central PP is the aortic pulse pressure . Carotid stiffness ( Cstif ) was calculated as DC−0 . 5 and circumferential wall stress as diastolic blood pressure×Dint/2×IMT . WCSA is wall cross-sectional area . Young's elastic modulus is calculated as DC−1x3 ( 1+LCSA/WCSA ) . Here we examined the association of one genetic variant ( rs9349379 ) with 14 interdependent traits in a sub-sample of 2458 participants ( 975 females and 1483 males ) . Hypertensive participants ( BP over 140 and/or 90 mmHg , and/or use of antihypertensive treatments ) were excluded to avoid confounding with genetic effects . We tested the association with FMD using logistic regression under the additive genetic model as implemented in PLINK[36] ( version 1 . 07 ) in discovery and follow-up studies . In the discovery analysis , we included the first five principal components axes as co-variates to control for hidden population stratification . The Bonferroni adjusted threshold for significance was set to P = 1 . 95×10−6 to account for multiple testing of 25 , 606 common variants . We used the inverse variance-weighted method for meta-analysis implemented in Metal [40] . Gene-based analyses of rare variants was performed using SKAT-O method as described in . [41] Heterogeneity was assessed with Cochran's Q statistics . We tested the association of rs9349379 with carotid parameters in 2 , 458 normotensive subjects ( 975 females and 1483 males ) from PPS3 using a linear regression on an additive genetic model including age , sex , body surface area ( BSA ) , and mean blood pressure ( MBP ) as covariates when relevant . Before performing the analyses , all parameters ( all quantitative traits ) were quantile-transformed to a standard normal distribution . The 14 traits fall into three main carotid parameters categories: geometry ( intima-media-thickness , external and internal carotid diameters , wall to lumen ratio and circumferential wall stress ) , arterial stiffness ( stiffness , cross-sectional distensibility , Young’s elastic modulus , cross-sectional compliance , and wall cross-sectional area ) and central blood pressure ( systolic , diastolic and aortic blood pressure ) . Considering that carotid parameters are interdependent , we applied a Bonferroni correction ( Padj = 0 . 017 ) for the three main categories as recommended [34] . We analyzed the expression of PHACTR1 from fibroblasts of skin biopsy samples of 104 individuals ( 51 FMD , 39 controls , 12 undetermined ) . Genotypes and detectable expression levels were available in 86 individuals . Fibroblasts of the DEFINE-FMD Study were derived from skin biopsy samples using standard explant techniques from 103 individuals ( 46 FMD and 57 controls ) . Briefly , skin biopsy samples were dissected into small pieces and cultured under cover slips in DMEM/F12 media containing 20% fetal bovine serum , 1% antibiotic-antimycotic solution , 1% 200 mM L-glutamine , 1% 100mM Sodium pyruvate , and 1% MEM Non-Essential Amino Acids ( all from Life Technologies , Grand Island , NY , USA ) at 37°C in 5% CO2 . Cell culture medium was replaced every 48 hours and confluent cells from passages 2–3 were used for RNA extractions . RNA was then converted to cDNA using iscript cDNA synthesis kit ( Bio Rad , Hercules , CA , USA ) and quantitative real time PCR ( qRT-PCR ) was performed using SYBR green fastmix low rox ( Quanta Biosciences , Gaithersburg , MD , USA ) . Fibroblasts from the University of Michigan were obtained through the Coriell Biorespository ( Camden , NJ , USA ) for 20 subjects with FMD and apparently 20 healthy control subjects and matched for passage number . DNAs and genotypes were not available . Confluent cells from passages 6–8 were used to extract RNA and generate cDNAs to assess PHACTR1 expression according to culture and extraction protocol described for the DEFINE-FMD study . Fibroblasts from the Department of genetic , HGEP study were derived from skin biopsy samples from patients followed up for unknown arterial diseases where FMD and Elhers Danlos syndrome were discarded . FMD status is not available . SNP rs9349379 was genotyped by direct forward and reverse sequencing ( BigDye Terminator kit v3 . 1 cycle sequencing kit ) and run on an ABI Prism 3730XL DNA Analyzer Sequencer ( Perkin Elmer Applied Biosystems , Foster City , CA ) . Cell culture and RNA extraction conditions and reagent were identical to the DEFINE-FMD study . Informed consent was obtained from all patients of all centers . In total , we analyzed the expression of PHACTR1 from fibroblasts of skin biopsy samples of 104 individuals ( 51 FMD , 39 controls , 12 undetermined ) , and 86 individuals had genotypes and relevant expression levels . The expression of PHACTR1 transcript variant 1 ( NCBI Reference Sequence: NM_030948 . 2 ) and GAPDH , which was used as normalization control , was determined by RT-qPCR following the MIQE guidelines . [42] All expression studies from the three laboratories used the same following primers: PHACTR1 fwd ATGACCGCAGGGCAGATAAG , rev TTCGGATGGCAGCTTTGTCT; GAPDH fwd GGGTGTGAACCATGAGAAGTATGA , rev GGTGCAGGAGGCATTGCT . The efficacy and efficiency of amplifications for PHACTR1 and GAPDH were equivalent as determined by the linearity tests: ( PHACTR1 efficacy 1 . 97 and efficiency 98 . 7% ) ; ( GAPDH efficacy 2 . 04 and efficiency 102% ) . The specificity of amplification products was verified by the presence of a single peak in the melting temperature curve analysis . Outliers displaying GAPDH Cq values different from the mean or a ΔCq value >14 ( Cq value PHACTR1 –Cq value GAPDH ) were filtered out . Good correlations for RT and technical qPCR replicates were obtained . Given that data obtained from three independent laboratories was equivalent: mean Cq for GAPDH 17 , 41 +/- 0 , 65 and mean Cq for PHACTR1 was 30 . 46 +/- 1 . 19 , we pooled them as indicated to gain power , especially for the correlation by genotypes where few samples per study were GG carriers . Data is presented as the mean fold change in the expression of PHACTR1 relative to GAPDH , calculated using the formula 2-ΔCq , as recommended . [43] Statistical analyses were performed using a non-parametric Mann-Whitney test for single comparison between FMD cases and controls and Kruskal Wallis test for comparisons according to rs9349379 genotypes . Protein staining for PHACTR1 was performed on paraffin embedded artery samples using primary anti-PHACTR1 antibody ( Sigma-Aldrich , St Louis , MI , USA ) and revealed using an ABC peroxidase kit with diaminobenzidine ( Vector laboratories , Burlingame , CA , USA ) . The paraffin blocks of arterial tissues were obtained from surgical pathology archives of the HEGP as remnants of the regular diagnostic procedure ( 2 renal arteries , 1 internal carotid , 1 popliteal and 1 femoral artery ) . FMD arteries belonged to patients who had surgical reparation for aneurysm resection ( 1 carotid , 1 radial , 1 coeliac , 1 ulnar and 1 pancreaticoduodenal artery ) . The arterial tissues were fixed in formalin and embedded in paraffin . Antigen retrieval were performed by incubating tissue sections in alkaline solution ( Dako , Trappes , France ) for 40 minutes at 94°C in a hot water bath . The sections were then incubated for 60 minutes with the primary anti-PHACTR1 rabbit polyclonal antibody diluted at 1/50 . For revelation we used ABC peroxidase kit with diaminobenzidine ( Vector laboratories , Burlingame , CA , USA ) . Anti-phactr1 morpholino was designed to target the exon7/intron7 boundary of zebrafish phactr1 transcript . A non-targeting morpholino of equivalent length but differing nucleotide composition was injected at equivalent concentrations as a control . For injections , 0 . 8nL of control of anti-phactr1 morpholino was injected into single cell embryos of the Tubingen/AB strain . For confirmation of morpholino efficacy , primers spanning the targeted boundary were used to amplify cDNA constructed from the isolated RNA of injected 72 hour post fertilization embryos . All the morphological analyses as well as the knockout status of the fish were performed blinded to the experimenter . For imaging studies , fish were reared under standard conditions with addition of phenylthiourea to media in order to inhibit pigmentation . At 72hpf , high-speed videography was performed as described[44] used to determine the diameter of major vessels . Measurement of vessels was manually performed using ImageJ . For confocal reconstructions of the embryonic vasculature , embryos obtained from lines expressing EGFP driven by the flk promoter were used . Confocal images were obtained on a Nikon A1SiR confocal . Two-dimensional projections of maximal intensity ( head and torso ) or averaged intensity ( trunk ) were used to visualize the overall vascular architecture using ImageJ .
|
Fibromuscular Dysplasia ( FMD ) is a vascular disease characterized by a succession of occlusions and dilatation of medium-sized arteries ( e . g renal , carotid or coronary arteries ) with important health consequences , mainly resistant hypertension and stroke . FMD is an atypical vascular disease because it is not associated with overweight or dyslipidemia and 80% of patients are early middle aged women . Our genetic study conducted in >1100 patients and >3800 controls demonstrate that a common variant rs9349379 located on chromosome 6 in the phosphatase and actin regulator 1 gene ( PHACTR1 ) increases by ~40% the risk of FMD . This is the first time a genetic risk factor is reported for FMD because it has been longtime considered rare and potentially under a Mendelian mode of inheritance . We also show that rs9349379 correlates with the expression of PHACTR1 in fibroblasts from FMD patients and controls . Interestingly , the same allele that increases the risk of FMD is at risk for cervical artery dissection and migraine , often reported in FMD patients but protective from myocardial infarction and coronary disease , where atherosclerosis is more common . The clear role of PHACTR1 in maintaining vascular well integrity is not fully elucidated . Using a specific antibody we detected PHACTR1 both on endothelial and smooth muscle cells of human FMD and control carotids , which suggests that PHACTR1 may have multiple functions depending on the cell type and the degree of atherosclerosis of the arteries .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"cardiovascular",
"anatomy",
"renal",
"arteries",
"fibroblasts",
"genetic",
"predisposition",
"arteries",
"connective",
"tissue",
"cells",
"blood",
"vessels",
"animal",
"cells",
"connective",
"tissue",
"biological",
"tissue",
"genetic",
"loci",
"anatomy",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"human",
"genetics",
"vascular",
"medicine",
"genetics",
"of",
"disease",
"vascular",
"diseases"
] |
2016
|
PHACTR1 Is a Genetic Susceptibility Locus for Fibromuscular Dysplasia Supporting Its Complex Genetic Pattern of Inheritance
|
The proper biological functioning of proteins often relies on the occurrence of coordinated fluctuations around their native structure , or on their ability to perform wider and sometimes highly elaborated motions . Hence , there is considerable interest in the definition of accurate coarse-grained descriptions of protein dynamics , as an alternative to more computationally expensive approaches . In particular , the elastic network model , in which residue motions are subjected to pairwise harmonic potentials , is known to capture essential aspects of conformational dynamics in proteins , but has so far remained mostly phenomenological , and unable to account for the chemical specificities of amino acids . We propose , for the first time , a method to derive residue- and distance-specific effective harmonic potentials from the statistical analysis of an extensive dataset of NMR conformational ensembles . These potentials constitute dynamical counterparts to the mean-force statistical potentials commonly used for static analyses of protein structures . In the context of the elastic network model , they yield a strongly improved description of the cooperative aspects of residue motions , and give the opportunity to systematically explore the influence of sequence details on protein dynamics .
Deciphering the motions that underlie many aspects of protein function is a major current challenge in molecular biology , with the potential to generate numerous applications in biomedical research and biotechnology . Although molecular dynamics ( MD ) hold a prominent position among computational approaches , considerable efforts have been devoted to the development of coarse-grained models of protein dynamics [1] . Besides their ability to follow motions on time scales that are usually not accessible to MD simulations , these models also give the possibility to better understand the general principles that rule the dynamical properties of proteins . The elegant simplicity of the elastic network models ( ENM ) certainly contributed to their popularity , and they have been successfully exploited in a wide range of applications [2]–[5] . In these models , the residues are usually represented as single particles and connected to their neighbors by Hookean springs [6] , [7] . The input structure is assumed to be the equilibrium state , i . e . the global energy minimum of the system . Common variants include the homogeneous ENM , in which springs of equal stiffness connect pairs of residues separated by a distance smaller than a predefined cutoff , and other versions in which the spring stiffness decays as the interresidue distance increases [8]–[10] . In all cases , the equations of motion can be either linearized around equilibrium , to perform a normal mode analysis of the system [11]–[13] , or integrated to obtain time-resolved relaxation trajectories [14] , [15] . Despite their many achievements , purely structural ENM also come with severe limitations . Notably , modeling the possible effects of mutations within this framework usually requires random local perturbations of the spring constants [16] , or a more drastic removal of links from the network [17] . A few attempts have been made to include sequence-specificity in the ENM by setting the spring constants proportional to the depth of the energy minima , as estimated by statistical contact potentials [18] , [19] . However , this approach cannot be extended to distance-dependent potentials , for they are not consistent with the ground hypothesis of the ENM , i . e . that all pairwise interaction potentials are at their minimum in the native structure . Other studies have led to the conclusion that the ENM behave as entropic models dominated by structural features , and that the level of coarse-graining is probably too high to incorporate sequence details [7] , [20] . Still , the chemical nature of residues at key positions can have significant effects on the main dynamical properties of a protein . Hinge motions [21] , for instance , obviously require some architectural conditions to be fulfilled , such as the presence of two domains capable of moving relatively independently . But the amplitude and preferred direction of the motion are most likely determined by fine tuning of specific interactions in the hinge region . In proteins subject to domain swapping , the hinge loops have indeed been shown to frequently include residues that are not optimal for stability [22] . The importance of the amino acid sequence has also been repeatedly emphasized by experimental studies of the impact of mutations on the conformational dynamics of proteins [23]–[25] . A major obstacle to the definition of accurate coarse-grained descriptions of protein dynamics lies in the highly cooperative nature of protein motions , which makes it difficult to identify the properties of the individual building blocks independently of the overall architecture of each fold . By condensing the information contained in a multitude of NMR ensembles , we build here a mean protein environment , in which the behavior of residue pairs can be tracked independently of each protein's specific structure . This methodology brings an efficient way of assessing coarse-grained models of protein dynamics and of deriving effective energy functions adapted to these models . In the context of the ENM , we identify a set of spring constants that depend on both the interresidue distances and the chemical nature of amino acids , and that markedly improve the performances of the model .
The mean-square fluctuations of individual residues ( MSRF ) have been extensively relied on to characterize protein flexibility and to evaluate coarse-grained models of protein dynamics [26] , in part because of their widespread availability as crystallographic B-factors . However , since the MSRF carry little information about the cooperative and anisotropic nature of residue motions , we propose to examine the dynamical behavior of proteins from the perspective of residue pairs rather than individual residues . Information about the fluctuations of interresidue distances is contained in the data of NMR experiments for numerous proteins , and will be exploited here . We define the apparent stiffness of a pair of residues in a protein : ( 1 ) where is the Boltzmann constant , the temperature , and the variance of the distance between residues and , in a structural ensemble representative of the equilibrium state . is defined up to a multiplicative factor , which corresponds to the temperature . We also introduce the uncorrelated apparent stiffness , to quantify the impact of the individual fluctuations of residues and on the fluctuations of the distance that separates them . This is achieved by using instead of in eq . 1 , where is computed after exclusion of all correlations between the motions of residues and ( see Methods ) . As illustrated in Figure 1 , can be quite different from one residue pair to another . Indeed , besides the impact of direct interactions , is also strongly dependent on the overall fold of the protein , and on the position of the pair within the structure . To remove the specific influence of each protein's architecture , we define the apparent stiffness in a mean protein environment : ( 2 ) where is one of 210 amino acid pairs , the discretized equilibrium distance between pairs of residues ( Å ) , the number of structures in the equilibrium ensemble of protein , and the number of residue pairs in protein . Pairs of consecutive residues were dismissed , so as to consider only non-bonded interactions . The mean protein environment is thus obtained by averaging over a large number of residue pairs in a dataset of different proteins ( see Methods ) . The influence of the distance separating two residues on the cooperativity of their motions can be investigated by considering amino acid types indistinctively in eq . 2 . Interestingly , follows approximately a power law , with an exponent of about −2 . 5 ( Figure 2 ) . Finer details include a first maximal value occurring for – distances between 5 and 5 . 5 Å , i . e . the separation between hydrogen-bonded residues within regular secondary structure elements , and a second around 9 Å , which corresponds to indirect , second neighbor , interactions . The high level of cooperativity in residue motions is well illustrated by the comparison of and its uncorrelated counterpart . Indeed , these two functions would take identical values if the variability of the distance between two residues could be explained solely by the extent of their individual fluctuations . In a mean protein environment , however , is about two orders of magnitude larger than at short-range , and the difference remains quite important up to about 30–40 Å . The comparison of values extracted from subsets containing exclusively small , large , all- , or all- proteins indicates that the content of the dataset has a remarkably limited impact on ( Figure S1 ) . This distance dependence can thus be seen as a general property of protein structures , a signature of protein cooperativity at the residue pair level . Of course , since is representative of a mean protein environment , deviations may occur for individual proteins , according to their specific structural organizations ( Figure S2 ) . The apparent stiffness is computed for each type of amino acid pair using eq . 2 , by considering only residue pairs separated by less than 10 Å . As shown in Figure 3A , the chemical nature of the interacting residues is a major determinant of their dynamical behavior . Unsurprisingly , Glycine and Proline appear as the most effective ingredients of flexibility . Pairs involving hydrophobic and aromatic amino acids tend to be considerably more rigid , with values up to 6 times larger . These differences originate in part in the individual propensities of different amino acids to be located in more or less flexible regions ( e . g . hydrophobic core vs . exposed surface loops ) . However , there is only a limited agreement between and ( Figure 3A–B ) : the correlation coefficient is equal to 0 . 71 , and spans a much wider range of values . Beyond individual amino acid preferences , the specifics of residue-residue interactions play thus a significant role in determining the extent of cooperativity in residue motions . The computation of the apparent stiffness of residue pairs in a mean protein environment provides an interesting tool to probe the dynamical properties of proteins . It also generates a very straightforward approach to assess the ability of coarse-grained models to reproduce accurately this general behavior . We focus here on four common variants of the residue-based ENM [27] , [28] , which differ only by the functional form of the spring constants . The dependence of on the interresidue distance is defined by two parameters: the cutoff distance , above which residues and are considered disconnected , and the exponent that determines how fast decreases with increasing distances: ( 3 ) where is the Heaviside function . The value of the temperature-related factor is obtained , for each protein independently , by fitting the predicted MSRF with the experimental ones . This ensures that the amplitude of the individual fluctuations of the beads in the network is on average equal to that observed in the corresponding NMR ensemble , and that the predicted values can thus be directly compared with those extracted from the NMR data . We consider the following models: , , , . These ENM variants were used to estimate the value of for each pair of residues in the 1500 proteins of our NMR dataset ( see Methods ) , and to subsequently compute and from eq . 2 . Strikingly , all ENM variants systematically predict values to be lower than the experimental ones , at least up to interresidue distances of 20–30 Å ( Figure 2 ) . These models overestimate thus the amplitude of pairwise fluctuations , relatively to the amplitude of individual fluctuations . For example , if two residues in a protein undergo highly correlated motions , the amount of thermal energy necessary to induce a moderate variance on the distance between them will generate high variances on their individual coordinates . Consequently , if the motions of the beads of the ENM are less coordinated , adjusting the scale of the spring constants to reproduce the amplitude of individual fluctuations leads to an overestimated variance on the interresidue distances , and thus to lower values . This problem is particularly apparent when is assumed to decrease proportionally to the square of the interresidue distance , in the . Although this model was shown to perform well in predicting MSRF values [10] , our results suggest that it negates almost completely the coordinated aspect of residue motions . Indeed , as shown in Figure 2 , the values predicted by this model are very close to those obtained from the experimental data after removal of the correlations between the motions of the different residues ( ) . This observation is consistent with the extremely short atom-atom correlation length characteristic of the , recently estimated on the basis of an X-ray structure of Staphylococcal nuclease [27] . The ENM is often considered as an entropic model , not detailed enough to include sequence information in a relevant way [7] , [20] . It is therefore hardly surprising that common ENM variants produce a poor description of the sequence specificities of protein dynamics . Individual amino acid preferences for more or less densely connected regions are responsible for some variety in the predicted values of ( Figure 3C–D ) . However , this variety is far from matching the one observed in the experimental data , as shown by a much narrower range of values , and a limited correlation coefficient with the experimental values , e . g . 0 . 64 for the and 0 . 62 for the ( Figure S3 ) . There is a much better agreement between the values predicted by the , and the experimental values of the uncorrelated apparent stiffness ( Figure 3B , D , correlation coefficient of 0 . 84 ) , which confirms that this model ignores the coordinated aspects of residue motions . Mean-force statistical potentials are commonly used to perform energetic evaluations of static protein structures [29]–[31] . These potentials do not describe explicitly the “true” physical interactions , but provide effective energies of interaction in a mean protein environment , in the context of a more or less simplified structural representation . Similarly , within the ENM framework , defines for each pair of residues an harmonic interaction potential . This potential is also effective in nature , accounting implicitly for everything that is not included in the model ( e . g . the surrounding water ) . Hence , we seek to identify the value of yielding the most accurate reproduction of the dynamical behavior of each type of pair in a mean protein environment , which is conveniently captured by the apparent stiffness . For that purpose , let us define as the energy of the elastic spring connecting two residues of type , in a mean protein environment: ( 4 ) where is the apparent stiffness extracted from the experimental data . is unknown and is expected to be different for different pair types . The knowledge of is thus not sufficient to estimate directly . However , from any approximate set of spring constants , we may build the ENM for all proteins in our dataset , to reproduce the mean protein environment , and compute for each pair type an estimated value of the apparent stiffness , , and bond energy , . Since the behavior of a given residue pair is highly dependent on its environment , we can make the assumption that is a relatively good approximation of , even if : ( 5 ) Indeed , if the spring stiffness of a residue pair is underestimated , it will also appear as less rigid in the ENM than in the experimental data . A more detailed discussion is given in Supporting Text S1 . From eqs . 4 and 5 , we devise thus an iterative procedure in which is updated at each step by confronting the predicted values of the apparent stiffness , , with the experimental ones , . It is expected to converge when , that is , when the predictions of the model agree with the experimental data: ( 6 ) We used this approach to derive , from the NMR data , four novel ENM variants: the distance-dependent dENM ; the sequence-dependent and , with a distance cutoff of 10 and 13 Å , respectively , and the sequence- and distance-dependent sdENM ( see Methods ) . Interestingly , the values for the 210 amino acid pairs in the are relatively well correlated with the corresponding contact potentials [30] , even though they result from totally different approaches ( Figure S4 ) . Some common general trends can be identified , e . g . hydrophobic contacts tend to be associated with both favorable interaction energies and large values ( Figure 4A ) . However , the overall correspondence remains limited , indicating that the determinants of protein rigidity and stability are related , but distinct . The distance dependence of in the dENM is remarkably similar to the power law that was previously obtained by fitting against a 1 . 5 ns MD trajectory of a C-phycocyanin dimer [8] ( Figure 4B ) , although our new model presents more detailed features . Notably , remains approximately constant up to interresidue distances of 5–6 Å , and then drops by about two orders of magnitude to reach a second plateau between 7 and 12 Å . The bootstrap estimates of the 90% confidence intervals displayed on Figure 4B underline the robustness of our derivation scheme , and indicate that the values determined here depend only marginally on the content of the dataset . The values of the sdENM are shown in Figure 4C , for a few amino acid pairs . This model not only combines the strengths of the sENM and the dENM , but also reveals the sequence specificity of the distance dependence . The D-R pair , for example , is almost as rigid as I-I at short distances consistent with the formation of a salt bridge , but almost as flexible as G-G at larger distances . There is of course a larger uncertainty on the determination of values , which is reflected by wider confidence intervals than those on in the dENM ( Figure 4B , C ) . This is due to the limited amount of available experimental data , and to the fact that the modelled dynamical behavior of a protein is obviously less sensitive to variations of the spring constant values for one type of amino acid pair , than for all amino acid pairs indistinctively . However , this uncertainty remains small enough to allow the identification of significant differences between the values determined for different types of amino acid pairs . In the example of Figure 4C , is consistently larger than over the whole range of inter-residue distances , whereas is significantly larger than at short-range ( 4–6 Å ) , and significantly smaller than at mid-range ( 6–12 Å ) . The sdENM yields a much more accurate reproduction of the dynamical behavior of residue pairs in a mean protein environment than the common ENM variants , as demonstrated by the good agreement between experimental and predicted values of ( Figures 5A , S5 ) , and ( Figure 5B ) . Beyond its performances in a mean protein environment , our new model also brings highly notable improvements with respect to previously described ENM variants when it is applied to the specific architecture of a given protein . This is illustrated by two examples , on Figure 6 . A more thorough assessment of the ability of the different ENM variants to capture the motions of individual proteins was performed on an independent dataset of 349 proteins . The correlation coefficient between predicted and observed MSRF ( ) has been widely used in the past but ignores the cooperativity inherent to protein dynamics , and presents other shortcomings . Therefore , we introduce a new measure ( ) that quantifies the relative error on the estimation of the variability of the distance between residue pairs , and is thus focused on the cooperative aspects of residue motions ( see Methods ) . Among the 4 previously described ENM variants , the is better at predicting the individual residue fluctuations ( Table 1 ) . Interestingly , the , with its simple cutoff distance , appears superior when it comes to the reproduction of cooperative motions ( ) . The new ENM variants based on our effective harmonic potentials present enhanced performances in comparison with the common models . In particular , the dENM reaches the same level of quality as the for individual fluctuations ( ) , but surpasses even the for the description of cooperativity ( ) . On the other hand , the impact of introducing sequence specificity can be examined by comparing with , and sdENM with dENM . It consists in a slight improvement of the correlation coefficient , and a pronounced decrease of the error , especially at short- ( 0–15 Å ) and mid- ( 15–30 Å ) range .
For the last decades , statistical potentials extracted from datasets of known protein structures [29]–[31] have played a critical role in static analyses of protein structures , with major applications including structure prediction , protein-protein docking , or rational mutant design . Our study demonstrates that a similar approach can be taken to derive effective energy functions that are specifically adapted to the coarse-grained modeling of protein dynamics . More precisely , in the context of the ENM , we exploited a dataset of 1500 NMR ensembles to determine the values of the spring constants that describe best the behavior of pairs of residues , as a function of both their chemical nature and the distance separating them . The success of our approach is attested by a drastic enhancement of the ability to accurately reproduce the cooperative nature of residue motions , with respect to previously described ENM variants . Moreover , a definite advantage of our method is that the effective parameters characterizing the strength of the virtual bonds are directly extracted from the experimental data without any a priori conception of their functional form . The fact that the distance dependence of the spring constants that we retrieve is quite similar to the power law , which was considered so far as underlying one of the best performing ENM variants [8] , [27] , also constitutes a major support to our approach . In our derivation scheme , the virtual bonds are parametrized so as to reproduce the behavior of amino acid pairs in a mean protein environment . The analysis of the ability of different models of protein dynamics to describe the motions of residues within this environment sheds an interesting new light on the properties of these models . In particular , our results indicate that previous ENM variants underestimate , sometimes dramatically , the rigidity of amino acid pairs at short- and mid-range . Our new model does however provide a much more accurate reproduction of the balance between short-range and long-range coordinated motions . This is arguably a crucial aspect when considering , for example , the consequences of localized alterations induced by ligand binding on signal transduction or global conformational changes , such as in ATP-powered molecular motors . Importantly , our results also demonstrate that the ENM does not have to be exclusively structural , and that sequence details may be allowed to play a major role in coarse-grained descriptions of protein dynamics . Thereby , this study paves the way towards comparative analyses of motions in proteins that share a similar structure but present differences in sequence . Such investigations will prove particularly interesting in the context of the rational design of ( modified ) proteins with controlled dynamical properties . On the other hand , the importance of orientational effects in protein dynamics has been underlined by both experimental and computational studies [5] , [7] , [32]–[36] . At the protein level , these effects are nicely illustrated by the strong anisotropy of a protein's response to applied external forces [33] , [34] , [36] . At the residue level , the anisotropy can be related to the directional variability of the packing density experienced by any given residue [5] , [35] . The accurate description of such orientational effects should benefit from the availability of sequence-specific models . Indeed , beyond the number of contacts established in each direction , the actual nature of these contacts can also have a substantial influence on the anisotropy of residue fluctuations . Although we focused here on residue-based elastic network models , our approach is not limited to this particular family , and can be readily implemented to use available dynamical data for the evaluation and optimization of other coarse-grained models of protein dynamics . Notably , the impact of chemical specificity on the dynamical behavior of residues should be even more accurately rendered by effective potentials based on a more detailed structural description .
We retrieved , from the Protein Data Bank [37] , ensembles of at least 20 models from solution NMR experiments , corresponding to monomeric proteins of at least 50 residues that present at most 30% sequence identity with one another . Entries under the SCOP classifications “Peptides” or “Membrane and cell surface proteins” were not considered . The presence of ligands , DNA or RNA molecules , chain breaks , non-natural amino acids , and differences in the number of residues per model were also grounds for rejection . These criteria led to the selection of 1849 distinct structural ensembles . A subset of 1500 ensembles was randomly selected for the main analysis , and the remaining 349 were used to assess the performances of the different ENM variants . Unfolded C- or N-terminal tails were automatically identified ( MSRF values larger than twice the average for all residues in the protein ) and removed from consideration . In each ensemble , the structure with the lowest root mean square deviation from the mean structure , after superposition , is chosen as representative and used to build the ENM . The network is built by considering each residue as a single bead , placed at the position of the corresponding atom in the input structure , and connecting neighboring beads with Hookean springs [6] , . The ENM variants considered here differ only by the form of the spring constant as a function of interresidue distance and of amino acid types . In all variants , bonded interactions are described by a larger value of , defined as ten times the value of for non-bonded interactions at a separation of 3 . 5 Å , averaged over all amino acid types . The potential energy of the network is given by: , where and are the instantaneous and equilibrium distances between residues and , respectively . By definition , the input structure corresponds to the global energy minimum , with . For a protein of residues , the Hessian of the system is the matrix of the second derivatives of with respect to the spatial coordinates of the residues . The eigenvalue decomposition of yields the covariance matrix of the spatial coordinates , which constitutes the output of the model: ( 7 ) where the sum is performed over the non-zero eigenvalues of , and are the corresponding eigenvectors . is a symmetrical matrix , constituted of submatrices : ( 8 ) where , , and correspond to the displacements of residue from its equilibrium position , along the three Cartesian coordinates . The predicted MSRF of residue is given by the trace of submatrix . For each pair of residues in a given protein , the experimental value of this variance is readily computed from the NMR data: ( 9 ) where is the number of structures in the NMR ensemble , the distance between the atoms of residues and in structure of protein , and the average distance over all structures . In the context of the ENM , values are estimated from the covariance matrix of the spatial coordinates , by standard statistical propagation of uncertainty: ( 10 ) where is the Jacobian of the distance as a function of the six spatial coordinates: ( 11 ) This estimation of takes into account the individual , anisotropic , fluctuations of both residues , as well as the correlations between their respective motions . It relies on the validity of the first order Taylor expansion of the distance as function of the coordinates in the vicinity of the average distance . We ensured that no systematic bias arose from this approximation ( Figure S6 ) . To quantify the impact of the individual motions of residues on their relative positions , we use eq . 10 to compute in an artificial construct where residue motions are not correlated . This is achieved by extracting the covariance matrix from the NMR data , and setting to zero all submatrices where . The values of the spring constants of the new ENM variants were derived from the dataset of 1500 NMR ensembles using eq 6 . For the dENM , and , the initial values of the spring constants were set equal to the experimental values of the apparent stiffness: or . Note that the values were computed by considering only residue pairs separated by a distance lower than the cutoff of 10 or 13 Å . For the sdENM , the values were set equal to the final values of the spring constants in the dENM , , for all amino acid types . A correction for sparse data was devised to ensure that tends to when the number of residue pairs of type is too small to obtain relevant estimations of . Instead of eq . 2 , we used the following definition to compute both the experimental and predicted apparent stiffness: ( 12 ) where , is the number of pairs of type in protein , and is the number of structures in the NMR ensemble of protein . The adjustable parameter can be understood as the number of occurrences of a residue pair , , that is needed to obtain a relevant estimation of . For a given type of residue pair , if , then no correction is necessary , and eq . 12 reduces to eq . 2 . On the contrary , if , then the data on pairs is considered too sparse to reliably estimate , and . We found that the value of has little impact on the overall quality of the model , as long as it is not too small ( ) , in which case aberrant values of are determined for some uncommon pairs , or too large ( ) , in which case the performances decrease because of a loss of information on sequence-specificity . The value of the parameter was set here to 500 . The values were rescaled after each iteration step , so that the average value of over all amino acid types is equal to 1 for pairs separated by a distance of 6 Å . Residue pairs of a given type for which ( after rescaling ) , were considered to establish no direct interaction: was set to 0 , and they were no longer considered in the iterative procedure . The performances of the new ENM variants after the first nine iteration steps are reported in Table S1 . The procedure converged rapidly for the dENM and the sdENM , and the final models were selected after 5 and 3 iteration steps , respectively . The sENM variants did not improve significantly with respect to the initial models ( ) , indicating that is a good approximation , contrary to . The procedure was thus stopped after one iteration step , for both the and the . To assess the robustness of the derivation scheme , and the sensitivity of the values determined for each type of residue pair to the content of the dataset , we calculated the bootstrap estimates of the 90% confidence intervals on , , and . For that purpose , the iterative procedure was repeated with 100 different datasets , each one consisting of 1500 NMR ensembles randomly picked , with replacement , from the original training dataset . All values , and the corresponding confidence intervals , are given in Dataset S1 . The ability of coarse-grained models to accurately describe protein dynamics is commonly evaluated by computing the Pearson correlation coefficient between predicted and experimental MSRF , , over all residues of a given protein: ( 13 ) where , for simplicity , was used instead of . There is indeed a direct relationship between the MSRF and the cristallographic B-factors: . and correspond thus here to the MSRF of residue extracted from the NMR data and predicted by the ENM , respectively . The scale of the predicted MSRF values depends on the scale of the spring constants , which are only defined up to a constant factor . This factor was determined , for each protein independently , by fitting the scales of the predicted and experimental MSRF , i . e . to ensure that: ( 14 ) Although it has been widely used in previous studies , is probably not the most adequate measure to evaluate the performances of coarse-grained models of protein dynamics . As pointed out previously [26] , [27] , it does indeed present several shortcomings: e . g . it is strongly affected by the presence of highly flexible regions , and does not account for possible flaws leading to an intercept of the regression line different from zero . Most importantly , the MSRF describe individual fluctuations but provide no information about the cooperative aspects of residue motions . The quality of the MSRF predictions gives thus no guarantee about the ability of the model to describe the cooperativity of protein dynamics . The provides an interesting example , for it performs quite well in predicting the MSRF but basically negates all cooperativity ( Figure 2 , Table 1 ) . Therefore , we introduce a new measure that exploits the information contained in the correlation matrix , to quantify the error on the estimation of the fluctuations of the interresidue distances: ( 15 ) where is the number of non-bonded residue pairs in protein , and are the experimental ( eq . 9 ) and predicted ( eq . 10 ) values of , respectively . is obtained after fitting the experimental MSRF with the predicted ones ( eq . 14 ) . The error is normalized by , which is the expected value of given the individual , anisotropic , fluctuations of both residues extracted from the NMR data , but neglecting all correlations between their respective motions . This normalization ensures that the contributions of the different pairs of residues are equivalent , and that the measure is not dominated by highly flexible regions . Both and are computed independently for each of the 349 proteins of our test set , and the average values are reported . We also report the short- ( ) , mid- ( ) , and long-range ( ) contributions to , obtained by considering only pairs separated by 0–15 Å , 15–30 Å , and more than 30 Å , respectively .
|
Decades of experimental evidence have underlined the fact that protein structures can hardly be considered as static objects . To understand how a protein achieves its biological purpose , it is therefore quite often necessary to unravel the complexity of its dynamical behavior . However , the definition of accurate and computationally tractable descriptions of protein dynamics remains a highly challenging task . Indeed , even though proteins are all built from a limited set of amino acids and local conformational arrangements , the specific nature of biologically relevant motions may vary widely from one protein to another , which constitutes a serious obstacle to the identification of common rules and properties . Here , instead of focusing on the study of a single protein , we adopt a more general perspective by condensing the information contained in a multitude of NMR conformational ensembles . This approach allows us to characterize the dynamical behavior of residues and residue pairs in a mean protein environment , independently of each protein's specific architecture . We describe how this analysis can be exploited to assess the performances of coarse-grained models of protein dynamics , to take advantage of existing experimental data for a more rational and efficient parametrization of these models and , ultimately , to improve our understanding of the intrinsic dynamical properties of amino acid chains .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2013
|
Effective Harmonic Potentials: Insights into the Internal Cooperativity and Sequence-Specificity of Protein Dynamics
|
Damage tolerance mechanisms mediating damage-bypass and gap-filling are crucial for genome integrity . A major damage tolerance pathway involves recombination and is referred to as template switch . Template switch intermediates were visualized by 2D gel electrophoresis in the proximity of replication forks as X-shaped structures involving sister chromatid junctions . The homologous recombination factor Rad51 is required for the formation/stabilization of these intermediates , but its mode of action remains to be investigated . By using a combination of genetic and physical approaches , we show that the homologous recombination factors Rad55 and Rad57 , but not Rad59 , are required for the formation of template switch intermediates . The replication-proficient but recombination-defective rfa1-t11 mutant is normal in triggering a checkpoint response following DNA damage but is impaired in X-structure formation . The Exo1 nuclease also has stimulatory roles in this process . The checkpoint kinase , Rad53 , is required for X-molecule formation and phosphorylates Rad55 robustly in response to DNA damage . Although Rad55 phosphorylation is thought to activate recombinational repair under conditions of genotoxic stress , we find that Rad55 phosphomutants do not affect the efficiency of X-molecule formation . We also examined the DNA polymerase implicated in the DNA synthesis step of template switch . Deficiencies in translesion synthesis polymerases do not affect X-molecule formation , whereas DNA polymerase δ , required also for bulk DNA synthesis , plays an important role . Our data indicate that a subset of homologous recombination factors , together with DNA polymerase δ , promote the formation of template switch intermediates that are then preferentially dissolved by the action of the Sgs1 helicase in association with the Top3 topoisomerase rather than resolved by Holliday Junction nucleases . Our results allow us to propose the choreography through which different players contribute to template switch in response to DNA damage and to distinguish this process from other recombination-mediated processes promoting DNA repair .
Proliferating cells are constantly exposed to DNA damage from both endogenous and exogenous sources . These DNA lesions can cause replication fork collapse and cell cycle arrest thereby posing a serious threat to genome integrity . To avoid the catastrophic consequences associated with fork demise , cells have evolved multiple mechanisms by which arrested or stalled replication forks can be rescued . These mechanisms are collectively referred to as DNA damage tolerance ( DDT ) mechanisms and involve factors belonging to two main repair pathways: the RAD52 homologous recombination ( HR ) and the RAD6/RAD18 post-replication repair ( PRR ) pathways [1] , [2] . The DDT mechanisms available in a cell are largely divided into two classes . One utilizes a combination of replicative and translesion synthesis ( TLS ) polymerases to replicate across the lesion , and in such situations the bypass can occur either in error-free or in error-prone manners [3] , [4] . The other DDT mechanism copies the information from undamaged segments of the genome , usually in an error-free manner and is referred to as template switch [2] , [5]–[7] . The mechanism , mode of action and factors implicated in template switch remain largely unknown [2] . Since template switch refers to a damage bypass process that operates in an error-free manner , it had been presumed to resemble and/or to involve recombination . Accordingly , distinct mechanisms involving recombination were proposed to account for template switch . One replication restart model of template switch , known also as the chicken foot model , proposes that the damage-bypass occurs at the site of fork stalling and involves pairing of the newly synthesized sister chromatids and replication fork regression [5] , [8] , [9] . The other model also proposes pairing of the newly synthesized sister chromatids at the fork or behind the fork in a manner that resembles the strand-exchange model of HR and leads to formation of sister chromatid junctions ( SCJs ) [6] , [7] , [10] . Whether template switch operates primarily at the fork or behind the fork could significantly affect the intermediate template switch DNA structure and has been an issue of debate [2] , [3] . Recent findings showing that restriction of the RAD18 pathway to G2 still supports lesion tolerance [11] , and that , during replication under damaging conditions when DDT factors are limiting , gaps accumulate behind the replication forks [12] , strongly corroborate the idea that template switch operates mainly in the rear of replication forks . Together with these findings , genetic and physical evidence have provided support for the model by which template switch occurs via recombination-like intermediates involving sister chromatid junctions ( SCJs ) [6] , [10] , [13] . In the recombination-like mode of template switch , annealing between the two newly synthesized sister-chromatids is expected to give rise to a D-loop recombination intermediate , which upon extension will lead to transient , hemicatenane-like or pseudo-double Holliday Junctions ( HJs ) structures ( Figure S1 ) or to double HJs [14] , [15] . In budding yeast , X-shaped intermediates with the expected biochemical properties of pseudo-double HJs and not of reversed forks or canonical HJs have been visualized during replication of damaged templates by using the 2D gel electrophoresis technique [10] . The resolution/dissolution of these DNA X-structures requires primarily the activity of the RecQ helicase Sgs1 ( BLM in mammalian cells ) and of the topoisomerase Top3 rather than that of Holliday junction nucleases [10] , [16]–[18] . If template switch operates mainly behind the forks to promote gap-filling , then factors required to promote replication completion and filling of gaps , such as those induced by UV irradiation , are expected to be required as well for the formation of template switch intermediates . Previous work in S . cerevisiae has shown that , following UV irradiation , DNA is initially synthesized as small discontinuous fragments , which are later converted to higher molecular-weight pieces similar in size to DNA from unirradiated cells [19] , [20] . Subsequent work has shown that these UV-induced gaps can be filled in a manner dependent on HR factors as well as proteins such as Rad18 , Rad5 and Mms2 , implicated in the error-free class of the PRR pathway [21]–[23] . Notably , both HR and error-free PRR factors have been shown to contribute to the formation of these template switch damage-bypass intermediates involving SCJs [10] , [13] , [15] , [24] . Altogether , these findings suggest that template switch represents a specific class of recombination process , involving in addition to traditional HR factors , other sets of enzymes with affinity for single-stranded ( ss ) DNA such as Rad5 and Rad18 [25] , [26] . The visualization of these intermediates in the proximity of replication forks , together with the evidence that these events are likely to be post-replicative , operating on the gaps left behind the forks [11]–[13] , suggest that template switch takes place during chromosomal replication although it does not interfere with the DNA synthesis process occurring at the replication fork . Thus , in terms of genetic requirements for error-free PRR and HR factors , post-replicative gap-repair and template switch appear to be similar . However , the exact role of HR and PRR factors in the formation/stabilization of template switch intermediates , the other players involved in this process and how these factors are coordinated with one another as well as with other gap-processing activities remain largely unknown . In this study , we planned to address these questions by dissecting the role of different factors in the formation of template switch intermediates . We analyzed factors that distinctly affect HR ( Rad55 , Rad57 , Rfa1 , Rad59 ) , factors implicated in gap processing and in the DNA damage/checkpoint response ( Exo1 , Rfa1 ) as well as the contribution of different DNA polymerase activities to the DNA synthesis step of template switch . HR mechanisms have been primarily modeled to explain double strand break ( DSB ) repair , and it has been demonstrated that the ends of a DSB are resected to expose 3′-single stranded ( ss ) tails that are bound by Rad51 and invade homologous duplex DNA , leading to a D-loop structure that can be subsequently extended and serve as a primer for DNA synthesis [27]–[29] . In S . cerevisiae , Rad52 plays an essential role in mediating strand exchange: the ssDNA is normally coated by the ssDNA binding protein RPA; Rad52 , which interacts with both Rad51 and RPA , overcomes the inhibitory role of RPA , recruits Rad51 , and promotes the formation of active Rad51 nucleofilaments that catalyze strand invasion [30] . The Rad51 paralogues , Rad55 and Rad57 , form a heterodimeric complex that interacts with Rad51 and has ssDNA binding activity but apparently no recombinase activity [31] . Similar to Rad52 , Rad55-Rad57 acts substoichiometrically to Rad51 to overcome the inhibitory role of RPA on Rad51-mediated strand exchange , indicative of a recombination mediator activity , although the mechanism of mediation is unknown [30] , [31] . Genetic and biochemical data suggests that Rad55-Rad57 also acts to stabilize the assembled Rad51 nucleofilaments [32] . The recombination defects of rad55 , rad57 are not always similar to the ones of rad51 . Notably , in spite of the generally much weaker phenotypes of rad55 , rad57 mutants in HR as compared to rad51 , rad57 cells are much more defective in spontaneous sister chromatid recombination ( SCR ) than rad51 [33] . Furthermore , in contrast to the defects of rad55 and rad57 mutants in DSB repair , which are suppressed by RAD51 overexpression , their SCR defect is only partly suppressed , suggesting that Rad55-Rad57 roles in DSB repair are distinct from their role in spontaneous SCR , which likely initiates from ssDNA gaps formed during replication [33] . Studies of the mammalian Rad51 paralogs Rad51C and Xrcc3 and of the rad57 mutants of Schizosaccharomyces pombe also suggested a possible role for Rad55-Rad57 in late recombination events , for instance by promoting the resolution of recombination intermediates or the displacement of the invading strand [34]–[37] . In budding yeast , the Rad55 protein is phosphorylated in a checkpoint-dependent manner under conditions of DNA damage , and this modification appears important for Rad55 function upon genome-wide genotoxic stress [38] . However , the effect of Rad55 phosphorylation on recombination and the recombination-mediator function of Rad55-Rad57 remain to be seen . The budding yeast Rad59 protein has similarity to the N-terminal region of Rad52 and is implicated in a subset of HR events , including spontaneous and damage-induced sister chromatid exchanges [39]–[42] and certain pathways of break-induced replication ( BIR ) [43] , [44]—an efficient HR process required to initiate replication when only one end of a DSB shares homology with a template [45]–[49] . In vitro studies have shown that Rad59 promotes strand annealing but is unable to stimulate Rad51-mediated strand exchange [50] . Understanding the contribution of different HR proteins to template switch will likely help elucidate the precise mechanism of this process and provide insights into how stalling or collapse of the replication fork triggers different recombination-mediated mechanisms in order to promote replication completion . Cells have a number of replicative and specialized TLS polymerases that participate in DNA replication as well as in different DNA repair events , but the replication activities required to promote the DNA synthesis step of template switch are presently unknown . The DNA polymerases α , δ , and ε ( Polα , Polδ , and Polε ) are the major replicative polymerases in eukaryotic cells , required to replicate DNA with high speed and fidelity [51] . Polα is tightly associated with the primase and is required for initiation of DNA synthesis on the leading strand as well as for the continuous synthesis of Okazaki fragments on the lagging strand . Although studies of Simian Virus ( SV40 ) DNA replication showed Polδ to be required for the extension of both leading and lagging strands [51] , and the polymerase activity of Polε in yeast cells is not essential for cell viability [52] , [53] , it is now generally agreed that both Polδ and Polε contribute to cellular DNA replication . Furthermore , mutational analyses of yeast suggest a differential involvement of Polδ and Polε in the synthesis of lagging and leading strands , respectively [54] , [55] . Loading of Polδ requires the proliferating cell nuclear antigen ( PCNA ) and replication factor C ( RFC ) , which function as a sliding clamp and a clamp loader , respectively . In addition , PCNA is also required for processive DNA synthesis by Polδ [51] and stimulates both Polε and Polδ in vitro [56] . In contrast to the replicative DNA polymerases , TLS polymerases such as Polη , Polζ and Rev1 in budding yeast , as well as their mammalian counterparts , have more open active sites , a property that allows these enzymes to accommodate bulky lesions and to promote replication through damaged templates [1] , [4] . It has been proposed that Polη has an additional role in promoting DNA synthesis during HR-mediated repair of DSBs [57] , [58] . In the present work we have examined the role of different recombination and replication factors in the formation of template switch intermediates during replication of damaged templates in vivo . By using a combination of genetic and physical assays , we show that factors implicated in the strand invasion step of HR , but not the strand annealing factor Rad59 , which is not essential for the strand exchange reaction , are required for the formation of the X-shaped template switch intermediates involving SCJs . Other factors , such as Exo1 , which is known to affect processing of recombination and replication intermediates , also play a role in promoting template switch . We demonstrate that TLS polymerases do not affect the efficiency of this process , while Polδ plays a major role in the DNA synthesis step of template switch . We thus identify a dual role for Polδ in genome replication and replication-associated repair and discuss mechanisms through which this functional versatility may be achieved .
Template switch events have been proposed to lead to the formation of SCJs in the proximity of damaged replication forks [10] , [17] . To define the factors that affect the efficiency of template switch , we used 2D gel electrophoresis to analyze the profile of replication intermediates formed at an early efficient origin of replication located on chromosome III in S . cerevisiae , ARS305 , and its flanking regions ( Figure 1 ) [13] . In this assay , synchronized yeast cells are released and allowed to undergo the following S phase in a medium containing the alkylating reagent methyl-methanesulfonate ( MMS ) . The pattern of replication intermediates is analyzed at different time points during replication . Previous results have shown that Rad51-dependent X-shaped intermediates sharing the properties of pseudo-double HJs form during replication of damaged templates and accumulate in mutants affecting the functionality of the Sgs1-Top3 complex [10] , most likely due to their impaired resolution [14] ( Figure S1 ) . Such molecules also form in wild-type cells , but are transient and scarce [10] , [13] , [24] . In order to facilitate our analysis of the contribution of different factors to the formation of the X-structures during replication of damaged templates , we took advantage of the sgs1 mutant background and compared the amount of X-molecules formed in sgs1Δ with those formed in double mutants of sgs1 and different repair genes . It has been demonstrated that Rad51 and Rad52 are required for template switch events leading to replication-associated SCJs in the proximity of replication forks [10] , [16] . Whether the function of Rad51 in this process is related to its ability to stabilize the X-structures , which could be achieved by binding of Rad51 to the ssDNA stretches of the hemicatenane-like intermediates and formation of paranemic junctions ( Figure S1 ) or of plectonemic DNA structures if one of the ssDNA strands is nicked , or rather to its active role in the formation of the structures , ( e . g . by promoting strand invasion as in typical HR reactions ) is not known . We examined the requirement of factors differentially affecting HR and strand exchange ( RPA , Rad55 , and Rad59 ) for template switch . Ablation of Rad55 , known to have mediator functions [28] , had an effect similar to that previously reported for RAD51 and RAD52 deletions [10] , [16] , abolishing the X-structures accumulating in the proximity of damaged replication forks in sgs1Δ ( Figure 2A ) . We note that in the graphs showing the quantification of the X-structure , the % of spike represents a normalized value to the maximum amount of X-molecules observed during the time course rather than the % of total replication intermediates ( see Materials and Methods for a detailed description of how quantification was performed ) . Although the mammalian orthologues of Rad55-Rad57 may also be implicated in late recombination events and/or resolution of recombination intermediates [35] , [36] , rad55 single mutants ( SGS1+ rad55Δ cells ) behaved similarly to wild-type cells in this assay ( Figure S2A ) . The replication and damage checkpoint Rad53 is required for the formation of the X-structure [10] and also affects HR , but its substrates involved in regulating HR remain unknown ( reviewed in [2] , [7] ) . Phosphorylation of Rad55 by the Rad53 checkpoint kinase was reported to be important for damage tolerance under conditions of genotoxic stress perhaps by promoting recombinational repair [38] . We analyzed the effect of the rad55 mutant in which the serines ( Ser , S ) 2 , 8 , and 14 phosphorylated by Rad53 were mutated to alanine ( Ala , A ) residues [38] . Unlike the RAD55 deletion , the rad55 phosphomutant did not affect the efficiency of template switch intermediates ( Figure 2A and see Figure S2B ) , suggesting that Rad55 phosphorylation by the replication checkpoint is not essential for this process . Differently from RAD55 and RAD51 deletions , ablation of RAD59 did not affect the X-molecule formation ( Figure 2B and see Figure S2B ) . Thus , the Rad55-Rad57 mediator of HR is required also for template switch , but the crucial substrate of Rad53 in this process is not Rad55 . To further examine the role of recombination mediators and possibly of the checkpoint response in this process , we examined the effect of mutations in the ssDNA binding protein RPA rfa1-t11 ( K45E ) , in template switch . Although RPA can exclude recombinases from HR substrates and therefore has an inhibitory role in the assembly of the presynaptic filament and strand exchange [28] , the rfa1-t11 mutation in the largest subunit of RPA is associated with a synapsis defect [59] , [60] , suggesting that RPA plays a role in DNA strand invasion during HR . Biochemical characterization of RPA containing the mutant Rpa1-K45E subunit showed it to be inefficient in Rad51-mediated strand exchange [60] . Consistent with this report , other studies have also found rfa1-t11 to be defective in recombinational repair [60]–[62] . We addressed whether rfa1-t11 impacts the accumulation of X-molecules in sgs1 mutants . As sgs1Δ was reported to be synthetic lethal or have severe growth defects with many mutants affecting replication and/or recombination [63]–[65] , we utilized a hypomorphic sgs1 mutant in which the helicase activity is impaired due to the insertion of the AUR1-C marker in the helicase domain of Sgs1 but which has milder phenotypes than sgs1Δ [66] . The sgs1::AUR1-C mutant was shown to accumulate X-molecules during replication of damaged templates [13] , [17] , in line with findings reported for other sgs1 alleles affecting the helicase activity of Sgs1 [67] . The rfa1-t11 mutation significantly decreased the accumulation of X-molecules in sgs1 ( Figure 3A ) . The original report on rfa1-t11 showed it to be proficient in DNA replication [62] . In agreement with this view , we also find that under conditions of DNA damage the profile of replication intermediates is not affected by the rfa1-t11 mutation at ARS305 or the flanking region ARS301 ( Figure S3 and data not shown ) . Thus , the effect of rfa1-t11 on the X-molecules formed in the proximity of replication forks cannot be attributed to general replication problems . Since RPA bound to single-stranded ( ss ) DNA is a signal for Rad53 checkpoint activation ( [7] and references therein ) , and Rad53 is required for the template switch X-formation [10] , it was important to establish whether the defects observed for rfa1-t11 in X-molecule formation under conditions of DNA damage are due to strand exchange defects and/or inability to boost Rad53 activation . The reports on the role of rfa1-t11 in checkpoint response are controversial: some studies found it defective for the replication/damage checkpoint [68]–[71] , while others found it proficient [62] , [72] . We found no evidence for impaired Rad53 activation in rfa1-t11 mutants either in spontaneous or MMS-treated conditions ( Figure 3B ) , suggesting that its effect in this context is more related to recombination and strand-exchange rather than checkpoint signaling . This result also allows us to conclude that the gaps formed during replication can still elicit a robust checkpoint response , mediated by RPA , in the absence of X-molecule formation . Exo1 is a member of the Rad2 family of structure-specific nucleases and possesses a 5′-3′ exonuclease activity ( [73] , [74] and references therein ) . Exo1 was implicated in processing abnormal structures arising at stalled replication forks [75] , [76] , in the checkpoint response [77] , DSB resection [78]–[80] , and other DNA repair events including mismatch and post-replication repair ( PRR ) ( reviewed in [74] ) . Here we addressed the involvement of Exo1 in the formation of SCJ molecules during replication of damaged templates . A combination of exo1Δ and sgs1Δ mutations leads to a severe growth defect , in accordance with previously published reports [63] . In attempts to overcome the cell-cycle delay and the general genome instability often associated with severe growth defects , we used the truncated sgs1 mutant described above and in previous works [66] . The sgs1 exo1Δ double mutant combination was still growing slowly in comparison with each single mutant , but the growth was not as severely affected as in sgs1Δ exo1Δ cells . We found that exo1Δ significantly reduced the amount of X-molecules accumulating in sgs1 mutants ( Figure 4 ) . The single mutant exo1 had a similar pattern of replication intermediates compared to wild-type cells ( Figure S4 ) . We addressed the possibility that specialized polymerases may be required for the DNA synthesis step of the template switch process . Mutations in DNA polymerases often sensitize cells to DNA damage , including MMS ( Figure S5 ) , but since this could reflect defects of these mutants in various DDT or repair pathways , it is hard to infer based on this sensitivity spectrum the contribution of the different polymerases to template switch . Previous work has shown that Polη can efficiently extend artificial D-loop substrates [57] and that chicken Polη affects Ig gene conversion tracts [58] . We thus analyzed the role of Polη , encoded by the RAD30 gene in yeast , in the formation of the X-structures accumulating in sgs1Δ , but observed no significant decrease in sgs1Δ rad30Δ mutants ( Figure 5A ) . In addition to Polη , other specialized TLS polymerases can facilitate damage-bypass; in budding yeast they are Polζ ( composed of the Rev3 catalytic subunit and the Rev7 non-catalytic subunit ) and Rev1 , which functions mostly in conjunction with Polζ but may also act to mediate the switching between TLS polymerases specialized for insertion and those required for extension [1] , [4] . We found that ablation of Polζ by REV7 deletion , or concomitant inactivation of Polζ and Rev1 ( rev7Δ rev1Δ ) , or of all TLS polymerases in yeast ( rev7Δ rev1Δ rad30Δ ) did not reduce the X-molecule accumulation in sgs1Δ cells ( Figure 5B and data not shown ) , suggesting that TLS polymerases do not play a major role in the DNA synthesis step required for template switch repair . The TLS mutants in a wild-type ( SGS1+ ) context did not affect the pattern of replication intermediates ( Figure S6 ) . We also note that this result does not imply that translesion synthesis is less important than template switch in DDT , as in our system TLS-mediated lesion bypass events not involving X-molecules are not detected . We also addressed the contribution of the main DNA polymerases required for elongation during eukaryotic genome replication: Polε and Polδ . We first examined by FACS the temperature at which these polymerase mutants do not impair cell cycle progression and found that at 30°C the Polδ mutant , cdc2-1 , and the Polε mutant , pol2-11 , are able to complete replication , while at 37°C these cells have a prominent delay in S-phase progression ( data not shown and see Figure S7 ) . This finding is in accordance with previous reports showing that cdc2-1 mutants fail to replicate approximately one third of their nuclear genome at restrictive temperatures [81] , [82] . To minimize the general replication defects inherently associated with Polδ and Polε mutations , we used permissive conditions of replication ( 30°C ) and analyzed the effect of these polymerase mutants on the X-molecules forming in the proximity of early origins of replications ( ARS305 ) , which are less prone to replication delays/problems as compared to later replication zones . The double mutants between sgs1 and either pol2-11 or cdc2-1 were viable , although sgs1 mutation induced lower viability of cdc2-1 cells at 30°C and increased the percentage of cells in G2/M under normal growing conditions ( Figure S7 ) . The cdc2-1 mutation in Polδ drastically diminished the amount of X-molecules in sgs1Δ , whereas sgs1 pol2-11 cells accumulated a similar amount of X-structures with sgs1 ( Figure 6 ) . The polymerase mutants , pol2-11 and cdc2-1 , in a wild-type ( SGS1+ ) context did not affect the pattern of recombination-like X-intermediates ( Figure S8 ) . To further establish that the effects of the cdc2-1 mutation on X-molecules are not due to general replication problems as opposed to a requirement for Polδ in template switch DNA synthesis , we attempted to gauge the differential effects of cdc2-1 in replication versus X-structure formation . For this purpose , we quantified the effect of cdc2-1 on both Y arcs ( representing replication forks ) and X-molecules . The results indicate that the reduction in the Y signal at ARS305 caused by the cdc2-1 mutation is much lower in magnitude than its effect on the X-molecules; accordingly the ratio of X-molecules versus Y arcs , which represents the amount of X-molecules normalized to the ongoing replication in the analyzed genomic fragment , is much lower in sgs1 cdc2-1 as compared to sgs1 ( Figure S9 ) . To further examine the effect of cdc2-1 on X-molecules versus DNA replication , we have also followed the progression of the forks to the ARS301 region ( Figure S10 ) , which is replicated passively by forks coming from ARS305 ( see Figure 1A ) . The progression of the replication forks in this region in a cdc2-1 background showed kinetics similar to those observed in SGS1+ cells . Notably , at all regions and time points analyzed the effect of cdc2-1 mutation on the X-signal was much more profound than its effect on the Y molecules ( Figures S9 and S10 ) . To further test the role of Polδ in template switch we examined the effects of the pol3-ct mutant , reported not to have defects in DNA replication [83] . We observed that the replication kinetics of pol3-ct as assessed by FACS are identical to that of wild-type ( Figure 7 and Figure S11 ) . When pol3-ct cells were crossed with sgs1 , we easily obtained pol3-ct sgs1 double mutants , and their doubling time at 30°C was similar to that observed for sgs1 single mutant ( pol3-ct: 90′±5′; sgs1: 100′±4′; pol3-ct sgs1: 100′±8′ ) . We found that the pol3-ct mutation reduces the amount of X-molecules accumulating in sgs1 to about 70%; although small , this effect was highly reproducible ( Figure 7 ) . The pol3-ct mutant had a similar pattern of replication intermediates with wild-type ( Figure S11 ) . Considering that sgs1 cdc2-1 is more slow-growing than cdc2-1 likely due to the accumulation of spontaneous lesions ( see Figure S7 ) , and pol3-ct has a minor effect on the X-accumulation ( Figure 7 ) , we decided to further analyze the effect of mutating the third , non-catalytic subunit of Polδ , Pol32 , known to affect the processivity of the Polδ complex [84] , [85] . The combination of pol32 and sgs1 mutations leads to marked slow growth at 28°C and 30°C , and lethality at lower temperatures , such as 23°C , which is still permissive for pol32 [11] . To override the undesirable effects on replication caused by the delayed growth of the double mutants , we employed a conditional SGS1 system , GAL-SGS1 , in combination with the pol32 mutation , previously reported [11] , in which SGS1 shut-down is induced only during the course of the experiment , by addition of glucose ( Figure 8 ) . The pol32 mutation had a clear effect in reducing the X-accumulation under such experimental settings . Notably , in these conditions the progression through S-phase of the double mutant or of the pol32 single mutant did not appear to be impaired ( Figure 8 and Figure S12 ) , in line with previous reports showing that the problems experienced by sgs1 pol32 cells and leading to low-viability are caused by G2 events [11] . Altogether , these last sets of results suggest that Polδ plays an important role in mediating the DNA synthesis step of template switch . Since Polδ is also required for bulk replication , and most template switch are likely to occur behind replication forks [11]–[13] , our results imply that Polδ acts in a distributive manner , both at the fork , to promote DNA replication , and behind the fork , to promote post-replicative repair events such as template switch .
Template switch , thought to be implicated in both gap-filling and restart of replication forks stalled by DNA lesions , plays an important role in DNA metabolism and may protect against chromosomal instability via stabilizing repetitive sequences , preventing translocations and instability associated with certain genomic disorders [2] , [86] . The mechanism and genetic factors promoting or controlling template switch are not well understood . The goal of our present study was to deepen our understanding of how template switch occurs and is regulated . To this purpose , we addressed the contribution of different factors to the formation of replication-associated SCJs , thought to represent template switch intermediates [5] , [6] , [10] , [13] . Several observations concur with the idea that template switch occurs mainly behind replication forks to fill in the ssDNA gaps accumulating under damaging conditions [11]–[13] ( Figure 9 ) . We do not exclude the possibility that a fraction of template switch events may occur at the site of lesion via other DNA intermediates such as reversed forks; future studies will be needed to both elucidate the proportion of bypass events occurring at the site of the lesion as well as to understand whether alternate pathways leading to regressed fork formation are subjected to regulation in the cells exposed to genotoxic stress . Besides alkylating bases , MMS may cause DSBs , although this notion remains controversial [87] , [88] . To what degree the factors implicated in DSB repair are required for replication-associated template switch and other SCR events is not known and thus , automatic extension of the existing genetic data aimed at elucidating the DSB response pathways to other recombination-like mechanisms involving HR factors , such as template switch , should be viewed cautiously . HR is most active in S and G2/M phases of the cell cycle , but it is likely that different lesions or DNA substrates will involve distinct crosstalks between repair proteins and cell-cycle or damage response signaling pathways in order to promote DDT [2] , [89] . Furthermore , whether DSB repair pathways operate primarily in S-phase to restart forks or in G2 to promote replication completion and DNA repair , remains an issue of debate . Considering that the X-structures generated during replication under conditions of MMS damage do not represent canonical Holliday Junctions , and furthermore , that Sgs1-Top3 , and not Holliday Junction nucleases , represent the main activities required for the X-resolution in S-phase [10] , [18] , it seems logical to assume that the template switch X-structures formed in the proximity of replication forks require distinct sets of factors and are , at least partly , different from the DNA intermediates generated during other DSB repair processes . On the other hand , it is also reasonable to foresee that the pathways implicated in replication-associated HR-mediated DSB repair ( BIR ) and template switch-mediated gap filling will share common enzymatic activities . In this study we have uncovered both similarities ( Rad51 , Rad55-Rad57 , Rfa1 , Polδ ) and dissimilarities ( Polε , Rad59 ) between the factors implicated in these two processes ( see also below ) , but future studies will be needed to deepen our understanding of the mechanisms through which different recombination-mediated pathways are coordinated with one another and other cell-cycle signaling networks to promote damage avoidance . By using 2D gel analysis of replication intermediates and a genetic set-up in which the template switch DNA structures are enriched by preventing their resolution mediated by Sgs1-Top3 [10] , [13] , we established here that , in addition to factors known to be required also for the strand invasion step of HR , Polδ plays an important role in the efficient formation of template switch intermediates . Previous genetic studies have also defined a role for Polδ in gap and DSB repair [46] , [90]–[93] . While the pol2-11 mutation affecting Polε function was previously reported to affect the DNA synthesis step of break-induced replication ( BIR ) [46] , it did not significantly affect the formation efficiency of forming template switch intermediates in the proximity of replication forks ( Figure 6A ) , suggesting that differences exist even at the elongation step between different recombinational repair pathways activated by replication problems . The cdc2-1 allele in the catalytic subunit of Polδ , previously shown to be defective in the repair of MMS-induced single-strand breaks at non-permissive temperatures [94] , [95] , also affects the efficiency of template switch intermediates at temperatures that are permissive for replication ( Figure 6B ) . The observation that cdc2-1 cells are able to establish forks even at late replication zones under conditions permissive for growth with kinetics similar to those observed in wild-type cells ( Figure 6 and Figure S10 ) compellingly suggests that the defect of cdc2-1 in X-formation reflects a bona-fide role of Polδ in template switch DNA synthesis ( Figure 8 ) and that this defect cannot be fully attributable to the replication defects of cdc2-1 cells [81] , [82] ( Figure 6 and Figure S9 ) . Work done on other alleles of Polδ , such as the pol3-ct allele having a truncation that removes the last four amino acids of the Pol3 protein , has uncovered a role for Polδ in HR-mediated DNA synthesis during gene conversion ( GC ) [93] , a DSB repair pathway occurring when homology with both the DSB ends is present , as well as in BIR [92] . In our system , we find that pol3-ct had a small but reproducible effect on the efficiency of forming template switch intermediates ( Figure 7 ) . Furthermore , by using a conditional sgs1 system in combination with a null mutation in the non-essential subunit of Polδ , Pol32 , we were able to establish that impairment of Polδ function and processivity by the pol32Δ mutation , largely impacts on the ability of cells to undergo template switch ( Figure 8 ) . Specialized TLS polymerases may also contribute to the DNA synthesis step of template switch . Although Polη can extend the invading 3′end of a D-loop intermediate in vitro and deletion of chicken Polη reduces the frequency of DSB-induced gene conversions [57] , [58] , we could not assign a significant role for Polη or other TLS polymerases in the DNA synthesis step of template switch ( Figure 5 ) . However , given the biochemical data indicating that Polη could promote GC of up to about 80 nucleotides in vivo [58] , while the gaps forming during replication of damaged templates in yeast have been estimated at approximately 400 nucleotides in size [12] , it is possible that Polη plays a role , redundant with other polymerases such as Polδ , but which may therefore be too subtle to appreciate in our system . Nevertheless , we note that our results are in line with previous studies that have reported no role for Polη in the DNA synthesis steps occurring during GC or BIR in budding yeast [92] , [93] . Our studies do not rule out a possible contribution of other DNA polymerases such as Polη and Polε to the DNA synthesis step of template switch; they do show , however , that TLS polymerases and Polε do not support wild-type levels of template switch when Polδ is inactive ( Figure 9 ) . Since most template switch events are likely to occur in the rear of replication forks [11] , [13] , the emerging question is how the function of Polδ is divided to suit both its role in catalyzing highly processive replication at replication forks and its repair role in template switch . One line of control could be achieved by upregulating the amounts of DNA polymerases to greater levels than the ones strictly needed to perform DNA replication [96]; this could explain why limiting amounts of Polδ were shown to be associated with repair defects and chromosomal instability [97] . The functional versatility and local distribution of Polδ may be mediated by its interaction with different sets of proteins , such as those between Polδ and differently modified forms of PCNA ( Figure 9 ) . Indeed Rad18-Rad5-Mms2-mediated polyubiquitination of PCNA [98] promotes template switch [13] and stimulates the repair activity of Polδ [11] , [99] , [100] . Recruitment of Rad18 to RPA-coated ssDNA containing DNA lesions [101] may subsequently influence a number of processes required for efficient template switch such as the remodeling of PCNA through posttranslational modifications [98] , the repair function of Polδ [11] , [99] , or the efficiency of HR per se [102] ( Figure 9 ) . The replication checkpoint kinase Rad53 is implicated in template switch [10] , but its role and targets in this process remain topics of investigation . Although strand exchange mediator activities , including that of Rad55 , are crucial for the formation of template switch intermediates ( Figure 2 and Figure S2 ) , we found that direct phosphorylation of Rad55 by Rad53 [38] is not required in this specific context ( Figure 2A ) . Gap extension before homology search can initiate is also expected to favor template switch; our results suggesting that Exo1 promotes template switch ( Figure 4 ) classify it as a new factor of the error-free PRR and may reflect its role in processing the gaps formed behind replication forks under conditions of genotoxic stress ( Figure 9 ) . Extension of the gaps could be expected to lead to longer stretches of ssDNA-RPA and robust Rad53 activation; although we did not observe any obvious defect of exo1 mutants in activating Rad53 following DNA damage ( Figure S13 ) , the effect may be too subtle or redundant with other nuclease-mediated pathways . Indeed , other nuclease activities , such as those of Sae2 and of the Mre11-Rad50-Xrs2 ( MRX ) complex have been shown to work together with Exo1 in other settings related to recombinational repair [74] , [78]–[80] . Although we could not directly assess the contribution of these nucleases to template switch due to the severe growth defects of the nuclease double mutants with sgs1 , we do not exclude the possibility that these three nucleases ( Exo1 , Sae2 , Mre11 ) may all contribute independently or work with each other to promote template switch . Uncontrolled and extended gap processing – as seen in checkpoint mutants [12] , [103] – should be avoided in order to preserve genome integrity , and if Exo1 is the key factor mediating these events [75] , [76] , it is reasonable to think that mechanisms exist to keep its activity under control . Indeed , recent studies have found that Rad53-dependent phosphorylation of Exo1 may limit ssDNA accumulation and act as a feedback to restrain checkpoint activation [77] . Since Rad53 phosphorylates Exo1 following MMS damage [77] , [104] , it is possible that the crosstalk between Exo1 and Rad53 is important for the regulation of template switch events as well as other mutagenic processes . Another possible role for Rad53 and the DNA damage checkpoint in template switch may be to favor or limit the strand invasion step or the processing of the recombination intermediates if the DNA synthesis step is hindered , for instance by controlling the activity of nucleases that may process the stalled/abortive recombination intermediates . Future challenges will lie in characterizing how other players , including factors required for chromatin organization or sensing the topological status of the DNA , cooperate with repair and replication factors to modulate the division of labor between polymerases and to enable the functional versatility of proteins such as PCNA and Polδ . Finally , understanding how different DNA synthetic and repair demands are orchestrated to prevent the accumulation of DNA damage and maintain chromosomal stability has important implications for enhancing our knowledge of how cells are protected from cancer-causing alterations .
The yeast strains used in this study are mostly derivatives of W303 and the relevant genotypes are shown in Table S1 . Unless otherwise indicated , strains were grown at 28°C in YP-media containing glucose ( 2% ) , YPD , as carbon source , with the exception of experiments presented in Figure 8 and S12 , where raffinose ( 1 . 8% ) and galactose ( 0 . 2% ) were used instead . Cells were synchronized either in metaphase by adding nocodazole to a final concentration of 10 mg/ml together with DMSO to a final of 1% v/v , for about 2 . 5 hr , or in G1 with α-factor to a final concentration of 3–4 µg/ml . The release from the synchronization was performed as previously described , in YPD containing MMS at a final concentration of 0 . 033% v/v [17] , with the exception of the experiment in Figure S12 , where the release was done in YP-media supplemented with raffinose ( 1 . 5% ) and galactose ( 0 . 5% ) and containing MMS 0 . 033% . Log phase cells were counted and 10-fold series dilutions were spotted on plates containing various concentrations of MMS and incubated at the indicated temperatures . Western blot analysis and TCA extraction of yeast proteins was performed as previously described [18] . Rad53 was detected with the mouse monoclonal EL7 antibody ( a gift from A . Pellicioli ) . FACS analysis was carried out as previously described by staining cells with propidium iodide as described in [10] , with the exception of the experiments presented in Figure 8 , Figure S7 , and Figure S12 , where SYTOX green ( Invitrogen ) solution was used instead as described in [18] . Purification of DNA intermediates and the 2D gel procedure were carried out as previously described [10] , [17] . We note that each experiment was performed independently at least twice with qualitatively identical results and that a representative result is shown in the figures . The DNA samples were digested with HindIII and EcoRV and analyzed by 2D gel with probes against ARS305 and/or the flanking region ARS301 , or alternatively digested with NcoI and analyzed with probes recognizing ARS305 . Quantification of signals of X-shaped intermediates was performed using the Image Quant software . For each time point , areas corresponding to the monomer spot ( M ) , the X-spike signal and a region without any replication intermediates as background reference were selected and the signal intensities ( SI ) in percentage of each signal were obtained . The values for the X and monomer were corrected by subtracting from the SI value the background value after the latter was multiplied for the ratio between the dimension of the area for the intermediate of interest and for background . Thus , the values for X and M were calculated in the following way: The relative signal intensity for the X was then determined by dividing the value for X with the sum of the total signals ( the sum of the X and monomer values ) . The resulting values for X signals were then normalized and converted to percentage by using the highest value number of X for each experiment as 100 and normalizing the other values to it . At least three independent experiments conducted with isogenic strains were used for calculation of standard deviation . When mentioned , the value for the Y arc signal was calculated in a manner analogous to the one for the X , Value for Y = SI ( Y ) - [SI ( background ) ( area ( Y ) /area ( background ) ) ] , and then the ratio X/Y was derived .
|
Completion of DNA replication is essential for cellular survival . Both endogenous processes and exogenous DNA damage can lead to lesions that impede DNA replication or result in an accumulation of DNA gaps . Recombination plays an important role in facilitating replication completion under conditions of replication stress or DNA damage . One DNA damage tolerance mechanism involving recombination factors , template switch , uses the information on the newly synthesized sister chromatid to fill in the gaps arising during replication under damaging conditions . This process leads to the formation of repair structures involving sister chromatid junctions in the proximity of replication forks . The template switch structures can be detected by 2D gel electrophoresis of replication intermediates as cruciform , X-shaped intermediates . Additional factors and regulatory pathways are required for the resolution of such structures to prevent their toxic effects . In this work , we have dissected the recombination/replication factors required for the formation of template switch intermediates . Another recombination mechanism , which has been implicated in the restart of collapsed forks , is break-induced replication ( BIR ) . This study allows us to identify the core factors required for template switch and to distinguish this process from other recombination-mediated processes promoting DNA repair .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry/replication",
"and",
"repair",
"genetics",
"and",
"genomics/gene",
"function",
"genetics",
"and",
"genomics/chromosome",
"biology"
] |
2010
|
Replication and Recombination Factors Contributing to Recombination-Dependent Bypass of DNA Lesions by Template Switch
|
Although increased capillary permeability is the major clinical feature associated with severe dengue infections the mechanisms underlying this phenomenon remain unclear . Dextran clearance methodology has been used to investigate the molecular sieving properties of the microvasculature in clinical situations associated with altered permeability , including during pregnancy and in various renal disorders . In order to better understand the characteristics of the vascular leak associated with dengue we undertook formal dextran clearance studies in Vietnamese dengue patients and healthy volunteers . We carried out serial clearance studies in 15 young adult males with acute dengue and evidence of vascular leakage a ) during the phase of maximal leakage and b ) one and three months later , as well as in 16 healthy control subjects . Interestingly we found no difference in the clearance profiles of neutral dextran solutions among the dengue patients at any time-point or in comparison to the healthy volunteers . The surface glycocalyx layer , a fibre-matrix of proteoglycans , glycosaminoglycans , and plasma proteins , forms a complex with the underlying endothelial cells to regulate plasma volume within circumscribed limits . It is likely that during dengue infections loss of plasma proteins from this layer alters the permeability characteristics of the complex; physical and/or electrostatic interactions between the dextran molecules and the glycocalyx structure may temporarily restore normal function , rendering the technique unsuitable for assessing permeability in these patients . The implications for resuscitation of patients with dengue shock syndrome ( DSS ) are potentially important . It is possible that continuous low-dose infusions of dextran may help to stabilize the permeability barrier in patients with profound or refractory shock , reducing the need for repeated boluses , limiting the total colloid volume required . Formal clinical studies should help to assess this strategy as an alternative to conventional fluid resuscitation for severe DSS .
Dengue infection is increasingly being recognised as a major burden on global health [1] . Infection with any of the four viral serotypes may result in asymptomatic infection , or cause a variety of disease manifestations ranging from non-specific fever to a severe syndrome characterised by increased vascular permeability , deranged haemostasis and thrombocytopenia [2] . Considerable volumes of plasma can be lost from the intravascular compartment resulting in potentially fatal dengue shock syndrome . Despite the fact that vascular leakage is the pathognomonic feature of severe dengue , little is known of the mechanisms underlying the change in permeability . The prevailing view is that dengue infection triggers an immunopathogenic cascade that alters microvascular structure or function in some as yet undefined way , resulting in a transient , spontaneously-reversible increase in permeability [3] . Plasma volume is normally regulated within tightly circumscribed limits by complex homeostatic mechanisms , with plasma and interstitial fluid existing in dynamic equilibrium separated by the semi-permeable capillary wall [4] , [5] . The surface glycocalyx layer , a highly anionic fibre-matrix of proteoglycans , glycosaminoglycans , and plasma proteins anchored in the plasma membrane of endothelial cells , is considered to be the primary barrier , functioning as a molecular sieve to selectively restrict molecules within the plasma and limit access to the endothelial cell layer which forms the secondary barrier . In general , small molecules are freely filtered , permeability decreases as molecular size increases , and large or negatively charged molecules are relatively protected within the circulation . Although the basic mechanisms governing filtration at the glomerulus are intrinsically similar to those operating in the rest of the microvascular circulation , additional highly specialised mechanisms serve to protect the intravascular albumin pool in the face of a glomerular filtration rate of the order of 180 litres/day in adults , resulting in production of virtually protein free urine in normal circumstances [5] , [6] . Increases in glomerular protein permeability demonstrated in patients without renal disease are likely to indicate substantial increases in systemic vascular permeability . Recent research in various diseases with altered capillary permeability such as diabetes and coronary atherosclerosis indicates a range of pathological effects on the surface glycocalyx layer of the systemic microvasculature , with early microalbuminuria thought to reflect these systemic changes rather than being directly of renal origin [7]–[10] . Knowledge of the sieving properties of capillary beds can be obtained by measuring the fractional clearance of test macromolecules [11] , [12] . The clearance of a molecule is equal to the excretion in urine per unit time divided by the concentration in plasma , provided no modification occurs in the kidneys . Relating this clearance to that of a freely filtered reference marker that is neither secreted nor absorbed adjusts for the glomerular filtration rate . Since many endogenous proteins are secreted and/or absorbed in the renal tubules , fractional clearance measurements of non-reabsorbable synthetic polymers such as dextrans are preferred . Both hypoalbuminaemia and proteinuria are seen during dengue infections , without evidence of renal impairment . Marked increases in fractional clearances of several endogenous proteins , notably albumin , have been documented during the critical phase for leakage [13] . In order to better understand the characteristics of the vascular leak associated with dengue , we undertook formal dextran clearance studies in a group of Vietnamese patients with dengue and a similar group of healthy volunteers as controls .
Serial dextran clearance studies were carried out on adult male patients at the Hospital for Tropical Diseases ( HTD ) of Ho Chi Minh City , all with dengue and evidence of vascular leakage , a ) during the phase of maximal leakage and b ) one and three months later . Similar clearance studies were performed on healthy male volunteers with no history of febrile illness for 3 months . All subjects gave written informed consent , and approval was obtained from HTD and the Oxford Tropical Research Ethics Committee . All studies took place in a dedicated pharmacology research room on a high-dependency ward supervised by an experienced clinician . Following established methodology , a 50 ml bolus of 10% Dextran 40 in normal saline ( Rheopolyglukin , Kraspharma , Russia ) plus 5% inulin ( Inutest® 25% , Fresenius Kabi , Austria ) was administered over 10 minutes to supine resting subjects , followed by a maintenance infusion of the same solution at 1 ml/min for 2 hours [12] . Inulin is commonly used as the standard for measuring ultrafiltration of small solutes . Blood samples ( 2 mL ) were collected from the contralateral arm at baseline , and again after 70 and 130 minutes , to coincide with the mid-point of timed urine collections performed between 60–80 and 120–140 minutes immediately after the subject had voided the bladder . Subjects were encouraged to drink plenty of water and vital signs were recorded before and every 30 minutes during the test . The exact timings of fluid administration and sample collection were carefully recorded . Samples were immediately separated and stored at −20°C until analysed subsequently in batches . All samples from one subject were analysed on the same day by high performance liquid chromatography ( HPLC ) using a LaChrom Elite system ( Merck , Germany ) including one auto-sampler with cooling unit ( L-2200 ) , two high throughput analysis pumps ( model L-2130 ) , a column oven ( model L-2350 ) and a Diode Array Detector ( model L-2455 ) . All plasma and urine samples were first deproteinized with 20% trichloroacetic acid . Fractional clearances of dextran were calculated as follows: θDex = ( U/P ) Dex/ ( U/P ) Inu , which refer to the urine-to-midpoint plasma concentration ratios of dextran and inulin respectively . The fractional volume was determined as Kav = ( Ve−V0 ) / ( Vt−V0 ) , where Ve is the elution volume of each dextran in individual fractions and Vt is the total volume of the column , estimated with glucose . The molecular radius was calculated by Rst = 0 . 33×M0 . 463 , M being the molecular weight at the peak elution position of each standard dextran . Rst of individual fractions was estimated by the relationship between Kav and Rst of the standard dextran solutions .
Between November 2008 and February 2009 15 male patients , median ( range ) 25 ( 18–30 ) years , were recruited to the study . All had clinically suspected dengue with evidence of vascular leakage ( progressively rising haematocrit ( 11/15 ) and/or pleural effusion or ascites on ultrasound ( 5/15 ) and/or hypoalbuminaemia ( 9/15 ) ) at the time of study , but were well-compensated cardiovascularly . One patient was diagnosed as having vascular leakage on the basis of development of significant hypoalbuminaemia alone , with the plasma albumin dropping from 47 g/L to 39 . 5 g/L on day 4 of illness . This patient received maintenance parenteral fluid therapy , likely masking the rise in haematocrit usually seen in association with such a reduction in plasma albumin . All initial studies took place between days 4–6 of illness ( the first day of illness was defined as the day of fever onset ) , with 13/15 on day 5 . No patient developed shock and all recovered fully with conventional symptomatic care . Throughout the course of the illness 11/15 patients experienced either skin or minor mucosal bleeding and the median ( range ) platelet nadir was 25 , 000 ( 8 , 000–111 , 000 ) cells/µl . All patients were subsequently confirmed to have dengue using standard serological/virological methods ( 5 patients sero-converted , while in 10 a virus was identified on RT-PCR , 5 DENV_1 , 3 DENV_2 and 2 DENV_3 ) [16] . Sixteen healthy male students , median ( range ) 24 ( 23–28 ) years , acted as controls . Clearance studies were generally well tolerated although one patient developed a minor febrile reaction shortly after the dextran infusion . Figure 1 depicts the dextran fractional clearance curves for the acute and convalescent studies performed in the dengue patients , together with results for the control subjects . All curves represent the results from the first timed collection . No detectable differences were apparent between any of the curves . There were no differences in inulin clearances between the groups during the acute illness or between acute illness measurements and follow-up measurements within any of the groups , consistent with the persistently normal renal function indices observed in all patients throughout the duration of the study .
Fractional clearance methodology is well established for investigating renal disorders and has provided useful insights into the pathogenesis of the systemic leak associated with meningococcal septicaemia and dengue , in both of which considerable increases in clearances of endogenous proteins were observed consistent with the severity of leakage [13] , [17] . However , in these formal dextran clearance studies in dengue patients we found no difference in the clearance profiles of polydisperse neutral dextran solutions either a ) during the period of maximal leakage compared to their own follow-up profiles , or b ) compared to the profiles observed in a similar group of healthy volunteers . Although all the patients had clear evidence of leakage , none developed DSS and they were thus comparatively less severe than the patient group we previously studied , in whom marked increases in protein clearances were demonstrated in association with DSS . Unfortunately nevertheless we did not measure clearances of endogenous proteins prior to the dextran clearance studies so we cannot be sure that they did have increased protein losses at baseline; however ongoing studies in our unit indicate that all patients with dengue have some degree of microalbuminuria and that most patients with demonstrable vascular leakage have proteinuria on dipstick testing . One possible explanation for this unexpected finding of normal dextran clearances during the period of vascular leakage is that a physical and/or electrostatic interaction occurs between the dextran molecules and the surface glycocalyx/endothelial cell complex , temporarily improving its permeability characteristics during the infusion and rendering the technique unsuitable for assessing permeability in such patients . Both structural and functional characteristics of the glycocalyx are known to depend in part on integration of endogenous plasma proteins , especially albumin , within the layer [4] , [6] . Although the pathogenesis of dengue associated vascular leakage remains unknown , significant losses of albumin and other proteins from the circulating plasma occur , and some washout of proteins from the glycocalyx layer must follow , compromising the function of the layer . However , replacement of lost proteins by synthetic colloid molecules may temporarily restore the permeability characteristics of the barrier . From animal studies incorporation of small dextran molecules into the glycocalyx layer is known to occur almost immediately , while larger molecules usually remain restricted within the circulating plasma for several hours [18] . In situations where the protein content of the glycocalyx is markedly reduced , it is possible that the rate and range of colloid molecules rapidly entering the layer increases . Colloid infusions have been shown to restore permeability in a pig heart model of ischaemia/reperfusion injuries [19] , [20] , and clinical experience using neutral dextran solutions for resuscitating patients with DSS indicates that the volume effect of a given bolus considerably exceeds the actual volume infused , supporting the idea that colloid molecules transiently restore the permeability barrier thereby reducing the leakage of proteins temporarily [21] . In recent years there has been increasing interest in investigating the mechanisms and regulation of glycocalyx synthesis and turnover , in the hope that such knowledge might lead to the development of novel therapeutic strategies to reduce pathologically increased permeability [5] . Thus , heparin injections are known to mitigate the severe protein losing enteropathy that develops in some children following complex cardiac surgery due to loss of heparan sulfate proteoglycans from the intestinal epithelial glycocalyx [22] , [23] . Use of heparin is limited by adverse effects , but alternative strategies to restore gut epithelial glycocalyx function using synthetic heparin-like compounds are being actively pursued [24] . Similarly , alterations to the glycocalyx layer in the coronary microcirculation are thought to contribute to myocardial ischaemia and the subsequent reperfusion injuries that result in localised vascular leakage; prevention of glycocalyx loss and/or restoration of damaged glycocalyx are being investigated as potential interventions to reduce myocardial damage in these circumstances [20] , [25] , [26] . The implications of these findings for resuscitation of patients with DSS are potentially important . Prompt restoration of circulating plasma volume is the cornerstone of therapy , and WHO management guidelines recommend initial resuscitation with crystalloid solutions followed by boluses of colloids for patients with recurrent or refractory shock , aiming to achieve cardiovascular stability with the minimum volume necessary to maintain vital organ functions until normal permeability is restored [2] , [17] . However , patients with severe leakage are intrinsically difficult to manage and often require multiple boluses of colloid during this period , putting them at significant risk of respiratory distress due to fluid overload , and of haemorrhagic complications due to colloid induced haemostatic dysfunction compounding the intrinsic coagulopathy and thrombocytopenia induced by dengue infection . These results indicate that an alternative resuscitation strategy may be beneficial in some circumstances – in severe cases continuous low-dose infusions of dextran may help to stabilise the permeability barrier , reducing the need for repeated boluses and limiting the total volume of colloid infused , thereby minimising adverse effects on coagulation and respiratory function . Further work is needed to confirm these findings in patients with demonstrably increased fractional protein clearances and to investigate whether the effect is similar with other colloid solutions . If the findings are confirmed , formal randomised and blinded intervention studies should help to address the question of whether continuous low dose colloid infusions improve outcome in severe DSS .
|
Dengue is a potentially serious common viral infection with no specific treatment . Plasma leakage from small blood vessels is the major severe problem , but we do not understand how this occurs . Techniques using controlled infusions of carbohydrate solutions , combined with careful measurement of the rate that the different-sized molecules clear from the circulation , have been successfully used to investigate leakage in other situations . We performed carbohydrate clearance studies in 15 Vietnamese adult males with dengue and plasma leakage , comparing results obtained during the acute illness with recovery values , and results from a group of healthy volunteers . However , we found no differences between any of the clearance profiles measured . One possible explanation may be that the carbohydrate molecules interact with blood vessels , temporarily restoring their normal barrier function . Although this means that the technique is unsuitable for investigating leakage in dengue patients , the implications for management of patients with severe leakage resulting in shock are potentially important . Patients with profound shock are usually managed with intermittent large boluses of carbohydrate or similar solutions , sometimes causing severe side-effects; however if continuous low-dose infusions actually stabilized the permeability barrier , this might reduce the need for repeated boluses , thereby minimizing these adverse effects .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"immune",
"physiology",
"clinical",
"research",
"design",
"integrative",
"physiology",
"dengue",
"anatomy",
"and",
"physiology",
"immunology",
"pediatrics",
"critical",
"care",
"team",
"organization",
"physiological",
"processes",
"hemorrhagic",
"fever",
"with",
"renal",
"syndrome",
"fluid",
"management",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"fluid",
"physiology",
"infectious",
"diseases",
"travel-associated",
"diseases",
"pediatric",
"critical",
"care",
"pediatrics",
"and",
"child",
"health",
"critical",
"care",
"and",
"emergency",
"medicine",
"clinical",
"immunology",
"physiology",
"viral",
"diseases",
"resuscitation"
] |
2011
|
Dextran Fractional Clearance Studies in Acute Dengue Infection
|
Efficient adaptation to iron starvation is an essential virulence determinant of the most common human mold pathogen , Aspergillus fumigatus . Here , we demonstrate that the cytosolic monothiol glutaredoxin GrxD plays an essential role in iron sensing in this fungus . Our studies revealed that ( i ) GrxD is essential for growth; ( ii ) expression of the encoding gene , grxD , is repressed by the transcription factor SreA in iron replete conditions and upregulated during iron starvation; ( iii ) during iron starvation but not iron sufficiency , GrxD displays predominant nuclear localization; ( iv ) downregulation of grxD expression results in de-repression of genes involved in iron-dependent pathways and repression of genes involved in iron acquisition during iron starvation , but did not significantly affect these genes during iron sufficiency; ( v ) GrxD displays protein-protein interaction with components of the cytosolic iron-sulfur cluster biosynthetic machinery , indicating a role in this process , and with the transcription factors SreA and HapX , which mediate iron regulation of iron acquisition and iron-dependent pathways; ( vi ) UV-Vis spectra of recombinant HapX or the complex of HapX and GrxD indicate coordination of iron-sulfur clusters; ( vii ) the cysteine required for iron-sulfur cluster coordination in GrxD is in vitro dispensable for interaction with HapX; and ( viii ) there is a GrxD-independent mechanism for sensing iron sufficiency by HapX; ( ix ) inactivation of SreA suppresses the lethal effect caused by GrxD inactivation . Taken together , this study demonstrates that GrxD is crucial for iron homeostasis in A . fumigatus .
Iron is an essential trace element for almost all organisms in all kingdoms of life . On the other hand , iron excess is toxic . Therefore , to maintain cell homeostasis , the balance between iron uptake and iron consumption has to be tightly regulated . Previous studies have shown that iron homeostasis in the pathogenic mold Aspergillus fumigatus is mainly regulated by two transcription factors , SreA , the repressor of siderophore biosynthesis and reductive iron assimilation [1] , and HapX , which is a repressor of iron-consuming pathways and activator of iron acquisition [2] . Moreover , HapX is essential for adaptation to iron excess . When iron concentrations increase , HapX changes its function from a repressor to an activator of iron-consuming and detoxifying pathways to avoid iron toxicity . Consequently , HapX is crucial for adaptation to both iron starvation ( -Fe ) and high iron concentrations ( hFe ) , i . e . lack of this regulator causes growth defects under -Fe as well as hFe [3] . Notably , both the -Fe and hFe functions of HapX require the HapB/HapC/HapE CCAAT-binding complex ( CBC ) as a DNA binding platform [4] . SreA and HapX are interconnected in a feedback-loop [5]: Expression of sreA is repressed by HapX during -Fe [2] and , in turn , hapX expression is repressed by SreA under iron sufficiency/excess [1] . Moreover , HapX induces sreA expression in response to iron . Fungal iron sensing has been studied most intensively so far in the yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe [6 , 7] . Remarkably , there is little similarity with respect to transcriptional iron regulation between S . cerevisiae and A . fumigatus . Despite the fact that both , HapX and SreA are conserved in most ascomycetes , S . cerevisiae lacks classical homologs of SreA and HapX . In this yeast , adaptation to iron starvation is mainly mediated by two paralogous transcription factors , termed Aft1 and Aft2 [8–10] . Adaptation to hFe by transcriptional activation iron detoxification is mediated by the bZIP transcription factor Yap5 [11] . Nevertheless , S . cerevisiae Yap5 and HapX show similarities . Both transcription factors are essential for iron detoxification by activation of vacuolar iron deposition . Moreover , they share a highly conserved cysteine-rich region ( CRR ) that is crucial for this function and which has been shown to coordinate a [2Fe-2S] cluster in Yap5 [3 , 12] . In contrast to HapX , however , Yap5 has no function during iron starvation . S . pombe employs a homolog of SreA , termed Fep1 [13] and a regulator displaying similarity with HapX , termed Php4 [14] . Similar to HapX , Php4 acts as repressor of iron-consuming functions during iron starvation , but in contrast to HapX it is not involved in activation of iron detoxification . Taken together , S . cerevisiae , S . pombe and A . fumigatus show significant differences with regard to the employed iron-regulatory transcription factors and the molecular mechanisms of iron sensing in A . fumigatus are largely uncharacterized . In both S . cerevisiae and S . pombe , the cytosolic monothiol glutaredoxins Grx3/4 respectively Grx4 have been shown to be involved in iron sensing [15 , 16] and coordination and transport of [2Fe-2S] clusters . These proteins contain a thioredoxin ( Trx ) -like domain , for which a canonical reductase activity has been excluded [17] , and a glutaredoxin ( Grx ) domain comprising a highly conserved CGFS motif . Coordination of [2Fe-2S] clusters is performed via the cysteine residue of the CGFS motif and two glutathione residues , which leads to dimerization of these monothiol glutaredoxins [18–20] . In the current study , we characterized the role of the cytosolic monothiol glutaredoxin of A . fumigatus ( Afu2g14960 ) , designated GrxD . We demonstrate that GrxD is essential for iron sensing by the iron-responsive transcription factors HapX and SreA , particularly for signaling iron starvation conditions . The study revealed both similarities and differences to iron sensing in other fungal species .
Protein BLAST searches identified the A . fumigatus homolog , termed GrxD , of S . cerevisiae Grx3/4 and S . pombe Grx4 , respectively . Alignment of GrxD homologs demonstrated high conservation , even between distantly related species ( Figs 1 and S1 ) . Compared to the Trx-like domain , the Grx domain shows significantly higher conservation including the [2Fe-2S] cluster coordinating CGFS motif . To investigate GrxD function in A . fumigatus , we aimed to delete the grxD gene via replacement by a hygromycin resistance-conferring cassette ( hph ) ( S2 Fig ) . Several attempts were unsuccessful , indicating that grxD is an essential gene , which we proved by heterokaryon rescue [21] . In short , this technique is based on the fact that A . fumigatus cells contain multiple nuclei . The fungal transformation procedure usually targets only the genome of one nucleus leading to heterokaryosity , in our case grxD+hph- ( wt; containing grxD but lacking hph ) nuclei and grxD-hph+ ( ΔgrxD; lacking grxD but containing hph conferring hygromycin resistance ) nuclei , which was proven by Southern blot analysis ( Fig 2A ) . During conidiation , nuclei are separated since conidia contain only a single nucleus . Conidia of eight heterokaryotic transformants were able to grow under non-selective conditions but not in the presence on hygromycin ( Fig 2A ) , demonstrating the inability of ΔgrxD ( grxD-hph+ ) conidia to grow; i . e . grxD is an essential gene . Due to the lethality of grxD deletion , we generated strains , in which grxD is under the control of the xylose-inducible xylP promoter ( PxylP , [22] ) . These strains were generated without and with C-terminal tagging of GrxD with the yellow fluorescent protein derivative Venus , yielding strains PxylP:grxD and PxylP:grxDvenus , respectively ( Fig 2B ) . PxylP displays xylose concentration-dependent activation . Without xylose supplementation , activity of this promoter is very low , i . e . expression of essential genes under this promoter in A . fumigatus led to the inability to grow [23] . Although we proved that grxD is essential ( Fig 2A ) , strains PxylP:grxD and PxylP:grxDvenus were able to grow without xylose-induction on solid minimal medium ( Fig 2C ) . This indicates that very low expression is sufficient to support growth . Nevertheless , we observed growth deficiencies under iron starvation , which were ameliorated with increasing iron concentrations ( Fig 2C ) , which indicates a role of GrxD in iron homeostasis . Overexpression of grxD with and without venus-tagging decreased growth under excess iron , but not under iron starvation or iron replete conditions ( Fig 2C ) , indicating that a surplus of GrxD impedes adaptation to high iron conditions . To further analyze GrxD function , we generated A . fumigatus mutants producing PxylP-driven GrxD variants lacking either the 19 N-terminal amino acids ( strain PxylP:grxDΔ19 ) or the whole Trx domain ( PxylP:grxDvenusΔtrx , Fig 1 ) , whereby in the latter strain GrxD was C-terminally tagged with Venus ( Fig 2B ) . Under non-inducing conditions ( without xylose ) , truncation of 19 N-terminal amino acids or truncation of the complete Trx domain , respectively , blocked growth during iron starvation and iron sufficiency ( Fig 2D ) . Growth of both mutant strains was rescued by xylose supplementation , whereby the strain expressing the Trx domain lacking GrxD required higher xylose supplementation indicating lower activity . Important to note , C-terminal tagging with Venus did not affect function of GrxD , at least judged by growth ability ( Fig 2E ) . The fact that , in contrast to strains PxylP:grxD and PxylP:grxDvenus , strains PxylP:grxDΔ19 and PxylP:grxDvenusΔtrx were unable to grow in -Fe conditions under non-induced conditions indicates that truncation of the N-terminal 19 amino acids or , even more pronounced , the truncation of the Trx domain decreases activity of GrxD . This might be due to decreased protein stability or hampered function . Nevertheless , under xylose-inducing conditions , all strains were able to grow under all conditions , which indicates that in contrast to the whole GrxD protein , the Trx domain is not essential for growth , at least when overexpressed . Consequently , the Grx domain is likely essential for growth . As shown above , N-terminal truncated GrxD versions ( PxylP:grxDΔ19 and PxylP:grxDvenusΔtrx ) were not able to grow at non-inducing conditions during iron starvation or iron sufficiency ( Fig 2D ) . However , high iron supplementation partially rescued the growth of these strains at non-inducing conditions ( Fig 2D ) . These data indicate that GrxD is involved in iron homeostasis with an important role especially during iron starvation . This is in agreement with decreased growth of strains with down-regulated GrxD , without and with C-terminal Venus-tagging , under iron starvation but not iron sufficiency and iron excess ( Fig 2C ) . Occasionally , cultivation of PxylP:grxDΔ19 conidia on plates resulted in suppressor mutants . We characterized one of these mutant strains , termed PxylP:grxDΔ19sup , in more detail . In contrast to PxylP:grxDΔ19 , PxylP:grxDΔ19sup was able to grow without xylose-induction under iron starvation and iron sufficiency ( Fig 2D ) . Under 0 . 1% xylose-inducing conditions , PxylP:grxDΔ19sup displayed a similar radial growth under iron starvation but decreased growth under iron sufficiency and high iron conditions compared to PxylP:grxDΔ19 ( Fig 2D ) . These results indicated that the suppressor mutation present in this strain leads to a defect in adaptation to iron excess . Northern analysis revealed an additional sreA transcript as well as de-repression of hapX and mirB ( encoding a siderophore transporter ) during iron sufficiency in strain PxylP:grxDΔ19sup compared to PxylP:grxDΔ19 ( S3A Fig ) . These results suggested that the suppressor mutation affects the function of SreA , which has previously been shown to repress transcription of these two genes [1] . PCR amplification analyses of the sreA locus ( S3B and S3F Fig ) followed by rapid amplification of cDNA ends ( 3´-RACE ) and nucleotide sequencing ( S3C and S3D Fig ) revealed that the suppressor mutation caused a chromosomal rearrangement ( S3E Fig ) , which results in truncation of SreA within the DNA-binding region . To independently confirm the genetic interaction between grxD and sreA , the sreA gene was deleted in a PxylP:grxDΔ19 background . This mutant , PxylP:grxDΔ19/Δ sreA , displayed the same growth pattern as PxylP:grxDΔ19sup ( Fig 2D ) , which affirms that sreA loss-of-function rescues the growth defect caused by down-regulation of grxD during iron starvation and sufficiency . To analyze whether inactivation of SreA rescues growth only in response to downregulation of GrxD ( PxylP:grxDΔ19/ΔsreA at non-inducing conditions ) or also complete lack of GrxD , we aimed to delete the grxD gene in a ΔsreA background . In contrast to wt background ( see above ) , this approach was successful . Compared to wt , the ΔgrxD/ΔsreA strain displayed severely decreased radial growth under iron starvation , iron sufficiency and iron excess , but it was viable ( Fig 2F ) . SreA is the repressor of iron uptake and SreA inactivation results in increased iron acquisition [1] . Consequently , the identified genetic interaction between grxD and sreA , together with the rescue of growth of the PxylP:grxDΔ19 strain under non-inducing conditions by high iron supplementation ( Fig 2D ) , indicate that lack of GrxD results in iron shortage , possibly caused by the requirement of GrxD for sensing iron starvation . To monitor endogenous and PxylP-controlled grxD expression , we performed Northern analysis . In wt grxD transcript levels decreased with increasing iron supplementation ( Fig 3A ) . In PxylP:grxDΔ19 , grxD expression was highly induced under xylose-induced conditions and decreased below detection limit upon xylose withdrawal demonstrating functionality of PxylP-mediated conditional grxD expression ( Fig 3A ) . In agreement with wt grxD transcript levels , Western blot analysis demonstrated that Venus-tagged GrxD ( GrxDVenus ) protein levels decreased with increasing iron availability when grxD was expressed from the endogenous promoter ( strain grxDvenus; Fig 3B ) . Under control of the xylP promoter , the protein level of Venus-tagged full-length GrxD ( GrxDVenus , strain PxylP:grxDvenus ) was highly decreased under non-inducing compared to inducing conditions ( Fig 3C ) . Interestingly , hFe conditions slightly decreased the GrxDVenus protein level under xylose-inducing conditions , which indicates an influence of iron on xylP promoter activity or on the grxD transcript stability . To analyze protein levels of the GrxD variant lacking the 19 N-terminal amino acids ( GrxDΔ19 ) , we generated a strain in which C-terminally Venus-tagged GrxDΔ19 is under the control of the xylP promoter ( strain PxylP:grxDvenusΔ19 ) . This strain showed identical growth compared to the untagged version PxylP:grxDΔ19 ( S4 Fig ) . Compared to GrxDVenus , the protein levels of the Venus-tagged GrxD variants lacking the 19 N-terminal amino acids ( GrxDVenusΔ19 , strain PxylP:grxDvenusΔ19 ) or the Trx domain ( GrxDVenusΔTrx , strain PxylP:grxDvenusΔtrx ) were slightly decreased under inducing conditions . Remarkably , under steady-state , non-inducing , high iron conditions ( Fig 3C ) , truncation of the 19 N-terminal amino acid residues ( GrxVenusΔ19 ) decreased the protein level compared to GrxDVenus although not as much as truncation of the entire Trx domain ( GrxVenusΔTrx ) . Due to the use of the same promoter , these data indicate higher protein stability of GrxDVenus compared to the truncated versions . These results most likely provide the explanation for the lack of growth of strains PxylP:grxDvenusΔ19 and PxylP:grxDvenusΔtrx during iron starvation and sufficiency under non-inducing conditions ( Figs 2D and S4 ) in contrast to strain PxylP:grxDvenus ( Fig 2C ) . Subcellular localization of Venus-tagged GrxD was determined by fluorescence . To visualize the nucleus , we expressed a gene encoding histone H2A tagged with monomeric red fluorescence protein ( H2AmRFP ) in recipient strains PxylP:grxDvenus and PxylP:grxDvenusΔtrx ( yielding strains PxylP:grxDvenus/H2AmRFP and PxylP:grxDvenusΔtrx/H2AmRFP ) . Fluorescence microscopy with these strains revealed that GrxDVenus and GrxDVenusΔTrx displayed predominant nuclear localization during iron starvation but not iron sufficiency ( Fig 4 ) . During iron sufficiency , we did not observe organelle-specific accumulation of GrxDVenus . The nuclear localization indicates a regulatory role of GrxD at least during iron starvation . Noteworthy , it has been demonstrated previously that HapX also accumulates in the nucleus during iron starvation [3] . To identify GrxD-interacting proteins , A . fumigatus strains wt , PxylP:grxDvenus and PxylP:grxDvenusΔtrx were cultivated under iron starvation ( -Fe ) , sufficiency ( 0 . 03 mM Fe ) and excess ( 5 mM Fe ) and the corresponding crude cell extracts were subjected to GFP-Trap affinity purification [24] . Here , wt served as a negative control to distinguish specifically interacting proteins from false positive bound ones . Effective enrichment of GrxDVenus and GrxDVenusΔTrx proteins was validated by SDS-PAGE and silver staining as well as Western blot analysis ( S5 Fig ) . Eluates from three independent biological GFP-Trap replicates were subsequently analyzed by nLC-MS/MS . For visualization of the specific enrichment of GrxD-interacting proteins , label-free quantification ( LFQ ) abundances of the most enriched proteins identified in PxylP:grxDvenus and PxylP:grxDvenusΔtrx GFP-Trap eluates were plotted against their LFQ abundances in wt control eluates ( Fig 5 and Tables 1 and S1 ) . We identified HapX as one of the most highly enriched proteins by GrxDVenus GFP-Trap under iron limitation ( Fig 5A ) . HapX was also detected in iron sufficient and high-iron conditions ( Fig 5B and 5C ) , however , with lower abundance , most likely due to its low protein level under these conditions [3] . Inversely , SreA was preferentially co-purified under iron sufficiency and excess ( Fig 5B and 5C ) , again reflecting the expression pattern of SreA [1] . These data indicate that GrxD constitutively interacts with HapX irrespective of the cellular iron status and at least under iron sufficiency and iron excess also with SreA; possibly , GrxD interacts also constitutively with SreA—the missing detection of the interaction during iron starvation might be due to the low expression of sreA during this condition [1] . In addition , proteins that are part of the cytosolic iron-sulfur protein assembly ( CIA ) machinery , namely Nbp35 ( Afu2g15960 ) , Dre2 ( AFUB_008090 ) and Mms19 ( Afu8g05370 ) , were enriched with high abundance under standard and excess iron levels ( Fig 5B and 5C ) . The CIA machinery was investigated extensively in the model organism S . cerevisiae . These studies showed that the monothiol glutaredoxins Grx3 and Grx4 play an indispensable role for cytosolic iron-sulfur ( FeS ) cluster biogenesis . An early step in cytosolic [4Fe-4S] cluster assembly involves Nbp35 forming a hetero-tetrameric scaffold complex with Cfd1 on which a [4Fe-4S] cluster is bound transiently [25 , 26] . Dre2 belongs to the CIA electron transfer complex and is needed for formation of the [4Fe-4S] cluster on Nbp35 [27 , 28] . Mms19 is part of the CIA targeting complex consisting of Cia1 , Cia2 and Mms19 , which , together with Nar1 transfers the [4Fe-4S] cluster to target apoproteins [29 , 30] . The precise site of requirement of monothiol glutaredoxins in the cytosolic FeS protein biogenesis has not been determined yet . In yeast , Grx3/4 is required for FeS cluster assembly on Dre2 and Nar1 [19] . How GrxD is exactly involved in the CIA of A . fumigatus remains to be elucidated . Nevertheless , these data underline the specificity of the approach . Grx4 protein interaction studies in S . pombe demonstrated that the Trx domain is essential for a stable protein interaction with both the iron regulators Fep1 ( SreA ortholog ) [31] and Php4 ( HapX ortholog ) [32] . Therefore , we were interested whether the GrxD Trx domain is necessary for all of the detected GrxD protein interactions in A . fumigatus . To address this topic , we analyzed our quantitative GrxDVenus and GrxDVenusΔTrx GFP-Trap co-purification data for selected interaction partners in detail ( Fig 6 ) . The GrxD Trx domain appeared to be dispensable for GrxD-HapX complex formation irrespective of the iron supplementation ( Fig 6B ) . In contrast , the GrxD Trx domain was essential for GrxD-SreA protein interaction ( Fig 6C ) indicated by a severely decreased SreA LFQ abundance in the absence of the Trx domain . Likewise , GrxDVenusΔTrx pull-down enrichment of the CIA proteins Dre2 , Nbp35 and Mms19 was less effective ( Fig 6D–6F ) . Unexpectedly , we identified the putative copper metallothionein CmtA ( encoded by Afu4g04318 ) as an interaction partner of GrxD , preferably under iron excess conditions ( Fig 6G ) . A recent study regarding cmtA regulation and CmtA protein function in A . fumigatus [33] revealed that cmtA expression is not regulated by copper availability and that CmtA is not required for copper detoxification . Consistently , the cmtA ortholog in A . nidulans ( AN7011 ) , termed MtlA , was found to be dispensable for copper ion tolerance [34] . Our GFP-Trap pull-down results may suggest that a GrxD-CmtA complex is involved in iron detoxification and/or transport , however this hypothesis has to be verified by future experiments . Furthermore , our data suggested an interaction of GrxD with two putative BolA family proteins , Bol1 ( Afu7g01520 ) and Bol3 ( Afu6g12490 ) . The Trx domain was dispensable for GrxD-Bol1 interaction , but GrxD-Bol3 interaction was dependent on its presence ( Fig 6H and 6I ) . However , both A . fumigatus proteins contain an N-terminal mitochondrial targeting sequence , suggesting that these proteins are localized in mitochondria . In support , homologs of A . fumigatus Bol1 and Bol3 from other Aspergillus species also contain N-terminal mitochondrial targeting sequences . In agreement , fluorescence microscopy of a strain ( PgpdA:bol1venus ) expressing Bol1 C-terminally tagged with Venus ( Bol1Venus ) suggested that Bol1 is mainly localized in mitochondria ( S6 Fig ) . It has been demonstrated previously that the homologous S . cerevisiae BolA proteins Bol1 and Bol3 form complexes with mitochondrial Grx5 , which lacks a Trx domain [35] . As GrxD is localized in the cytosol and nucleus , the interaction with both mitochondrial Bol1 and Bol3 proteins in vivo appears unlikely . One possible explanation for their detected GrxD interaction is the artificial mixture of the proteins when cellular compartments are disrupted during sample preparation . A similar phenomenon has been observed in S . cerevisiae for interaction of Grx3/4 with Bol1 , respectively Bol3 [36] , which are both localized in mitochondria [35] . Nevertheless , we can neither exclude that a minor fraction of Bol1 is localized in the cytosol nor that Bol3 is exclusively or partially localized in the cytosol and that GrxD indeed interacts with these BolA-like proteins in vivo as described in other organisms [15 , 31 , 37 , 38] . To exemplary confirm GFP trap affinity purification results , we performed co-immunoprecipitation ( co-IP ) with subsequent Western blot detection ( S7 Fig ) . HapX or SreA , respectively , was immunoprecipitated and purified from PxylP:grxDvenus and PxylP:grxDvenusΔtrx whole cell lysates using rabbit α-HapX , or rabbit α-SreA antibodies covalently linked to Protein-A-Sepharose . Western blot analysis demonstrated co-IP of GrxDVenus with both HapX and SreA ( S7 Fig ) . These experiments confirmed that GrxDVenus interacts with both HapX and SreA , while truncation of the Trx domain GrxDVenusΔTrx blocks interaction with SreA but not with HapX . For in vitro co-purification experiments , A . fumigatus GrxD was fused with a C-terminal His-tag ( GrxDHis6 ) and bicistronically co-expressed in Escherichia coli with a polypeptide representing the A . fumigatus HapX C-terminus ( HapX161-491 ) that contains all four cysteine-rich regions ( CRR; Fig 7A ) . To investigate the interaction between both proteins , GrxDHis6 was enriched from crude cell extract via its His-tag using a Ni-Sepharose column . Consequently , co-purification of HapX161-491 requires binding to GrxDHis6 . After initial Ni-chelate chromatography , we observed that GrxDHis6 and HapX161-491 were co-enriched ( Fig 7B ) . The GrxD His-tag was subsequently removed by tobacco etch virus ( TEV ) protease treatment and the GrxD-HapX161-491 complex stability was further analyzed by preparative size exclusion chromatography ( SEC ) . Two major peaks appeared during SEC and their apparent molecular masses were estimated based on the elution volumes of protein calibration standards . For peak 1 , a molecular mass of 152 . 9 kDa ( Fig 7B ) approximately corresponding to a heterotetrameric complex consisting of two HapX161-491 and two GrxD subunits ( theoretical mass: 130 . 4 kDa ) was calculated . For peak 2 , a molecular mass of 27 . 7 kDa corresponding to a theoretical molecular mass of a GrxD monomer ( 29 . 75 kDa ) was determined . Additionally , UV-Vis spectra ( 250–550 nm ) were recorded for peak 1 and 2 ( Fig 7C ) . The reddish-brown color of the GrxD-HapX161-491 complex ( peak 1 ) as well as the absorption maxima at 322 and 415 nm indicated the incorporation of a [2Fe-2S] ligand , as spectra of [2Fe-2S] proteins are typically more complex than those of [4Fe-4S] proteins , which display only one characteristic peak around 400–420 nm [39] . In contrast , GrxD separated in excess from the GrxD-HapX161-491 complex ( peak 2 ) appeared colorless and displayed no absorption at 322 and 415 nm ( Fig 7C ) . We hypothesized that the reddish-brown color of the GrxD-HapX161-491 complex is mainly derived from binding of an FeS ligand by HapX161-491 CRR . This was supported by SEC purification of HapX161-491 in the absence of GrxD , which yielded a reddish-brown colored SEC fraction displaying a UV-Vis spectrum almost identical to that of the GrxD-HapX161-491 complex ( Fig 7D and 7E ) . These data strongly indicate that HapX is able to coordinate FeS cluster ( s ) without GrxD . To analyze the in vitro GrxD-HapX161-491 protein-protein interaction in more detail , two GrxDHis6 mutants were constructed , co-produced with HapX161-491 and purified from E . coli crude cell extracts . Based on the results of the in vivo co-purification experiments , the Trx domain was deleted first . Consistent with our in vivo data , removal of the GrxD Trx domain had no impact on GrxDΔTrx-HapX161-491 protein interaction in vitro ( Fig 7F ) . In a second step , GrxD cysteine ( C ) residue 191 was mutated to alanine ( A ) . GrxD C191 is part of the CGFS active site motif in the Grx domain , which is highly conserved and known to be important for iron sensing through binding of a [2Fe-2S] cluster in S . cerevisiae [19 , 40] and S . pombe [32 , 41] . In S . pombe , the CGFS site’s cysteine is required for iron-dependent Grx4-Php4 complex formation [32] . In this study , mutation of the GrxD C191 to A did not influence binding to the HapX161-491 CRR in E . coli ( Fig 7G ) . HapX harbors four CRR , which might participate in iron sensing . As reported previously [3] , CRR-A and B ( Fig 7A ) are crucial for adaptation to iron excess . In particular , the mutation of C 203 to A in CRR-A or exchange of C277 to A in CRR-B rendered A . fumigatus more susceptible to iron overload . C277 is part of the CRR-B C277GFCSDGTPCIC motif , which is reminiscent to the CGFCNDNTTCVC [2Fe-2S] cluster binding site in S . cerevisiae Yap5 [12] . To elucidate the impact of both C203 and C277 on GrxD-HapX161-491 complex formation , we targeted C203 and C277 by site-directed mutagenesis and replaced them by alanine . Neither HapX161-491 C203A exchange nor C277A substitution affected binding of the respective HapX versions to GrxD ( Fig 7H and 7I ) . In summary , we conclude that the Trx domain and residue C191 of GrxD as well as residues C203 and C277 in HapX are not required for in vitro complex formation between GrxD and HapX . As gene deletion was not possible in wt cells , we developed a protein depletion strategy to investigate the effects of GrxD deficiency . We avoided to use strain ΔgrxD/ΔsreA as it was not possible to measure effects of GrxD deficiency on SreA in this strain and as growth of this mutant was severely impaired . To study the effects of GrxD depletion on iron regulation , we employed PxylP:grxDΔ19 , which allowed to decrease grxD expression to a lethal amount without xylose induction , while growth was fully rescued with a moderate ( 0 . 1% ) concentration of xylose ( see above , Fig 2 ) . To analyze the effect of GrxD depletion on iron regulation , we performed Northern analysis of iron regulated genes during iron starvation and sufficiency . For GrxD depletion , PxylP:grxDΔ19 was grown under inducing conditions for 20 h at 25°C and subsequently grown for another 20h at 37°C without xylose to repress grxD expression . This method was used previously to investigate essential genes [23] . During iron starvation , GrxD depletion decreased transcript levels of hapX and mirB , which were upregulated during iron starvation in wt ( Fig 8A and 8B ) . On the other hand , GrxD depletion increased transcript levels of sreA ( Fig 8A and 8B ) and cccA ( Fig 8B ) , which are downregulated during iron starvation in wt . During iron sufficiency , GrxD depletion did not significantly affect transcript levels of these genes . These data emphasize that GrxD is involved in iron regulation and is important for adaptation to iron starvation rather than iron sufficiency . Repression of sreA and cccA during iron starvation has previously been shown to depend exclusively on HapX [2 , 3] . Therefore , the de-repression of these genes found upon GrxD depletion indicates that GrxD is required for signaling iron starvation to HapX . To test whether the effects on mirB are linked to SreA or HapX , we also depleted GrxD in strains lacking SreA ( strain PxylP:grxDΔ19/ΔsreA ) . It has been shown previously that sreA is downregulated in wt during iron starvation and lack of SreA results in de-repression of iron-uptake genes ( mirB , hapX ) during iron sufficiency [1] . Deletion of sreA in PxylP:grxDΔ19 increased expression of mirB upon GrxD depletion , albeit not to wt level . This indicated that GrxD is required to inactivate the repressing function of SreA under iron starvation . The absence of full induction in GrxD depleted PxylP:grxDΔ19/ΔsreA compared to the appropriate reference ( ΔsreA ) indicates that GrxD is not only required to inactivate SreA-mediated repression of mirB , but also for the induction of mirB expression , likely via activation of HapX inducing function . Interestingly , grxD was also de-repressed during iron sufficiency in a SreA deficient strain ( Fig 8A ) , suggesting that SreA is a repressor of grxD transcription during iron sufficiency . In agreement , MEME analysis [42] of grxD promoter regions of 20 different Aspergillus species identified the highly conserved motif 5´-ATCWGATAA-3´ ( S8 Fig ) , which was previously shown to be the consensus motif for DNA-binding by SreA [1] . This regulatory pattern is similar to that in S . pombe , since grx4 transcript levels are about 2-fold elevated in iron-starved cells [43] , but contrasts the situation in S . cerevisiae because grx4 is here under control of Yap5 , which activates grx4 gene expression in iron excess conditions [44] . Previously , HapX was shown to be essential for transcriptional short-term induction of iron-consuming genes [3] . To investigate whether this induction depends on GrxD , we shifted GrxD-depleted cells from iron starvation to iron sufficiency ( Fig 8B ) . Such a shift causes extensive transcriptional rearrangements including repression of iron uptake ( mainly via SreA , [1] ) and induction of iron-consuming genes ( mainly via HapX , [3] ) . Remarkably , GrxD depletion did not completely block induction of sreA and cccA in this set-up indicating independence of GrxD . To prove that this induction is not mediated by remaining GrxD protein levels upon GrxD depletion , Northern blot analysis was performed using strain ΔgrxD/ΔsreA , which lacks GrxD and SreA . The shift from iron starvation to iron sufficiency still induced cccA in this mutant , although the response was decreased compared to wt ( Fig 8C ) . cccA is exclusively regulated by HapX [3] and therefore its induction during sFe proves that GrxD is , at least partially , dispensable for HapX function during iron excess . The most likely explanation for the decreased response is the transcriptional downregulation of iron acquisition mechanisms during iron starvation in GrxD-lacking cells ( see above ) , which decreases iron uptake in the iron shift . In summary , these data indicate that GrxD is required during iron starvation conditions to activate HapX iron starvation function ( i . e . repression of iron-consuming genes and induction of iron uptake ) and to inactivate SreA function ( i . e . repression of iron uptake ) , but not for iron sensing by HapX under iron excess . FeS clusters in GrxD homologs are coordinated by C191 in the CGFS motif located in the Grx domain ( Fig 1 ) . To analyze the function of this cysteine residue in A . fumigatus iron-regulation , we overexpressed C-terminal venus-tagged grxD-variants ( targeted to the pksP locus and expressed under control of the strong constitutive PgpdA promoter of glyceraldehyde-3-phosphate dehydrogenase encoding gene ) using PxylP:grxDΔ19 as recipient strain ( Fig 9A ) . This strategy allowed for growth during induction with xylose regardless of the functionality of the pksP-targeted grxD-variant due to grxDΔ19 expression of the endogenous PxylP-controlled grxD gene . Without xylose induction , only the pksP-located version is expressed allowing phenotypical characterization of the pksP-targeted grxD variant . Overexpression of grxDC191A was unable to rescue the growth defect caused by lack of GrxD ( non-inducing conditions ) during iron starvation and iron sufficiency , demonstrating that replacement of cysteine residue 191 by alanine blocks GrxD function ( Fig 9B ) . In contrast , expression of grxDvenusC191S was able to rescue the lack of GrxD during iron sufficiency but not iron starvation ( Fig 9B ) . Similarly , serine can partially compensate for the function of this cysteine residue in the S . cerevisiae GrxD homolog [19 , 45] . Endogenous ( wt ) GrxD protein levels are highest under iron starvation ( Fig 3B ) , indicating a higher GrxD requirement under iron starvation , which might explain the lack of compensation by GrxDVenusC191S under this condition . Alternatively , C191 might be particularly important for adaptation to iron starvation . Interestingly , under xylose-inducing conditions ( leading to expression of grxDΔ19 ) overexpression of grxDC191A or grxDC191S decreased growth particularly during iron starvation indicating a dominant negative effect of these GrxD variants . As overexpression of grxDvenusC191S was partially able to compensate downregulation of grxDΔ19 , we generated a mutant strain expressing exclusively PxylP-driven grxDvenusC191S ( Fig 9A ) . Indeed , overexpression ( xylose-induction ) of grxDvenusC191S also enabled growth in this set-up in an iron supply-dependent manner: wt-like ( or even better than wt ) growth during high iron conditions , decreased growth during iron sufficiency but only poor growth during iron starvation ( Fig 9C ) , as observed above in PxylP:grxDΔ19/PgpdA:grxDvenusC191S ( Fig 9B ) . Northern analysis revealed that overexpression of either grxDvenus or grxDvenusC191S increased expression of hapX during iron starvation ( Fig 9D ) . As hapX expression is mainly regulated by SreA repression , these data indicate that overexpression of either grxDvenus or grxDvenusC191S inactivates SreA . In agreement , GrxD deficiency constitutively activated SreA ( Fig 8A ) . Remarkably , overexpression of grxDvenusC191S but not grxDvenus decreased expression of mirB during iron starvation ( Fig 9D ) . This result resembles GrxD deficiency ( Fig 8B ) and indicates that the residual function of GrxDVenusC191S is not sufficient to maintain the iron-regulatory function under iron starvation . As mirB expression requires not only inactivation of SreA ( and SreA is highly inactivated as judged by the hapX expression ) but also induction by HapX , these findings indicate that GrxDVenusC191S fails to activate HapX in contrast to GrxDVenus . In contrast to iron starvation , overexpression of grxDvenusC191S or grxDvenus had no significant effect on these genes during iron sufficiency ( Fig 9D ) . Taken together , these data underline the importance of GrxD for sensing of iron starvation . As shown previously [23] and above ( Fig 8B ) , a short-term shift from iron starvation to iron sufficiency upregulates sreA and cccA . This response was previously shown to be mediated by HapX [3] and does not require GrxD as shown here ( Fig 8B and 8C ) . Remarkably , however , this regulation was blocked by overexpression of GrxDVenus but not GrxDVenusC191S ( Fig 9D ) . As GrxD dimers are capable of [2Fe-2S] cluster coordination , these data might indicate that GrxDVenus but not GrxDVenusC191S competes with HapX for [2Fe-2S] and thereby blocks activation of the high-iron function of HapX . In agreement , a grxDvenus overexpressing strain displayed severe growth deficiencies at excess iron conditions ( Fig 2C ) . The observed difference between GrxDVenus and GrxDVenusC191S in these experiments is most likely based on the decreased [2Fe-2S] binding affinity of GrxDVenusC191S compared to GrxDVenus .
Recently , we have shown that iron sensing in A . fumigatus depends on a signal from mitochondrial ( ISC ) but not on cytosolic ( CIA ) iron-sulfur cluster biosynthesis and on glutathione biosynthesis [23] . Here we demonstrate that A . fumigatus monothiol glutaredoxin GrxD is required to activate HapX-mediated adaptation to iron starvation as well as for inactivation of SreA during iron starvation . Thereby GrxD acts as sensor for iron starvation , most likely by modulating the signal for iron availability , which is generated by ISC . GrxD homologs have previously been shown to be involved in iron sensing in the ascomycetous yeast species S . cerevisiae , S . pombe and the basidiomycetous yeast species Cryptococcus neoformans [15 , 19 , 46 , 47] . Yet , these fungal species and the filamentous ascomycete A . fumigatus display significant differences with respect to transcriptional iron regulators and the role of the GrxD homologs . S . cerevisiae employs two paralogs , Grx3/4 , which are essential for growth dependent on the genetic background [19]; in S . pombe , mutants lacking Grx4 are viable only under microaerophilic conditions [43 , 46]; in C . neoformans , deletion of the entire Grx4 gene but not truncation of the Grx domain is lethal [47] . Here we demonstrate that in A . fumigatus GrxD is essential for growth , whereby the cysteine residue in the Grx domain plays a crucial role , while the Trx domain is dispensable for growth , at least when the Grx domain is overexpressed . As shown for Grx4 in S . cerevisiae [19] , GrxD has most likely also a dual function in A . fumigatus: a regulatory role in iron sensing as well as in transport of [2Fe-2S] clusters in cellular metabolism . Moreover , Grx3/4 have been suggested to be involved in stress resistance in S . cerevisiae via affecting actin dynamics and Sir2 glutathionylation [48 , 49] . In agreement , co-IP approaches revealed physical interaction of GrxD not only with the iron regulators SreA and HapX , but also with CIA components . Likewise , physical interaction of Arabidopsis thaliana GrxD homolog GRXS17 and CIA components has been observed previously [50] . Lethality of lack of GrxD might be a synergistic effect of its dual roles . The fact that we found that lack of SreA suppresses the lethal effect of lack of GrxD and that high iron supplementation suppresses the growth defect caused by GrxD downregulation indicates however that the role in iron sensing is the major reason for its essentiality under standard conditions . Our in vivo approaches indicated that the Trx domain of GrxD is required for interaction with SreA but not HapX . In agreement , in vitro studies with recombinant proteins revealed that neither the Trx domain nor the cysteine residue in CGFS motif in the Grx domain , which is essential for the [2Fe-2S] cluster coordination , are required for physical interaction of GrxD with HapX . Moreover , cysteine residues , which have previously been shown to be essential for in vivo function of HapX under high-iron conditions [3] , were found to be dispensable for physical interaction of GrxD with HapX . The paralogous S . cerevisiae transcription factors mediating adaptation to iron starvation , Aft1/2 , are conserved exclusively in closely related Saccharomycotina and do not display any similarity to HapX or SreA . In S . cerevisiae , lack of Grx3/4 results in constitutive activation of Aft1/2 irrespective of the iron status . Thus , Grx3/4 is required for inactivation of Aft1/2 during iron sufficiency [15] , i . e . sensing of iron sufficiency . The S . cerevisiae transcription factor mediating adaption to iron excess , Yap5 shows similarities to HapX , but has no function during iron starvation [11] . This indicates that HapX homologs have evolved in a modular manner , whereby A . fumigatus HapX combines protein modules and respective functions for adaption to iron excess from S . cerevisiae Yap5 and functions for adaption to iron starvation from S . pombe Php4 ( see below ) . Similar to Yap5 , HapX contains two cysteine-rich regions ( CRR ) , which are crucial for high iron functions [3] , whereby one of these contains a perfectly conserved CGFC motif , which was shown to be essential for Yap5 function and [2Fe-2S] cluster coordination [12] . We found in the current study that recombinant HapX displays a reddish-brown color and a UV-Vis spectrum indicative of [2Fe-2S] coordination . Together with our previous observation that activation of the HapX high-iron function depends on ISC but not CIA , our data indicate that HapX senses high iron conditions via [2Fe-2S] coordination similar to Yap5 . Remarkably , [2Fe-2S] coordination by Yap5 was shown to be independent of Grx3/4 [12] . Similarly , we also observed that GrxD is dispensable for the activation of the HapX high-iron function in A . fumigatus ( Figs 8B and 8C and 10 ) . The transcription factors maintaining iron homeostasis in S . pombe are termed Fep1 and Php4 [13 , 51] . Fep1 is a homolog of SreA and shares the same function . The HapX homolog Php4 lacks a bZIP-type DNA-binding region but , similar to HapX , interacts with the Php2/Php3/Php5 CBC via its N-terminal CBC-binding domain resulting in repression of iron-consuming pathways under iron starvation [51] . However , in contrast to HapX , Php4 appears to lack a function in activation of iron acquisition during iron starvation and is not involved in adaptation to iron excess . In agreement , the CRR that are conserved and essential for high-iron functions in S . cerevisiae Yap5 and A . fumigatus HapX are not conserved in Php4 . In S . pombe , lack of Grx4 caused constitutive activation of the repressing functions of both Php4 and Fep1 [46] , i . e . it caused repression of iron acquisition during iron starvation via Fep1 and repression of iron-consuming pathways during iron sufficiency via Php4 and , therefore , deleterious effects during both iron starvation and sufficiency . This finding contrasts the situation in A . fumigatus , in which lack of GrxD caused regulatory defects only during iron starvation . Thus , GrxD appears to modulate the activity of SreA in A . fumigatus in a similar way as Grx4 affects Fep1 in S . pombe ( Fig 10 ) . In contrast to Php4 in S . pombe , however , lack of GrxD did not trigger constitutive HapX iron starvation functions . On the contrary , GrxD depletion impaired HapX mediated adaptation to iron starvation ( Fig 10 ) , which indicates significant mechanistic differences in the mode of action of the monothiol glutaredoxin in regulation of S . pombe Php4 and A . fumigatus HapX . In S . pombe , Php4 and Grx4 form a heterodimer , irrespective of the cellular iron status via the Trx domain of Grx4 [32] . During iron sufficiency Php4 and Grx4 are suggested to coordinate a [2Fe-2S] cluster with GSH as additional ligand [16] , which causes export from the nucleus to block Php4 activity . In contrast to Php4 , HapX appears to coordinate [2Fe-2S] clusters also without GrxD , similar to S . cerevisiae Yap5 ( see above ) . Unlike S . pombe Php4 , A . fumigatus HapX also has a function in high-iron conditions and therefore it is unlikely that inactivation of HapX iron-starvation functions ( repression of iron-consuming pathways , activation of iron acquisition ) involves export of HapX from the nucleus , which could explain evolution of mechanistic differences in modulation of activity of Php4 and HapX . C . neoformans employs homologs of A . fumigatus SreA and HapX , termed Cir1 and HapX , respectively [52 , 53] . In contrast to A . fumigatus SreA , however , Cir1 is also involved in adaptation to iron starvation , e . g . activation of iron acquisition . Recently , the GrxD homologue Grx4 was demonstrated to be essential for activation of Cir1 functions via physical interaction , i . e . lack of GrxD phenocopied lack of Cir1 [47] . This differs from the situation in A . fumigatus , in which lack of GrxD renders SreA constitutively active . Taken together , the role of GrxD homologs in iron sensing has been demonstrated in different fungal species . In all these species , GrxD homologs display physical interaction with the employed iron regulators . However , these transcription factors show in part significant differences in protein domains and mode of action . These differences are most likely the reason for the different regulatory consequences of lack of GrxD in the analyzed species . Moreover , GrxD homologs show different regulatory patterns in different fungal species . In S . cerevisiae , expression of the Grx3/4-encoding genes is upregulated during iron sufficiency compared to iron starvation , which is mediated by Yap5 [44] . In contrast , in S . pombe and C . neoformans , Grx4 is upregulated during iron starvation compared to iron sufficiency [43 , 47] . In these species , Grx4 is preferentially located in the nucleus . C . neoformans , Grx4 , however , shows increased nuclear localization under iron starvation compared to iron sufficiency [46 , 47] . For A . fumigatus GrxD we found a similar expression and localization pattern as in C . neoformans . Moreover , we discovered a negative feedback-loop between GrxD and SreA: GrxD is required to repress the function of SreA during iron starvation , while SreA transcriptionally represses expression of the GrxD encoding gene during iron sufficiency ( Fig 10 ) . Iron sensing by S . cerevisiae Aft1/2 and S . pombe Fep1 has been shown to involve not only a GrxD homolog but also a cytosolic BolA2-like protein , termed Fra2 . In both organisms Fra2 deficiency resembles Grx3/4 or Grx4 deficiency , i . e . a constitutive increase of iron uptake in S . cerevisiae and constitutive repression of iron uptake in S . pombe [54 , 55] . Similar to S . cerevisiae and S . pombe [56] , the genome of A . fumigatus and other Aspergillus species encodes two BolA-like proteins containing mitochondrial targeting sequences . However , in contrast to S . cerevisiae and S . pombe , Aspergillus spp . appear to lack a cytosolic BolA2-like protein ( although dual localization cannot be excluded ) indicating another possible difference in the iron sensing apparatus in these molds . An intriguing question is of course how GrxD mechanistically modulates the function of SreA and HapX . For S . pombe it has been suggested that GrxD signals iron starvation to Fep1 by removing iron , not [2Fe-2S] , bound by Fep1 [46] . Later on , it was shown that Fep1 coordinates a [2Fe-2S] cluster , not iron , by a highly conserved CRR [57] . Nevertheless , GrxD-mediated removal of [2Fe-2S] clusters bound by SreA and HapX appears to be a conceivable mode of action for signaling iron starvation . Such a model is supported by the fact that overexpression of grxD impaired adaptation to iron sufficiency , i . e it blocked short-term induction of cccA expression , which depends exclusively on HapX [3] . This effect was not seen when the [2Fe-2S] cluster coordinating cysteine residue in the CGFS motif of GrxD was replaced by a serine residue , which decreases the affinity for the [2Fe-2S] cluster [19] . These data might suggest that in this set-up GrxD competes for [2Fe-2S] clusters with HapX , which impairs iron sensing by HapX . Moreover , this cysteine to serine exchange also impaired transcriptional adaptation to iron starvation , i . e . high-affinity [2Fe-2S] binding by GrxD is crucial for sensing iron starvation . The severe growth defect of downregulation of GrxD in A . fumigatus is likely a combination of deficiencies in iron sensing and [2Fe-2S] transport . Alternative to GrxD-mediated removal of [2Fe-2S] clusters bound by SreA and HapX , GrxD might signal iron starvation in complexes with HapX and SreA by inducing conformational changes upon [2Fe-2S] cluster coordination . Thus , the cytosolic monothiol glutaredoxin GrxD is involved in iron sensing in A . fumigatus as shown previously for other fungal species . However , our studies revealed significant differences in the mode of action of GrxD and the consequences of the lack of GrxD in this mold , which underlines a remarkable plasticity in iron sensing in fungi . The virulence defect of A . fumigatus mutants lacking siderophore biosynthesis [58–60] or HapX [2] , as well as the transcriptional upregulation of iron acquisition pathways [61] in murine infection models indicate that A . fumigatus faces iron starvation in vivo . Moreover , plasma was recently shown to inhibit growth of A . fumigatus as long as transferrin was not iron saturated , i . e . , in the absence of”non-transferrin bound iron” [62] . In line with A . fumigatus facing iron starvation during growth in plasma we found that GrxD localizes to the nucleus during growth in plasma ( S9 Fig ) similar to growth during iron starvation in minimal medium ( Fig 4 ) . In contrast , supplementation of plasma with high amounts of iron blocked the predominant nuclear localization ( S9 Fig ) similar to growth under iron sufficiency in minimal medium ( Fig 4 ) . Taken together , these data implicate that GrxD plays a role in adaptation to iron starvation during infection . In this regard noteworthy , lack of the Grx domain in the GrxD ortholog renders of C . neoformans avirulent in a murine infection model [47] . Moreover , the essential role of GrxD for viability of A . fumigatus underlines the importance of iron metabolism and homeostasis .
Strains used in this study are listed in S2 Table . Oligonucleotides used in this study are listed in S3 Table . Growth assays were performed in Aspergillus minimal medium ( 1% ( w/v ) glucose , 20 mM glutamine , salt solution and iron-free trace elements according to [63] and Aspergillus complex medium ( 2% ( w/v ) glucose , 0 . 2% ( w/v ) peptone , 0 . 1% ( w/v ) yeast extract , 0 . 1% ( w/v ) casamino acids , salt solution and iron-free trace elements according to [63] . Iron ( FeSO4 ) was added separately as indicated in the respective figures . However , -Fe , +Fe and sFe stands for iron starvation ( no iron ) , 0 . 03 mM iron , and shift to 0 . 03 mM iron after precedent iron starvation , respectively . PxylP-driven genes are repressed unless xylose ( w/v ) is added to the medium , which is indicated in the respective Figures . For solid growth , the medium was solidified with 1 . 8% ( w/v ) agarose . In phase one , 108 spores of strains of interest were shaken in 50 ml minimal medium +Fe at 25°C with 0 . 1% ( w/v ) xylose ( inducing conditions to enable GrxDΔ19 production and thereby growth ) for 20 h . Germlings were centrifuged and washed once with water to remove iron and xylose before being re-suspended in 100 ml minimal medium containing no xylose . To deplete already produced GrxDΔ19 in phase two , growth was continued for 20 h at 37°C . During phase two , the growth conditions were -Fe , +Fe or sFe . Controls were treated the same way . For microscopy in minimal medium , strains were grown in well chamber slides ( Ibidi ) with 2 x 104 spores/well ( final concentration 105/ml ) for 18h at 37°C with 0 . 05% ( w/v ) xylose under iron starvation ( -Fe ) or iron sufficiency ( +Fe ) . Growth in these chamber slides was hardly sufficient to generate iron starvation after 18 h . To increase iron starvation , -Fe media contained 0 . 5 mM of the ferrous-iron chelator bathophenanthroline disulfonic acid ( BPS ) . For growth in human blood plasma , spores were inoculated in plasma without or with spiking with 0 . 1 mM iron to override iron starvation . Spore inoculation and incubation was identical to microscopy with minimal medium . Human plasma was obtained from the bloodbank of Medical University Innsbruck and treated as described previously [62] . Mycelia were examined with a spinning-disc confocal microscopic system ( Ultra VIEW VoX; PerkinElmer , Waltham , MA ) that was connected to a Zeiss AxioObserver Z1 inverted microscope ( Zeiss , Oberkochen , Germany ) . Images were acquired with Volocity software ( PerkinElmer ) with a 63x oil immersion objective with a numerical aperture of 1 . 4 . The laser wavelengths used for excitation of Venus and mRFP were 488 and 561 nm , respectively A schematic overview for the generation of all mutant strains is given in S2 Fig . To simultaneously exchange the endogenous promoter of grxD and include a Venus-tag , a plasmid containing grxD 5’-region , hph , PxylP , grxD ( including 3’-region ) and pUC19 backbone was generated . Parts of this plasmid were amplified with primers oKM11-16 and pMMHL15 [23] or A . fumigatus wt gDNA as template and finally assembled with NEBuilder ( New England Biolabs ) in pUC19 yielding plasmid pKM1 . Subsequently pKM1 was linearized with oKM26 and oKM27 to integrate the venus-tag ( amplified with oKM28 and oKM29 from phapXVENUS-hph [3] ) via seamless cloning ( NEBuilder; New England Biolabs ) yielding pKM1+venus . The insert of pKM1+venus was amplified with primers oKM11 and oKM16 and transformed into a wt recipient strain via homologous recombination . Thereby endogenous grxD was exchanged . As two possibilities for homologous recombination at the grxD locus were available ( S2 Fig ) , we received two types of transformants , PxylP:grxD and PxylP:grxDvenus . Site-directed mutagenesis ( Q5 Site-Directed Mutagenesis Kit; New England Biolabs ) was performed with pKM1+venus ( see above ) and primers oMM182 and oMM184 yielding pMMHL43 . The insert was amplified with oKM11 and oKM16 and transformed into a wt recipient strain yielding strain PxylP:grxDvenusΔtrx via homologous recombination at the grxD locus . Thereby endogenous grxD was exchanged . To integrate mRFP-tagged histone H2A driven by constitutive gpdA promoter , a plasmid containing fragment PgpdA:mRFP:H2A , a phleomycin resistance cassette ( ble ) , a pksP homologous site and pUC19 backbone was generated . Subunits of this plasmid were amplified with primers oMM189-194 and plasmid pME3173 [64] , A . fumigatus wt gDNA or pAN8-1 [65] , respectively , as template and finally assembled with NEBuilder ( New England Biolabs ) in pUC19 yielding plasmid pMMHL44 . The plasmid was linearized with BamHI and integrated into the pksP locus of recipient strains ( PxylP:grxDvenus or PxylP:grxDvenusΔtrx ) via homologous recombination at the pksP locus . This gene encodes for a polyketide synthase , which is involved in conidial pigmentation [66] . Disruption of pksP allows for fast screening of positive integrations , as ΔpksP strains produce white conidia . To delete sreA , a plasmid containing sreA 5’-region , a pyrithiamine resistance cassette ( ptrA ) , sreA 3’-region and pUC19 backbone was generated . Subunits of this plasmid were amplified with primers oMM164-169 and A . fumigatus wt gDNA or pSK275 ( syn . pME3024 [67] ) as template and finally assembled with NEBuilder ( New England Biolabs ) in pUC19 yielding plasmid pMMHL38 . The insert of pMMHL38 was amplified with oMM164 and oMM169 and transformed into a wt recipient strain . Thereby sreA was deleted via homologous recombination . venus-tagging of grxD was performed by employing CRISPR technology as described previously [68] . We used the hygromycin resistance-mediating AMA-plasmid pFC332 and grxD targeting sequence AGGCTCCTGCCAGCGCTTGA as protospacer sequence , yielding pMMHL49 . A repair template was amplified with oKM15 and oKM16 from pKM1+venus ( see above ) . The repair template and pMMHL49 were together transformed into a wt recipient strain . This procedure caused cleavage at the grxD locus by CRISPR and integration of the repair template via homologous recombination . By subsequent growth on non-selective media the CRISPR plasmid was lost yielding grxDvenus , a marker-free strain , in which endogenous grxD is tagged with venus without further manipulation of the grxD locus . The 5’-region of grxD was amplified with primers oAfgrx4-oe1 and oAfgrx4-oe2 and digested with AvrII ( fragment A ) . Truncated grxD was amplified with primers oAfgrx4-oe4 and oAfgrx4-oe5 and digested with NcoI . The PxylP sequence was liberated from plasmid pxylPp [69] by digestion with NotI and NcoI . Both , truncated grxD and PxylP were ligated via their NotI overhang , the fragment was amplified with primers oAfgrx4-oe6 and oAfgrx4-oe7 and digested with XbaI ( fragment B ) . The hygromycin resistance cassette was released from plasmid pAN7-1 by digestion with XbaI and AvrII ( fragment C ) . Fragments A , B and C were ligated via AvrII and XbaI overhangs . The resulting fragment was amplified with primers oAfgrx4-oe3 and Afgrx4-oe8 and integrated into a wt recipient strain via homologous recombination at the grxD locus yielding PxylP:grxDΔ19 . Thereby endogenous grxD was exchanged . As grxD is essential ( see Results ) growth under non-inducing conditions ( no xylose ) was inhibited . However , streaking out >108 spores on non-inducing agar plates yielded colonies . At least one of these , designated as PxylP:grxDΔ19sup , harbored a mutation suppressing the lethal effect caused by grxD deficiency . A construct containing grxD 5’-region , hph , PxylP and the 19 aa truncated version of grxD as 3’-homologous region was amplified from strain PxylP:grxDΔ19 gDNA with primers oAfgrx4-1 and oAfgrx4-oe5 and transformed into grxDvenus as recipient strain via homologous recombination . To inactivate sreA or hapX in a PxylP:grxDΔ19 background , the knockout constructs were amplified from ΔsreA or ΔhapX gDNA with primers oMM164 and oMM169 or oAfhapX-1 and oAfhapX-2 , respectively , and transformed into a PxylP:grxDΔ19 recipient strain via homologous recombination yielding strains PxylP:grxDΔ19/ΔsreA and PxylP:grxDΔ19/ΔhapX . To inactivate grxD , a plasmid containing grxD 5’-region , hph , grxD 3’-region and pUC19 backbone was generated . Subunits of this plasmid were amplified with primers oMM301-306 and A . fumigatus wt gDNA or pAN7-1 [70] as template and finally assembled with NEBuilder ( New England Biolabs ) in pUC19 yielding plasmid pMMHL61 . The insert of pMMHL61 was amplified with oMM301 and oMM306 and transformed into wt as recipient strain . This procedure yielded heterokaryotic transformants , containing two different nuclei ( grxD+hph-; wt; containing grxD but lacking hph and grxD-hph+; ΔgrxD; lacking grxD but containing hph ) as described in Results . The amplified cassette was also transformed into ΔsreA as recipient strain . Thereby grxD was deleted via homologous recombination . To rescue grxD deficiency , a plasmid was generated containing pksP and PgpdA:grxD:venus in backbone PgpdA-lacZ-trpCT-pJET1 . 2 [71] . The pksP fragment was amplified with oAf-pksP1-f and oAf-pksP2-r and integrated into the HindIII site of PgpdA-lacZ-trpCT-pJET1 . 2 yielding pMMHL6 . Subsequently , pMMHL6 was partially amplified with oMM156_HL6fwd and oMM157_HL6rev and assembled with grxD:venus amplified from pKM1+venus with primers oMM158_grxDfwd and oMM159_venus_rev using NEBuilder ( New England Biolabs ) . The resulting plasmid pMMHL37 was used for site-directed mutagenesis ( Q5 Site-Directed Mutagenesis Kit; New England Biolabs ) with primers oMM313 and oMM314 or oMM314 and oMM315 to generate pMMHL63 or pMMHL64 , respectively . pMMHL37 , pMMHL63 and pMMHL64 were linearized with FseI and transformed into PxylP:grxDΔ19 as recipient strain to obtain strains PxylP:grxDΔ19/PgpdA:grxDvenus , PxylP:grxDΔ19/PgpdA:grxDvenusC191A and PxylP:grxDΔ19/PgpdA:grxDvenusC191S via homologous recombination in locus pksP . To exchange endogenous grxD by a PxylP-driven grxD version in which cysteine 191 is replaced by serine , pKM1+venus was used for site directed mutagenesis ( Q5 Site-Directed Mutagenesis Kit; New England Biolabs ) with primers oMM313 and oMM314 yielding pMMHL65 . The insert was amplified with oKM11 and oKM16 and transformed into a wt recipient strain yielding strain PxylP:grxDvenusC191S via homologous recombination in locus grxD . Thereby endogenous grxD was exchanged . To constitutively express venus tagged bol1 from the pksP locus , a plasmid was generated consisting of PgpdA-driven bol1 followed by venus assembled in pMMHL37 as backbone . Therefore , bol1 was amplified with primers oMM358 and oMM359 from A . fumigatus wt gDNA and assembled with linearized pMMHL37 ( linearized with primers oMM356 and oMM357 ) in a NEBuilder ( New England Biolabs ) reaction yielding the final plasmid pMMHL83 . This plasmid was subsequently linearized with FseI and integrated into locus pksP via homologous recombination . RNA was isolated using TRI Reagents ( Sigma ) according to the manufacturer’s manual . 10 μg of RNA was used for electrophoresis on 2 . 2 M formaldehyde agarose gels and subsequently blotted onto Amersham Hybond-N Membranes ( ThermoFisher ) . Transcripts of interest were detected with DIG-labeled probes amplified by PCR . DNA was isolated by PCI extraction and isopropanol precipitation . To confirm the gene-specific restriction pattern of the genetic manipulations , DNA was digested with restriction enzymes specific for the respective gene . The resulting restriction fragments were separated on an agarose gel and transferred to Amersham Hybond-N Membranes ( ThermoFisher ) by capillary blotting with NaOH . Signals for correct integration were detected with DIG-labeled probes amplified by PCR . Rabbits were immunized with polypeptides corresponding to the amino acid residues of HapX161-491 and SreA308-546 . Sequences were PCR-amplified as NdeI-NotI fragments from cDNA and inserted into a pET-21b ( + ) vector ( Novagen ) to obtain polypeptides with a C-terminal 6x-His tag . The resulting plasmids were introduced into E . coli Rosetta BL21 cells ( Novagen ) , designed to enhance the expression of eukaryotic proteins that contain codons rarely used in E . coli . Expression was induced for 4 h at 37°C with 0 . 1 mM isopropyl-β-D-thiogalactopyranoside ( IPTG ) . Proteins were purified from cleared lysates by incubation , 2 h at 4°C , with 0 . 5 ml of Ni-NTA Agarose Resin ( Qiagen ) . Beads were washed repeatedly with phosphate buffer saline ( PBS ) containing 75 mM imidazole followed by PBS with 100 mM imidazole before proteins were eluted with 500 mM imidazole . Imidazole was removed by extensive dialysis against PBS . Protein material was lyophilized and used to immunize rabbits ( by Davids Biotechnologie GmbH , Regensburg , Germany ) . The specificity of the obtained antibodies was tested by Western analysis ( S10 Fig ) . Proteins were extracted using a reported procedure [72] involving solubilization from lyophilized mycelial biomass with NaOH , followed by their precipitation with trichloroacetic acid ( TCA ) . Aliquots were resolved in 10–12% ( w/v ) SDS-polycrylamide gels and transferred to nitrocellulose membranes . Western blots were reacted with rabbit α-HapX or rabbit α-SreA antiserum ( 1:20 , 000 ) , mouse α-GFP antibody ( 1:10 , 000; Roche , 11814460001 ) mouse α‐Tub antibody ( 1:10 , 000; Sigma , T6199 ) as primary antibodies and with peroxidase coupled antibodies as secondary antibodies ( 1:10 , 000; anti-Rabbit; Sigma , A1949 or 1:10 , 000; anti-Mouse; Sigma , A4416 ) . Proteins were detected using Amersham Biosciences ECL . Covalent coupling of rabbit α-HapX respectively rabbit α-SreA antibodies ( antiserum ) to Protein-A-Sepharose ( GE Healthcare ) was performed according to [73] . For the negative control IgGs contained in preserum were covalently linked to Protein-A-Sepharose . In short: 1 ml of Protein-A-Sepharose slurry ( 50% ) was mixed with 0 . 5 ml ( anti ) serum and treated with 20 mM dimethylpimelidate in 0 . 2 M Na-tetraborate . The reaction was stopped with 0 . 2 M ethanolamine . Immunoprecipitation assays were performed according to [74] . Mycelia were grown for 16 h in minimal medium containing 0 . 1% xylose and no iron supplementation for HapX immunoprecipitation , or 0 . 03 mM iron for SreA immunoprecipitation . For protein extracts , 40 mg of mycelia were grinded and dissolved in 1 ml protein extraction buffer containing 20 mM Tris-HCl pH 8 , 110 mM KCl , 10% ( v/v ) glycerol , 0 . 1% ( v/v ) Triton X-100 , 1μl BitNuclease ( Biotool ) and protease inhibitor ( cOmplete ULTRA EDTA-free , Roche ) . Extracts were mixed with 50 μl of covalently linked rabbit α-HapX , rabbit α-SreA or rabbit preserum beads and incubated for 3h at 4°C in a rotating wheel . Subsequently the beads were washed three times ( 10 min at 4°C in a rotating wheel ) with chilled protein extraction buffer and increasing salt concentrations ( 110 mM , 500 mM and 750 mM KCl ) . Bound proteins were eluted in 40 μl of Laemmli sample buffer at 95°C . Twenty microliters of aliquots were resolved in 10% SDS-polyacrylamide gels and transferred to nitrocellulose for Venus detection . Venus tagged GrxD or GrxDΔTrx was detected with mouse α-GFP ( 1:10 , 000; Roche , 11814460001 ) and α‐Tubulin was detected with mouse α‐Tub ( 1:10 , 000; Sigma , T6199 ) as primary antibody and with a peroxidase-coupled secondary antibody ( 1:10 , 000; anti-Mouse IgG; Sigma , A4416 ) . HapX respectively SreA were detected with rabbit α-HapX or rabbit α-SreA antisera ( 1:20 , 000 ) . To avoid the detection of rabbit IgGs , which were used for the co-IP , a conformation specific anti-Rabbit IgG antibody ( 1:1000; Cell Signaling Technology , L27A9 ) was used in combination with a peroxidase-coupled anti-Mouse IgG secondary antibody ( 1:10 , 000; Sigma , A4416 ) . For the detection Amersham Biosciences ECL was used . 3’ RACE was performed using FirstChoice RLM-RACE Kit ( ThermoFisher ) . Total RNA from PxylP:grxDΔ19sup was reverse transcribed with the oligo-dT containing primer 3' RACE Adapter . The resulting cDNA was used for Touchdown PCR with sreA ( 5’-UTR ) -specific forward primer oKM31 and adapter-specific reverse primer 3’ RACE Outer Primer . To increase specificity , the resulting PCR product ( s ) were amplified in a second PCR with nested primers oKM30 and 3’ RACE Inner Primer . This procedure yielded a fragment ( ~900bp ) which was isolated and sequenced ( S3 Fig ) . A . fumigatus mycelia were harvested in Stop buffer [75] at 4°C after growth for 22 h and freeze-dried . Protein extraction was performed according to a modified procedure from [75] using HK buffer for total protein extraction . All steps were carried out at 4°C in the cold room . In short , 100 mg of mycelium powder was dissolved in 1 ml HK buffer , centrifuged twice at 20 , 187 x g for 15 min and 500 μl of the supernatant was incubated with GFP-Trap agarose beads ( ChromoTek ) for 1 h . The beads were washed twice in HK buffer without IGPAL , twice in wash buffer ( 25 mM Tris/HCl pH 7 . 5 , 500 mM NaCl , 5 mM EDTA and 15 mM EGTA ) and once in ultrapure water . Proteins were eluted in 10% ( v/v ) acetonitrile and 5% ( v/v ) acetic acid and used for nLC-MS/MS measurement , Western blot detection and silver staining . LC-MS/MS analysis was carried out on an Ultimate 3000 nano ( n ) RSLC system coupled to a QExactive Plus mass spectrometer ( both Thermo Fisher Scientific , Waltham , MA , USA ) . Peptides were trapped for 5 min on an Acclaim Pep Map 100 column ( 2 cm x 75 μm , 3 μm ) at 5 μl/min followed by gradient elution separation on an Acclaim Pep Map RSLC column ( 50 cm x 75 μm , 2μm ) . Eluent A ( 0 . 1% ( v/v ) formic acid in water ) was mixed with eluent B ( 0 . 1% ( v/v ) formic acid in 90/10 acetonitrile/water ) as follows: 0 min at 4% B , 6 min at 6% B , 14 min at 10% B , 20 min at 14% B , 35 min at 20% B , 42 min at 26% B , 46 min at 32% B , 52 min at 42% B , 55 min at 50% B , 58min at 65% B , 60–64 . 9 min at 96% B , 65–90 min at 4% B . Positively charged ions were generated at 2 . 2 kV using a stainless steel emitter and a Nanospray Flex Ion Source ( Thermo Fisher Scientific ) . The QExactive Plus was operated in Full MS / data-dependent MS2 ( Top10 ) mode . Precursor ions were monitored at m/z 300–1500 at a resolution of 70 , 000 FWHM ( full width at half maximum ) using a maximum injection time ( ITmax ) of 120 ms and an AGC ( automatic gain control ) target of 1e6 . Precursor ions with a charge state of z = 2–5 were filtered at an isolation width of m/z 1 . 6 amu for HCD fragmentation at 30% normalized collision energy ( NCE ) . MS2 ions were scanned at 17 , 500 FWHM ( ITmax = 120 ms , AGC = 2e5 ) . Dynamic exclusion of precursor ions was set to 20 s . The LC-MS/MS instrument was controlled by QExactive Plus Tune 2 . 9 and Xcalibur 3 . 0 with DCMS Link . Tandem mass spectra were searched against the Aspergillus Genome Database ( AspGD ) of Aspergillus fumigatus Af293 ( http://www . aspergillusgenome . org/download/sequence/A_fumigatus_ Af293/current/A_fumigatus_Af293_current_orf_trans_all . fasta . gz; 2018/09/18 ) and the protein sequence of Dre2 ( AFUB_008090; the Dre2 ortholog is not present in the Af293 gene annotation ) as well as further modified protein sequences ( e . g . Venus-tag ) using Proteome Discoverer ( PD ) 2 . 2 ( Thermo ) and the algorithms of Sequest HT ( version of PD2 . 2 ) and MS Amanda 2 . 0 . Two missed cleavages were allowed for the tryptic digestion . The precursor mass tolerance was set to 10 ppm and the fragment mass tolerance was set to 0 . 02 Da . Modifications were defined as dynamic oxidation of Met , acetylation of Ser , phosphorylation of Ser , Thr , and Tyr and ubiquitination ( GG ) of Lys as well as static Cys carbamidomethylation . At least two peptides per protein and a strict false discovery rate ( FDR ) < 1% were required for positive protein hits . The Percolator node of PD2 . 2 and a reverse decoy database was used for q-value validation of spectral matches . Only rank 1 proteins and peptides of the top scored proteins were counted . The Minora algorithm of PD2 . 2 was applied for relative label-free quantification . GFP-Trap eluates from wt A . fumigatus mycelial extracts were used for quantification of nonspecifically co-purified proteins . Proteins were separated by SDS-PAGE using NuPAGE 4–12% ( w/v ) Bis-Tris gradient gels ( Invitrogen ) . Silver staining was performed using the SilverQuest Silver Staining Kit ( Invitrogen ) according to the manufacturer’s protocol . For Western detection , proteins were transferred onto a PVDF membrane using the iBlot 2 dry blotting system ( Invitrogen ) . The membrane was blocked in 3% ( w/v ) bovine serum albumin ( BSA ) dissolved in 1x PBST ( 137 mM NaCl , 2 . 7 mM KCl , 10 mM Na2HPO4 , 2 mM KH2PO4 , 0 . 05% ( v/v ) Tween 20 ) . As primary antibody rabbit α-GFP ( abcam , ab290 ) was used , followed by secondary antibody HRP-conjugated anti-Rabbit IgG ( ICL ) incubation . The membrane was developed using the 1-Step Ultra TMB-Blotting chromogenic substrate ( Thermo Scientific ) . For individual expression and protein purification , synthetic genes coding for full-length GrxD and HapX amino acids 161–491 ( cysteine-rich C-terminus ) were cloned into the NdeI and BamHI sites of the pET-MCN vector pnEA/vH [76] producing C-terminally His6-tagged GrxD ( pnEA/vH-GrxD ) and HapX161-491 ( pnEA/vH-HapX161-491 ) fused to a TEV cleavage site . For co-expression , the synthetic gene coding for HapX161-491 was initially cloned into the NdeI and BamHI sites of the pET-MCN vector pnCS producing untagged HapX161-491 ( pnCS-HapX161-491 ) . Subsequently , the BglII/XbaI fragment from pnCS-HapX161-491 was subcloned into the BglII and SpeI sites of pnEA/vH-GrxD generating a bicistronic expression cassette . Site-directed mutagenesis was performed with the QuikChange II site-directed mutagenesis kit ( Agilent ) according to the manufacturer’s protocol . Primers used for mutagenesis are listed in S3 Table . E . coli BL21 ( DE3 ) cells ( New England Biolabs ) were transformed with the respective plasmid for autoinduction in Overnight Express Instant TB medium ( Novagen ) . Wet biomass was harvested by centrifugation ( 10 , 543 x g ) and the cell paste was stored at -80°C . Frozen bacterial cells were resuspended in lysis buffer ( 50 mM HEPES pH 8 . 0 , 300 mM NaCl , 2 mM glutathione , 10 mM imidazole , 1 mM AEBSF ) and disrupted at 1000 bar using a high-pressure homogenizer ( Avestin Emulsiflex C5 ) . Cell debris were removed by centrifugation ( 48 , 384 x g ) , the pH was adjusted to 8 . 0 and the supernatant clarified by filtration through a 1 . 2 μm membrane . His6-tagged proteins were then purified by Ni-chelate affinity chromatography using a 20 ml Ni-Sepharose FF column ( GE Healthcare ) and proteins were eluted with 500 mM imidazole . Fractions containing either HapX161-491-His6 or the HapX161-491/GrxD-His6 complex were digested with TEV protease for 4 h at room temperature and loaded onto a Superdex 200 prep grade 26/60 size exclusion chromatography column ( GE Healthcare ) that was equilibrated with 25 mM HEPES pH 7 . 5 , 150 mM NaCl , 2 mM glutathione . UV-Vis absorption spectra were recorded in the range from 250 to 550 nm with a JASCO V-630 spectrophotometer .
|
Aspergillus fumigatus is a ubiquitous saprophytic mold and the major causative pathogen causing life-threatening aspergillosis . To improve therapy , there is an urgent need for a better understanding of the fungal physiology . We have previously shown that adaptation to iron starvation is an essential virulence attribute of A . fumigatus . In the present study , we characterized the mechanism employed by A . fumigatus to sense the cellular iron status , which is essential for iron homeostasis . We demonstrate that the transcription factors SreA and HapX , which coordinate iron acquisition , iron consumption and iron detoxification require physical interaction with the monothiol glutaredoxin GrxD to sense iron starvation . Moreover , we show that there is a GrxD-independent mechanism for sensing excess of iron .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] |
[
"aspergillus",
"fumigatus",
"medicine",
"and",
"health",
"sciences",
"protein",
"interactions",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"aspergillus",
"gene",
"regulation",
"pathogens",
"regulatory",
"proteins",
"microbiology",
"dna-binding",
"proteins",
"carbohydrates",
"organic",
"compounds",
"fungal",
"molds",
"xylose",
"fungi",
"model",
"organisms",
"experimental",
"organism",
"systems",
"transcription",
"factors",
"amino",
"acids",
"molecular",
"biology",
"techniques",
"fungal",
"pathogens",
"schizosaccharomyces",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"cysteine",
"mycology",
"animal",
"studies",
"proteins",
"medical",
"microbiology",
"hyperexpression",
"techniques",
"gene",
"expression",
"microbial",
"pathogens",
"schizosaccharomyces",
"pombe",
"chemistry",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"yeast",
"sulfur",
"containing",
"amino",
"acids",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"eukaryota",
"organic",
"chemistry",
"monosaccharides",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"physical",
"sciences",
"saccharomyces",
"cerevisiae",
"organisms"
] |
2019
|
The monothiol glutaredoxin GrxD is essential for sensing iron starvation in Aspergillus fumigatus
|
Partial duplication of genetic material is prevalent in eukaryotes and provides potential for evolution of new traits . Prokaryotes , which are generally haploid in nature , can evolve new genes by partial chromosome duplication , known as merodiploidy . Little is known about merodiploid formation during genetic exchange processes , although merodiploids have been serendipitously observed in early studies of bacterial transformation . Natural bacterial transformation involves internalization of exogenous donor DNA and its subsequent integration into the recipient genome by homology . It contributes to the remarkable plasticity of the human pathogen Streptococcus pneumoniae through intra and interspecies genetic exchange . We report that lethal cassette transformation produced merodiploids possessing both intact and cassette-inactivated copies of the essential target gene , bordered by repeats ( R ) corresponding to incomplete copies of IS861 . We show that merodiploidy is transiently stimulated by transformation , and only requires uptake of a ∼3-kb DNA fragment partly repeated in the chromosome . We propose and validate a model for merodiploid formation , providing evidence that tandem-duplication ( TD ) formation involves unequal crossing-over resulting from alternative pairing and interchromatid integration of R . This unequal crossing-over produces a chromosome dimer , resolution of which generates a chromosome with the TD and an abortive chromosome lacking the duplicated region . We document occurrence of TDs ranging from ∼100 to ∼900 kb in size at various chromosomal locations , including by self-transformation ( transformation with recipient chromosomal DNA ) . We show that self-transformation produces a population containing many different merodiploid cells . Merodiploidy provides opportunities for evolution of new genetic traits via alteration of duplicated genes , unrestricted by functional selective pressure . Transient stimulation of a varied population of merodiploids by transformation , which can be triggered by stresses such as antibiotic treatment in S . pneumoniae , reinforces the plasticity potential of this bacterium and transformable species generally .
Partial duplications of genetic material provide potential for evolution of new traits in all kingdoms of life , with duplicated material able to evolve in the absence of functional selective pressure [1] . Such duplications are prevalent in eukaryotes , and were recently shown to be associated with susceptibility to diseases in the human population [2] . Duplications are also thought to provide genetic novelty in plants and yeast [3] , [4] . Although diploidy favors gene evolution , prokaryotes are generally haploid in nature , possessing a single copy of most genetic material . Despite this , partial chromosomal duplications , known as merodiploids , allow prokaryotes to evolve new genes . Historically , merodiploidy in bacteria was first suggested in studies of binary encapsulation created by genetic transformation in Haemophilus influenzae [5] and Streptococcus pneumoniae [6] , [7] . Subsequent studies have identified bacterial merodiploids , with various duplications uncovered in S . pneumoniae [8] , [9] , Bacillus subtilis [10] , Escherichia coli [11] and Salmonella spp . [12] , [13] . However , these studies frequently involved specific markers or chromosomal regions , and the structure and mechanisms of formation of these merodiploids have remained elusive . Extensive analysis conducted in Salmonella enterica led to the conclusion that in unselected laboratory cultures 0 . 005–3% of cells possess duplication of a specific gene [14] , with the trade-off between their high formation rate and genetic instability resulting in a steady state after ∼30 generations [15] . However , while it is generally accepted that natural merodiploids occur at relatively high rates in bacteria [16] , similar detailed information is not available for any other species . The most common form of merodiploidy is tandem duplication ( TD ) , where duplicated regions remain adjacent in the chromosome . TDs are generally thought to form spontaneously by homologous recombination between direct repeat sequences such as insertion sequences ( IS ) or rRNA sequences [17]–[19] , an idea supported by the fact that mutation of the recombinase RecA abrogates >90% of TD formation in S . enterica [15] . Such homologous recombination events may involve unequal crossing over between different repeat regions of sister chromatids during replication , resulting in production of a TD with a hybrid repeat sequence separating duplicated regions [16] . However , in S . pneumoniae the site of spontaneous duplication within the capsule locus appeared to be random and sequences flanking the different duplications were unique [20] . A recent study in S . enterica also suggested that RecA-driven recombination may not be the sole mechanism for TD formation . Authors showed that most TDs formed on a replicative plasmid were dependent on an active transposase and conjugative apparatus , and suggested single-strand annealing between transposase-nicked sister chromosomes as an alternative mechanism of TD formation [21] . There is little information regarding TD formation during genetic transfer processes such as genetic transformation , conjugation and transduction . TDs formed by phage transduction have previously been reported but formation was dependent on transduction with non-natural junctions , defined as those not naturally present in the host genome , and therefore somewhat artificial [11] , [12] , [22] . Similarly , non-natural junctions were used to document formation of duplications during transformation in B . subtilis [23] . In the human pathogen S . pneumoniae , merodiploids spontaneously generated by transformation have been identified , with cells expressing two polysaccharide capsules isolated [6] , [7] . However , the corresponding merodiploid structures have remained undefined and the mechanisms for their formation unknown . The process of transformation involves internalization of single-stranded ( ss ) DNA fragments created from an exogenous double-stranded ( ds ) donor , and their integration into the recipient chromosome by homology . Here , we demonstrate that transformation of S . pneumoniae with homologous DNA generates TDs . Formation of these TDs does not require active transposases and is not dependent on non-natural donor junctions but only repeats present in both donor and host DNA . We establish that merodiploid formation involves alternative pairing of partly repeated ( R ) exogenous sequences resulting in interchromatid integration of internalized R ssDNA . These merodiploids produced by a genetic cross represent true merozygotes and occur throughout the genome . Transformation thus modulates the frequency of merodiploids in a pneumococcal population . These observations reveal a new , non-conservative facet of transformation in the major human pathogen S . pneumoniae , reinforcing the view that this species exhibits a great plasticity potential [24] . We discuss both the likelihood that the proposed model , relying on fundamental homologous recombination steps , applies to other transformable species , and the evolutionary potential offered by transformation-generated merodiploidy .
During study of the essential pneumococcal gene codY , we identified two independent suppressing mutations allowing survival of otherwise lethal codY::trim mutant cells in the encapsulated strain D39 [25] . We attempted to recreate this genotype in laboratory strain R1502 by transformation with chromosomal DNA possessing the codY::trim cassette and the suppressing mutations . In reasonable agreement with our previous report [25] , trimethoprim resistant ( TrimR ) transformants appeared with a frequency of ∼0 . 0020 relative to the rpsL41 reference marker ( conferring resistance to streptomycin , SmR ) . However , a selected TrimR transformant ( R2597 ) possessed neither suppressing mutation , but harboured both wildtype and codY::trim loci ( Figure S1 ) . This suggested that duplication of the codY region allowed R2597 to tolerate inactivation of one copy of codY . To identify the extent of the duplication , we sequenced the entire R2597 genome . Results showed that R2597 possesses a large duplication ( Figure 1A ) , bordered by truncated IS861 sequences ( Figure 1B and Figure S2A ) . This 107 . 4 kb duplication includes the codY region ( Figure 1C ) . Sequence analysis revealed that total coverage of the duplicated region was lower than twice that of a strain lacking the duplication ( Figure S2B ) and trim sequence had lower coverage than codY sequence ( Figure 1D ) indicating that the codY::trim duplication was under-represented . This suggested instability of the duplicated material , which would be expected from a TD in the chromosome ( Figure 2A ) or a circular 107 . 4 kb extrachromosomal element ( pop-out; Figure S3AB ) . Both of these should harbor an identical new junction ( E-R2/1-A; Figure 2A and Figure S3AB ) , which was detected by PCR of R2597 ( Figure 2A ) , sequencing of which revealed a chimeric IS861 made of parts of upstream ( R1 ) and downstream ( R2 ) IS861 sequences ( Figure 2B ) . To distinguish between a chromosomal TD and an extrachromosomal element , we analyzed R2597 by pulse-field gel electrophoresis ( PFGE ) , and hybridization with codY and trim probes , as different restriction maps were predicted ( Table S1 ) . A ∼133-kb codY-positive SmaI fragment was detected in R2597 ( Figure 2C ) . This band was predicted only for a 107 . 4 kb TD with a codY::trim/codY+ arrangement; trim-specific fragments predicted for this arrangement were also observed ( Figure S3C ) . However , R2597 contained in addition a wildtype-like ( 224 . 6-kb ) codY-positive SmaI fragment ( Figure 2C ) . Together with the low band intensity of the ∼133-kb codY-positive fragment ( Figure 2C ) , this revealed a mixed population with both non-TD ( i . e . , wildtype-like ) and TD cells present in R2597 culture . Most likely , growth in absence of Trim prior to analyses resulted in partial loss of the duplication , explaining the under-representation observed in both genome sequencing and PFGE . We conclude that an unstable TD of 107 . 4 kb was created in R2597 , rendering it merodiploid for codY and allowing mutation of one copy of the gene , and subsequent survival of otherwise lethal recombinants during TrimR selection . R2597 was isolated following transformation with chromosomal DNA ( containing codY::trim ) . To determine whether a sequence in the chromosomal DNA could stimulate merodiploid formation , we transformed cells with codY::trim ( conferring trimethoprim resistance ) or codY::spc ( conferring spectinomycin resistance ) PCR fragments in the presence or absence of isogenic chromosomal DNA . Co-transformation of chromosomal DNA stimulated formation of merodiploids between 6 and 8-fold ( Figure 3A ) . PCR analysis of 10 clones recovered from the codY::trim transformation or the codY::trim plus R800 DNA transformation showed that both codY and trim sequences were present in all clones ( Figure 3B ) . Out of 10 clones tested ( 5 from each transformation ) , 9 possessed the TD junction ( Figure 3C ) . PFGE and hybridization analysis , and sequencing of the junction of clone 11 ( R3023 ) confirmed the presence of a 107 . 4 kb TD but with a codY+/codY::trim arrangement instead of the opposite arrangement observed with R2597 ( Figure 3DE , Figure S4A and S4C , and Table S1 ) . Analysis of clone 1 ( R3022 ) , lacking the TD junction , uncovered a small duplication of the codY locus with a codY+/codY::trim arrangement lacking flanking repeats ( duplication of 3 , 276 bp , 1 , 422 , 105–1 , 425 , 381 on R6 genome , Figure S4B–E ) . Although the mechanisms of formation of such a duplication remain unclear , similar small duplications lacking flanking repeats have previously been documented in S . pneumoniae [20] . We confirmed formation of merodiploids independently of competence in pneumococcal cells by detecting the TD junction by PCR in a strain unable to develop competence ( R1501; unpublished observations ) . These results suggested firstly that merodiploids occur at basal levels in a S . pneumoniae population , accounting for the appearance of transformants with only codY::trim or codY::spc PCR fragments as donor , and secondly that transformation transiently stimulated their formation . However , the alternative possibility that transformation was facilitating , by some ill-defined mechanism , the emergence of pre-existing merodiploids could not be formally excluded . Transformation-induced merodiploidy was investigated further with the aim of distinguishing between the two possibilities , duplication made by transformation or simple trapping of a pre-existing duplication . Since the identified TDs were flanked by IS861 repeat sequences ( Figure 2A ) , and separated by a hybrid IS861 sequence ( Figure 2B ) , we hypothesized that an alternative pairing event between donor DNA and host chromosome , followed by an unequal crossing over , could generate merodiploids during transformation . This would be dependent on a donor fragment containing R as well as associated non-repeated flanking sequence ( NR-f ) . To test this , we amplified R-NRf PCR fragments representing R1-A and R2-Z ( Figure 2A ) and co-transformed cells with codY::trim and R1-A or R2-Z . Co-transformation increased the frequency of codY::trim ‘lethal cassette’ transformants up to 8-fold ( Figure 4A ) compared to transformation with codY::trim alone . All tested clones possessed the same TD , and sequencing of the junction of three of them identified three distinct R2/1 recombination junctions ( Figure 4B ) . Identical transformation with similarly-sized PCR fragments lacking repeat sequences , AB , had no effect on frequency of lethal cassette transformants ( Figure 4A ) . Furthermore , donor fragments containing R1 or R2 alone ( i . e . , without NRf ) also had no effect ( Figure 4C ) , showing that the R-NRf ‘hybrid’ structure of the donor fragment is key to merodiploid formation . Based on these observations , we elaborated a general model for merodiploid formation by transformation ( Figure 5A ) , where alternative pairing of R-NRf donor fragments ( red ) leads to interchromatid integration , bridging the neosynthesized ( blue ) and parental ( black ) chromatids of a partially replicated chromosome . Alternative pairing is crucial to our model , occurring here by initial pairing of donor R1 with R2 in the host chromosome , displacing the neosynthesized strand , followed by pairing of flanking A with A on the sister chromatid ( Figure 5A ) . Subsequent restoration of sister chromatid continuity ( Figure 5B ) results in unequal crossing-over , creating a chromosome dimer . Resolution of this dimer generates a merodiploid chromosome ( possessing two copies of codY ) and an abortive chromosome lacking codY ( Figure 5D ) . Our results showed that R1-A and R2-Z were equally efficient at generating merodiploids ( Figure 4A ) . We suggest that this is because R1-A can bridge chromatids associating loci 2 and 3 and R2-Z can bridge loci 1 and 4 ( Figure 5A ) , the resulting recombinant chromosomes being indistinguishable . The break in the complementary chromatid resulting from interchromatid integration of R1-A can be repaired spontaneously through invasion of the sister chromatid by the displaced R2 neosynthesized strand ( blue ) ( Figure 5B ) . We also envisioned an alternative donor DNA-directed break repair mechanism involving interchromatid integration of donor R2-Z ( Figure 5C ) . R2-Z fragment integration would require displacement of both the parental R1 strand ( creating R1/2 ) and neosynthesized Z strand , bridging the gap between chromatids . Consistent with this model , transformation with both R1-A and R2-Z donor fragments increased the frequency of TrimR transformants by up to 10-fold , an increased efficiency compared to either fragment alone ( Figure 4A ) . To determine whether observed cooperativity of donor fragments was dependent on polarity , we co-transformed cells with combinations of four donor fragments ( R1-A , R2-Z , X-R1 , E-R2; Figures 2A and 4D ) . Results showed that only donor fragments of the same polarity are able to cooperate to restore complementary strand integrity during merodiploid formation ( Figure 4E ) . Thus , restoration of complementary strand integrity can occur either spontaneously or via interchromatid integration of an exogenous fragment , allowing chromosome dimer creation , and subsequent merodiploid formation ( Figure 5D ) . A predicted consequence of the repair of the complementary chromatid is the formation of an abortive chromosome , subsequent to the creation of the X-R1/2-Z junction ( Figure 5E ) . We were able to detect this abortive chromosome junction by specific PCR on unselected cultures 30 minutes after transformation with R-NRf fragments ( Figure 5E ) , showing that this junction is indeed produced . Analysis of the junction sequence produced by PCR on R1-A-transformed culture showed a long region of mixed sequence ( Figure 5E ) , suggesting that a number of different alternative pairing events had occurred within the population of merodiploids produced . While sequencing of junctions provided evidence for alternative pairing of R segments ( Figures 2B , 3E and 4B ) , the overall frequency of such events in the transformed population remained unknown . To further validate our model and obtain an estimate of this frequency , we took advantage of a previous observation that the generalized mismatch repair system ( Hex ) of S . pneumoniae can be saturated by excess mismatches [26] , [27] . Hex operates on transformation intermediates , ejecting donor DNA from transformation heteroduplexes when 1–2 mismatches are present [28] , [29] . A larger number of mismatches can result in saturation of Hex , rendering it unable to repair a single mismatch elsewhere in the genome . The R1 and R2 fragments share 96% sequence identity , with the 4% divergence a target for Hex if alternative pairing occurs . We therefore investigated the ability of these fragments to saturate Hex by comparing transformation efficiency of two Hex sensitive point mutations , rif23 and nov1 ( conferring resistance to rifampicin , RifR , and to novobiocin , NovR , respectively ) on chromosomal donor DNA with or without R-NRf fragments . Addition of R-NRf fragments resulted in a net increase in RifR and NovR transformation efficiencies in a hex+ recipient , indicative of escape from mismatch repair ( Figure 5F ) . We conclude that Hex was saturated in >50% of cells , establishing that alternative pairing is a frequent event when a culture is transformed with a DNA fragment harboring a region repeated in the recipient chromosome . Detection of abortive chromosome junctions on unselected cells shortly after transformation ( Figure 5E ) supported the idea of merodiploid formation by transformation irrespective of lethal selection . To provide further evidence for this , we detected TD and abortive chromosome junctions by PCR on transformed cultures at different times post-transformation with R-NRf fragments . The TD junction was readily detected in the transformed population as early as 10 min after DNA internalization ( Figure 6A ) . Such timing is consistent with previous data indicating completion of ssDNA integration within ∼10 min after uptake [30] . This result confirmed that R-NRf fragments promote merodiploidy in the absence of the codY::trim lethal cassette and therefore independently of selection . Since self-transformation generated the same merodiploids as R-NRf transformation ( Figure 3 ) , TD junctions were expected to be present in cultures transformed with self DNA in the absence of selection . PCR on self-transformed cultures confirmed their presence ( Figure 6C ) supporting the view that transformation creates de novo merodiploids . The abortive chromosome was detected by PCR on the same transformed cultures ( Figure 6B ) . Similarly to TD junctions , abortive chromosome junctions were readily detected in cultures transformed with self DNA in the absence of selection ( Figure 6C ) . Interestingly , the abortive chromosome junctions disappeared 100 minutes after uptake , presumably due to cell division , and loss of abortive cells , whereas TD junctions persisted ( Figure 6B and 6C ) . To definitely prove that the detected junctions were created by transformation and specifically depend upon physical integration of the exogenous R-NRf donor fragment , we mutated three bases at the 5′ end of R in R1-A to insert a restriction site ( BamHI ) , creating R1*-A . We transformed cells with codY::trim and R1*-A PCR fragments , and recovered TrimR clones . TD junction PCRs of 3/4 tested clones were sensitive to BamHI restriction , confirming integration of donor R1*-A ( Figure 6D , left panel ) . The single insensitive PCR fragment was presumably derived from a spontaneous TD . Furthermore , PCR fragments amplified from cultures transformed with R1*-A alone were sensitive to BamHI restriction , confirming integration of the R1*-A donor fragment in the absence of any selection ( Figure 6D , right panel ) . Taken together , these results demonstrate that transformation does not simply select pre-existing merodiploids from a population but actively promotes formation of de novo merodiploids . In order to determine whether creation of merodiploids by transformation is a general phenomenon , we investigated the triggering of merodiploidy at different chromosomal sites . We identified two other pairs of truncated IS repeat regions , each sharing >90% homology and separated respectively by 144 . 1 ( site #2 ) and 210 . 6 ( site #3 ) kb ( Figure 7A and Figure S5AB ) . Transformation with R-NRf fragments corresponding to sites #2 and #3 triggered appropriate merodiploid formation , detected by PCR specific for the TD junctions ( i . e . , E-R4/3-A and E-R6/5-A; Figure S5CD ) ( Figure 7B ) . Sequencing of these junctions identified R4/3 and R6/5 mixed recombination sites ( Figure S5EF ) . Self-transformation should also provide the recipient population with the R-NRf fragments required to produce any of the TDs shown in Figure 7A . PCR detection of junctions specific for site #1-3 TDs on the same self-transformed cells harvested 40 minutes after uptake of isogenic chromosomal DNA confirmed that all three TDs were created within the same transformed population ( Figure 7C ) . We conclude that self-transformation is capable of simultaneously producing a large variety of merodiploids in a population . Finally , to determine whether formation of a much larger merodiploid could be stimulated by transformation , we transformed cells with the R4-Z donor fragment , and selected TrimR transformants . R4 shares 98% identity with R2 over 1002 bp , and alternative pairing of donor R4 with chromosomal R2 should promote formation of a 935 . 6 kb merodiploid , duplicating codY and allowing insertion of codY::trim ( Figure 7A ) . R4-Z increased transformation efficiency of codY::trim PCR 3-fold , suggesting merodiploid formation ( data not shown ) . This was confirmed by transformation with only R4-Z and detection of the specific TD junction ( E-R2/4-A; Figure S5G ) by PCR on transformed culture ( Figure 7D ) . Sequencing of this TD junction confirmed R2/4 recombination ( Figure S5H ) . These results demonstrate that transformation-induced merodiploidy is a general process which can occur at many chromosomal locations , and that over 40% of the pneumococcal genome can be duplicated as the result of the uptake and processing of a single short R-NRf ssDNA fragment .
Alternative pairing of R is a crucial step of our model ( Figure 5A ) . RecA-directed homology search being random , R presumably has an equal chance of pairing with each chromosomal repeat and our data show that alternative pairing is frequent , occurring in >50% of cells ( Figure 5F ) . Whilst pairing initiating at the ‘normal’ site ( defined by its NRf ) would result in RecA-driven synapsis extending into the NRf region , alternative pairing of R occurring first allows NRf to pair with its homologue in the sister chromatid , the second key feature of our model ( Figure 5A ) . Such pairing is presumably facilitated by three-dimensional homology search [31] involving intersegmental contact sampling catalyzed by RecA [32] . Alternative pairing leading to interchromatid integration introduces a break in the complementary chromatid ( Figure 5B ) . Restoration of complementary strand integrity , required to recover a double-stranded chromosome structure , can occur through alternative pairing and interchromatid integration of a second independently taken up R-NRf fragment ( Figures 5C and 4A ) . Restoration of complementary strand integrity can also occur spontaneously ( Figure 5B ) . Owing to the conservation in nature of annealing of ssDNA displaced by recombinase-mediated strand exchange [33] , annealing of the displaced R recipient strand to the sister chromatid R single strand by RecO could readily take place . Alternatively , this annealing could be catalyzed by the transformation-dedicated RecA loader , DprA , which has this ability [34] . Restoration of strand continuity produces a chromosome dimer which , upon resolution , generates both a merodiploid and an abortive chromosome ( Figure 5D ) . Dimer resolution prior to division of the transformant could involve either site-specific recombination [35] or homologous recombination . In any case , merodiploidy triggered by transformation is a complex process associating both ssDNA integration ( as a trigger ) and subsequent dsDNA processing mechanisms . Formed by recombination during a genetic exchange process , these merodiploids are thus true merozygotes , defined as bacterial cells containing a second copy of part of the chromosome . Remarkably , alternative pairing of a donor fragment as short as ∼3 kb triggers duplication of long chromosomal regions , ranging from the 107 . 4 kb TD observed in the codY::trim merodiploid R2597 ( Figures 1 and 2 ) up to 935 . 6 kb ( Figure 7A ) . This is reminiscent of an old observation that P1 phage transduction allowed the production of a transductant in which a duplication much larger than the quantity of P1 phage genetic material was produced [11] . The authors concluded that “…the requirements for transduction of the “condition of merodiploidy” appear to be the cotransduction of the ( TD junction ) …” and implicated “A mechanism whereby two recipient chromosomes interact with the transduced ( junction ) … to regenerate the TD …” . This outcome is in essence very similar to what we describe , although the mechanisms involved differ significantly . Merodiploid formation by transformation , involving internalization and processing of ssDNA rather than dsDNA , does not require uptake of a pre-existing TD junction but only a small R-NRf fragment . Interchromatid integration of this fragment is sufficient to trigger a cascade of recombination/repair events , ultimately leading to production of a merodiploid chromosome with a long TD . This process potentially occurs at many places in the pneumococcal chromosome , as demonstrated in Figure 7 . The frequency of duplications in a growing bacterial population is considered to range in Escherichia coli and Salmonella typhimurium from >10−2 to ∼10−4 [16] , [17] , and 0 . 005–3% of cells in an unselected laboratory culture of Salmonella enterica were estimated to contain duplication of a specified gene [14] . In S . pneumoniae , transformation frequencies with codY::trim PCR alone ( Figure 3A ) suggest that ∼0 . 05% of the recipient cells already possessed a codY duplication . We showed that 4 out of 5 transformants tested had the same TD junction as the transformation-generated merodiploids ( Figure 3C ) . S . pneumoniae thus has basal levels of merodiploidy in noncompetent cells , and transformation can be viewed as a mechanism that transiently augments the formation of merodiploids in a population . For merodiploids generated by transformation with a R-NRf fragment , a rough estimate of the expected frequency of codY::trim transformants , assuming that every alternative pairing event promoted merodiploid formation ( i . e . , that each cell could accept the lethal cassette ) , indicates that the observed frequency is between 100 and 1000-fold lower than this expectation . This could result from a failure to restore complementary strand integrity and/or a reduced probability of interchromatid interactions , hence of interchromatid pairing of NRf , due to nucleoid organization and/or choreography [36] . Nevertheless , the isolation of the R2597 merodiploid after transformation with chromosomal DNA carrying the lethal codY::trim cassette and two independent suppressors , rather than a transformant having acquired the two suppressors , indicates that merodiploid formation is more frequent than simultaneous transformation by two independent mutations . The initial merodiploid in this study was identified via transformation with chromosomal DNA containing a lethal cassette , mutating the essential codY gene [25] . We have shown that both pre-existing and transformation-generated merodiploids allowed tolerance of the otherwise lethal codY::trim cassette . These observations show that the study of essential pneumococcal genes through inactivation by transformation carries its own potential drawback , as merodiploidy can obscure the results . This was illustrated when the clpX gene , encoding part of the Clp protease in S . pneumoniae , was shown to be an essential gene [37] after initial publication of a clpX mutant strain [38] . Authors suggested that mutation of clpX occurred in a merodiploid cell which maintained wild-type and mutated copies of the gene [37] . Coupled with our observations that a basal level of spontaneous merodiploid formation exists within a population and that transformation with self DNA can transiently increase the formation of merodiploids , these results suggest that great care should be taken when attempting to mutate genes with important roles in pneumococcal physiology , and that rigorous validation of successful mutation is essential . We have demonstrated that pneumococcal transformation increases the formation of merodiploids . It is thus particularly relevant that competence for genetic transformation is induced in response to antibiotics [39] , constituting a pneumococcal SOS substitute [30] . This provides the pneumococcus with maximum potential for plasticity , including merodiploid formation , at a time when adaptation is crucial for survival . Interestingly , the increase in merodiploidy via transformation does not require the presence of non-self exogenous DNA but readily occurs with chromosomal DNA that could be released by daughter cells in the culture . Thus , the occurrence of fratricide within pneumococcal cultures [40] could provide the competent cells encountering adverse conditions with a transiently increased adaptive potential subsequent to transformation-generated gene duplication . The production of merodiploids by self-transformation thus constitutes a new , non-conservative facet of pneumococcal transformation in the sense that it creates novel junctions ( the TD junction ) . In addition , the resulting gene duplications , allowing mutation of duplicated material without the constraints of selective pressure , are likely to be of great evolution potential [16] . We conclude that merodiploidy stimulated by transformation produces a wide variety of merodiploids within a population , maximising the adaptive potential of the transformed population in response to conditions of stress ( Figure 7E ) . Our validated model for merodiploid formation relies on canonical homologous recombination and repair steps . These are likely to be shared by any organism . This is supported by the fact that the vast majority of duplications ( >90% ) are RecA-dependent [15] and none of the steps leading to merodiploidy during pneumococcal transformation differ from those documented in E . coli or Salmonella species [11] , [14] , [17] . In light of the conservation of the transformation machinery , including the presence of the transformation-dedicated RecA loader DprA in transformable species [34] , [41] , it is likely that merodiploidy can be triggered by transformation in most species . The previous findings that chromosomal integration of foreign DNA linked , on one side , to a piece of DNA homologous to the recipient chromosome occurred with similar characteristics in species as phylogenetically distant as S . pneumoniae , Acinetobacter sp . and Pseudomonas stutzeri [42]–[44] further support the view that basic transformation mechanisms are conserved . It is therefore quite likely that merodiploidy promoted by self-transformation is a transient plasticity mechanism shared by all transformable species .
S . pneumoniae strains and primers are described in Table S2 . CSP-induced transformation was performed as described previously [45] , using precompetent cells treated at 37°C for 10 min with synthetic CSP1 ( 100 ng mL−1 ) . After addition of transforming DNA , cells were incubated for 20 min at 30°C . Transformants were selected by plating on CAT-agar supplemented with 4% horse blood , followed by challenge with a 10 mL overlay containing kanamycin ( 250 µg mL−1 ) , Nov ( 4 µg mL−1 ) , Rif ( 2 µg mL−1 ) , spectinomycin ( 100 µg mL−1 ) , Sm ( 200 µg mL−1 ) or Trim ( 20 µg mL−1 ) , after phenotypic expression for 120 min at 37°C . Specifics of experiments for merodiploid formation with R-NRf PCR fragments , for creation and detection of IS861 TD and AC junctions , for detection of other TD junctions and for self-transformation are described in Text S1 . Roche 454 FLX whole genome sequencing was performed by LGC Genomics ( Berlin , Germany ) using genomic DNA isolated from mid-log cultures by the Genomic DNA kit ( Qiagen ) . For each strain , a single read library and a 3-kb span paired-end library were generated and sequenced according to Roche standard protocols . A total of 451 , 127 reads ( 96 , 405 of which contained paired ends passing quality filtering ) were obtained for R1502 ( 59-fold coverage ) , and 367 , 697 reads with 137 , 475 paired ends were obtained for R2597 ( 64-fold coverage ) . Data from the sequencing runs were mapped to the reference R6 strain ( Acc . no . : NC_003098 ) using Roche GsMapper [Release 2 . 3 ( 091027_1459 ) ] , and coverage data was extracted from the alignment results . Sequence coverage was defined as the number of times any given genomic base is represented in sequence reads . The loess function as implemented in the Loess R package was used to plot a smoothed line of the coverage as a function of genomic position . PFGE analysis was based on a published protocol [46] , [47] . Strains were grown in THY medium until OD550 0 . 2 , and digestions carried out with ApaI , SacII and SmaI enzymes . To create hybridization probes , codY or trim PCR fragments were amplified with scodY1/scodY2 ( D39 ) or strim1/strim2 ( TD80 ) primer pairs .
|
Merodiploids are defined as cells possessing a partial duplication of their genetic material , which potentially allows evolution of new genes . Historically , some have been observed in studies of natural genetic transformation . Transformation allows the bacteria to take up foreign DNA and incorporate it into their genome by homology . It is key to the high diversity observed in the human pathogen Streptococcus pneumoniae ( the pneumococcus ) . Here we show that transformation with self DNA generates a population of merodiploids with varied chromosomal duplications , up to almost half a genome in size . We show that formation of merodiploids by transformation only requires uptake of a small donor DNA fragment partially repeated in the chromosome . The donor repeat recombines with an alternative repeat on one arm of a replicating chromosome , whilst the non-repeated part recombines with its complement on the other arm , bridging the two . Subsequent recombination events generate a merodiploid chromosome with the region between the two repeats duplicated . Our results demonstrate that transformation , which is induced by stresses such as antibiotic treatments , transiently increases the ability of a population to form merodiploids . We suggest that creating a variety of merodiploids at a time of stress maximizes the adaptive potential of this pathogen .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Natural Genetic Transformation Generates a Population of Merodiploids in Streptococcus pneumoniae
|
Spo11 is the topoisomerase-like enzyme responsible for the induction of the meiosis-specific double strand breaks ( DSBs ) , which initiates the recombination events responsible for proper chromosome segregation . Nineteen PCR-induced alleles of SPO11 were identified and characterized genetically and cytologically . Recombination , spore viability and synaptonemal complex ( SC ) formation were decreased to varying extents in these mutants . Arrest by ndt80 restored these events in two severe hypomorphic mutants , suggesting that ndt80-arrested nuclei are capable of extended DSB activity . While crossing-over , spore viability and synaptonemal complex ( SC ) formation defects correlated , the extent of such defects was not predictive of the level of heteroallelic gene conversions ( prototrophs ) exhibited by each mutant . High throughput sequencing of tetrads from spo11 hypomorphs revealed that gene conversion tracts associated with COs are significantly longer and gene conversion tracts unassociated with COs are significantly shorter than in wild type . By modeling the extent of these tract changes , we could account for the discrepancy in genetic measurements of prototrophy and crossover association . These findings provide an explanation for the unexpectedly low prototroph levels exhibited by spo11 hypomorphs and have important implications for genetic studies that assume an unbiased recovery of prototrophs , such as measurements of CO homeostasis . Our genetic and physical data support previous observations of DSB-limited meioses , in which COs are disproportionally maintained over NCOs ( CO homeostasis ) .
An important function of meiosis is to precisely segregate one copy of each chromosome into cells that become gametes . This accurate division is accomplished by presenting the meiosis I spindle with homolog pairs that have been linked by crossovers ( COs ) , so that partner chromosomes can segregate from each other . Chromosomes that fail to sustain a CO are at risk of undergoing nondisjunction , and may produce aneuploid gametes that are usually inviable . A single CO can be sufficient for proper segregation . However , those COs that occur too close or too far from the centromere can also be detrimental [1]–[3] . Thus , it is essential for the meiotic cell to control the number and distribution of COs . Meiotic DSBs formed by Spo11 are regarded as the major initiating events for CO and non-crossover ( NCO ) recombination in all organisms studied to date [4] . In budding yeast , it has been estimated that 140–220 DSBs result in ∼95 COs per meiosis [5] . Only a broad estimate of how the remaining DSBs are repaired can be made due to limitation in resolution and the current inability to detect certain repair events in absolute amounts . Nevertheless , given the aforementioned limitations , roughly 40 become detectable NCOs [5] , [6] and the rest are undetectable events , including restorations ( ∼40 ) [7] , intersister ( ∼10–30 ) and possibly NCOs with very short gene conversion ( GC ) tracts . Thus the majority of DSBs formed during meiosis are directed to interhomolog repair , in contrast to mitosis where intersister repair dominates . Efficient interhomolog repair requires that the two homologs pair , at least locally , at DSBs . In yeast and mice , Spo11-initiated recombination has been shown to be essential for chromosomes to find their homologous partners and establish stable pairing [8] , [9] . Spo11-initiated recombination is furthermore necessary for SC assembly . Homologous synapsis via SC deposition along the paired chromosomes further biases repair towards a CO outcome [10] . Although a large number of synapsis initiation events are presumed to occur at recombination sites themselves , the earliest synapsis events initiate at the centromeres [11]–[13] . A small set of spo11 alleles has been previously used to assess the consequences of reduced numbers of DSBs on CO distribution and frequency [14] , [15] . These studies revealed the existence of a phenomenon whereby CO numbers remain stable despite fluctuations in total recombination events . It has been proposed that such “CO homeostasis” reflects a mechanism to ensure that COs form at the expense of NCOs when overall DSB levels are low [15] . In budding yeast meiosis , most COs result from resolution of a double Holliday junction , while NCOs arise predominantly through synthesis-dependent strand annealing ( SDSA ) [7] , [16] . Both processes involve intermediates containing heteroduplex DNA . Mismatch repair of the heteroduplex can result in gene conversion that is generally contiguous; however , it is now appreciated that gene conversion tracts may also be discontinuous [6] , [7] , [17] . Since gene conversions are associated with both COs and NCOs , they have historically been used as a metric for overall recombination . Gene conversion can easily be measured when heteroalleles are used to select prototrophs . In meiosis , repair of the DSB is initiated by basepairing one free end with the non-sister chromatid , forming a heteroduplex . Depending on the length of the gene conversion tract , when one end of the heteroduplex stops between two heteroalleles , a functional gene can be generated , making the cell prototrophic . Prototrophs are a subset of gene conversions that are detectable . The analysis of flanking marker exchange among prototrophs in the ARG4 region was used to substantiate the existence of CO homeostasis [15] , [18] . Increasingly defective spo11 alleles generated prototrophs with increasingly higher levels of CO association ( i . e . , flanking marker exchange in selected prototrophs ) , suggesting that , as DSBs become scarce , a given recombination event is more likely to be repaired as a CO . A second way to define CO homeostasis was described [19] where the fluctuation of CO number remained low compared to larger fluctuations of total recombination events in wild-type meioses . This alternate analysis refocused the phenomenon of CO control in more physiological ranges of DSBs and paved the way to study CO homeostasis in other organisms [20] . CO homeostasis has now been documented in worm , mouse and human meioses using cytological markers to measure lower than expected fluctuations in CO number [21]–[23] . It is noteworthy that measurement of GC tract lengths is constrained by the density of heterozygous markers . Two studies in yeast [24] , [25] measured GC tract lengths , averaging 1–2 kb , among identified recombinants within small well-marked regions . Recently , high density arrays and next-generation sequencing have further refined tract length measurements and expanded the data to encompass the whole genome [5] , [6] . CO-associated tract lengths were found to be quite variable , averaging ∼2000 bp , and being ∼200 bp longer than conversion tracts not associated with a reciprocal exchange . Importantly , several recombination mutants have altered GC tract lengths , revealing that mutants can affect recombination in unexpected ways [5] , [19] . Since Spo11 is the enzyme that catalyzes DSB formation , the identification of additional alleles can be a valuable tool for deciphering the role of DSB levels in various aspects of meiosis . Here we isolate nineteen new spo11 alleles and characterize them using genetics and sequencing to evaluate the relationship between recombination initiation and recombination outcomes . Historically , prototroph assays have been used as a genetic measure for DSB levels . We find , that especially for meioses with severely reduced Spo11 activity , prototroph frequencies critically underestimate DSB formation . Our findings indicate that a reduction in DSB activity influences detectability of recombination events due to alterations in the lengths of GC tracts . The extent and direction of tract length change depends on whether the event results in a CO or NCO .
Hypomorphic spo11 alleles were isolated by screening transformants carrying PCR-derived mutant SPO11 sequences for defects in gene conversions and spore viability ( Materials and Methods ) . Nineteen non-null mutants were recovered and characterized . The sequence changes in these mutants are presented in Figure S1 . Diploids homozygous for each mutation were constructed and used to quantify CO frequencies . Genetic analysis of spo11 hypomorphs must take into account that mutants with severe defects will generate mostly dead spores due to nondisjunction of nonrecombinant chromosomes [26] . Thus , the spores that are viable may have undergone higher than average numbers of COs . To eliminate this bias , we incorporated a spo13 mutation in the strains used to make genetic measurements . Meiosis in spo11 spo13 diploids generates viable diploid spores as a result of a single , predominantly equational round of chromosome segregation , independent of recombination . [27] . A wild-type diploid and twenty mutants ( 19 novel alleles and a previously characterized allele , spo11-D290A6HA [28] ) in an isogenic background ( Figure 1A ) were induced to undergo meiosis at three different temperatures ( 18°C , 22°C and 31 . 5°C ) . To isolate meiotic products for measurements of CO frequencies , arginine prototrophs were selected from purified spore populations . Arg prototrophs result from a GC between heteroalleles at the ARG4 locus . Arg+ recombinants were scored for the segregation of four chromosome 3 markers; CO frequencies on chromosome 3 were then summed to obtain an overall measure of CO frequency ( Table S1 ) . The 20 mutants characterized exhibit a wide range of recombination levels . Among the different mutants , crossover defects ranged from 50 fold reduced ( for spo11-751 ) to 1 . 6 fold reduced ( for spo11-118 ) , at the most restrictive temperature . Interestingly , the vast majority of mutants are cold sensitive; for example , the spo11-245 mutant exhibits 55% of the wild-type levels of COs at 31 . 5°C , but only 3 . 8% of the wild-type level at 18°C . To obtain an independent measure of crossing over , we conducted a physical analysis of reciprocal recombination between linear and circular versions of chromosome 3 ( [29]; Materials and Methods ) . Wild type ( WT ) and three spo11 hypomorphs were sporulated at 31 . 5°C , and chromosomes from populations of sporulated cells were then fractionated on a CHEF gel and analyzed by Southern blot hybridization using a probe from chromosome 3 ( Figure 1B ) . A single CO between circular and linear chromosomes 3 results in a linear chromosome twice as long as the original . A CO between the linear dimer and the other circle generates a chromosome three times as long as the original . The abundance of chromosome 3 CO products as measured in this assay correlates well with the frequency of COs as determined genetically ( Figure 1C ) ( correlation coefficient = 0 . 96; p-value<0 . 05 . ) . DSBs were also physically measured with the same mutant alleles in a sae2 background and were found to have a reduction in observable DSBs ( Figure 1B , C ) . Since crossing over is required for proper chromosome segregation at meiosis I , a simple expectation is that spore viability varies with CO frequency in the spo11 mutants . Spore viability was measured in SPO13 versions of the spo11 mutants used to measure crossing over ( Table S2 ) . In wild-type yeast , COs are in significant excess to the number of chromosome pairs ( ∼95 COs vs . 16 pairs of chromosomes ) . Thus , modest decreases in crossing over are not expected to decrease spore viability dramatically as long as COs are properly distributed among chromosomes [30] . Accordingly , the relationship between CO frequency and spore viability was non-linear; dramatic decreases in viability were observed only as the number of measured COs approached the number of chromosome pairs ( Figure 1D ) . The apparent requirement for more than one CO per chromosome may reflect a defect in crossover control or the inability for a functioning CO distribution system to deal with so few COs . However , we cannot eliminate the possibility that we have a selection bias for high functioning cells within a population due to our method of first selecting prototrophs . Assembly of the SC begins at the sites of synapsis initiation complexes ( SICs ) [12] , [31] and these complexes are thought to mark the sites of future COs [32] , [33] . Thus , the extent of SC formation is presumed to parallel CO level [14] . We characterized spo11 hypomorphs for the extent of SC formation by staining surface-spread nuclei with antibodies to the SC component , Zip1 ( Figures 1E and 1F ) . The synapsis phenotypes of the mutants encompassed the entire spectrum from mostly dotty Zip1 localization with only rare linear stretches to wild-type levels of full-length SC . Overall , the extent of SC formation in spo11 mutants roughly follows CO levels with those mutants undergoing more COs achieving more complete synapsis ( Figure 1G , Table S2 ) . In order to score synapsis in a uniformly arrested population for several severely defective spo11 hypomorphs , we measured SC formation in an ndt80 background . Ndt80 promotes progression beyond the pachytene stage of meiosis , thus SPO11 ndt80 cells arrest with fully synapsed chromosomes [34] . Interestingly , while some severe spo11 mutants only produced rare nuclei with partial SCs , in the ndt80 background such mutants displayed up to 96% of meiotic chromosome spreads with SC ( Figure 2A , Table S2 ) . This ndt80-associated restoration of synapsis was apparent for several alleles ( 179 , 217 , 245 , 1025 and 240 , Table S2 and data not shown ) but either weakly restored or absent in other alleles ( 246 , 117 and 845 , data not shown ) . To further examine this apparent rescue of the spo11 phenotype , recombination levels were measured by scoring Arg prototroph formation after 48 hours of ndt80-arrest by a return to growth assay . While previous work reported small increases in heteroallelic prototrophy levels in ndt80- and cdc28-arrested cells [34] , [35] our WT ndt80 strain did not increase Arg prototrophy levels compared to WT ( Figure 2 ) . However , Arg prototroph formation was increased approximately 10 and 60-fold in spo11-179 ndt80 and spo11-217 ndt80 , respectively ( Figure 2B , Table S1 ) compared to the mutants in the NDT80 background , and reached about 10% of WT ( and SPO11 ndt80 ) levels . If recombination and synapsis are restored , then spore viability should increase as well in spo11 hypomorphs arrested at late pachytene . To explore this question we used an estrogen-inducible NDT80 allele [36] , so that after a prolonged prophase arrest , cells could be induced to sporulate . Dissection of the spo11 hypomorphs after 48 hours of ndt80 arrest produced a large increase in spore viability ( spo11-179 increased from 7% to 71% and spo11-217 increased from 5% to 88% , Figure 2C , Table S2 ) . Moreover , crossing over , increased to about 60% of wild-type values ( Figure 2D , Tables S1 and S3 ) . The effect of spo11 mutations on meiotic GC was assessed by determining the frequency of prototroph formation at ARG4 . We observed insignificant temperature effects in the wild-type strain for prototroph frequency ( 2 . 3×10−4 vs . 1 . 8×10−4 when sporulated at 22° or 31 . 5° respectively ) . Although CO values appeared to be somewhat temperature sensitive ( 78 cM vs . 91 cM at 18° and 31 . 5° respectively , Table S1 ) , no effects of temperature on CO numbers or tract lengths were found by deep sequencing analysis in a limited number of tetrads ( see below ) . Diaz et al . [28] found that certain spo11 alleles have altered cut sites within an artificial hotspot , suggesting that alterations of the Spo11 complex can affect more than just quantity of cuts . While we have checked that a subset of our spo11 alleles have decreased DSBs ( Figure 1B ) we do not know if they have additional alterations in the locations of cutting . For simplicity of this analysis , we will assume that each allele is a simple hypomorph , and that temperature effects change only the number of events . ( Note that in our simulations below , we find that varying the usage of DSB sites within the ARG4 region does not significantly alter the prototrophy levels ( Figure S2 ) If prototroph formation and crossing over are similarly impacted by decreased Spo11 activity , then a directly proportional relationship should be observed when CO frequencies are plotted against prototroph frequencies . However , prototroph formation is affected more severely than crossing over ( Figure 3A ) and this trend is apparent in the entire spectrum of mutants and temperatures . The disparity between crossover and gene conversion levels increased as Spo11 activity declined . For example , a mutant with less than 1% of the wild-type level of prototroph formation exhibits as much as 15% of the wild-type level of crossing over ( e . g . , spo11-179 at 18°C , Table S1 ) . This is seen even when mutants with greater than 20% of wild-type CO levels are plotted by temperature i . e . the three regression lines would have a positive X intercept rather than at zero . The use of spo13 dyads to generate spores avoids a viability bias in these analyses . However , if there is significant heterogeneity on a cell-to-cell basis for recombination within a genotype , it is possible that the observed disparity of CO and prototroph frequencies may be somewhat exaggerated . A similar trend to Arg prototrophy was observed when prototroph formation was measured at the LEU2 locus on chromosome 3 , using a subset of SPO11 hypomorphs ( Figure 3B , Table S4 ) . To better understand the relationship between GC and crossing over , we examined Arg prototrophs for the frequency of crossing over between flanking markers . In WT , the fraction of Arg prototrophs associated with a CO was ∼59% ( Table S1 ) . In the mutants , the fraction of Arg prototrophs associated with a CO was about 1 . 4 fold higher than in WT ( Figure 3C ) , reaching a maximum of 78% crossover association , at the point where overall CO frequency ( relative CO frequency on chromosome 3 ) decreased to ∼50% of the wild-type level ( spo11-845 at 22°C , Table S1 ) . However , when overall CO values decreased below 50% , the fraction of Arg prototrophs associated with a CO fell with the severity of the allele . A similar trend was observed when conversion-associated COs were measured at LEU2 ( Figure 3D , Table S4 ) . Since the majority of the mutants are cold-sensitive , the bulk of the data providing the downward trend in CO association ( i . e . , the most severe alleles ) were derived from sporulation at 18°C . Thus , it is possible that this may reflect a temperature effect on CO association rather than a result of reduced Spo11 activity . However , inspection of the few alleles that provide similar Spo11 activity at high and low temperatures do not support the notion that cold temperatures influence CO association outcomes ( Table S1 , S2 ) . With this caveat , we analyze the data using the simpler model that Spo11 activity is responsible for changes in CO association . Our genetic experiments reveal two unexpected results . First , heteroallelic gene conversion levels ( measured at ARG4 and LEU2 ) fell at a much greater rate especially in alleles with severely compromised Spo11 activity than other indicators of recombination ability ( e . g . CO levels , spore viability and SC formation ) . Second , although flanking marker exchange among Arg and Leu prototrophs increased as expected in spo11 alleles with moderate defects ( presumably due to CO homeostasis ) , in the most severe alleles , CO association surprisingly decreased compared to less severe mutants . Previous work using moderate spo11 alleles noticed a weakening of CO homeostasis as DSB activity decreased and proposed that CO homeostasis was strongest in cells with relatively high levels of DSBs [15] . If the trend of CO homeostasis had continued , we would expect that the alleles where Spo11 activity is severely compromised would show a continued gradual increase in CO/NCO ratio . Together , the hyper-reduction of prototrophs and the decrease in CO association suggests that the effects of severely reduced DSBs results not only in fewer events , but influences their repair outcome . To better understand the effects of spo11 mutations on meiotic gene conversion and crossing over , we monitored meiotic recombination on a genome-wide scale by sequencing the genomes of spores from tetrads containing four viable spores . For this analysis , we used a diploid derived by mating a typical laboratory strain ( S96 ) to a clinical isolate ( YJM789 ) . The parental strains differ from each other at ∼55 , 000 locations ( single-nucleotide polymorphisms ( SNPs ) ) allowing most interhomolog recombination events to be detected [6] , [37] . Spores from five tetrads derived from spo11 hypomorphs ( three from spo11-217 and two from spo11-32 , sporulated at 24 or 31 . 5 oC ) were subjected to high-throughput , multiplexed sequencing . The data were compared to sequenced data from three wild-type tetrads sporulated at 30°C of which two were reported previously [6]; the wild-type tetrads had an average of 92 COs ( Table 1 ) . We also analyzed one wild-type tetrad at 22°C and two wild-type tetrads at 26°C . The recombination characteristics of the wild-type tetrads do not differ with temperature ( Table S5 ) . Two of the tetrads from the spo11-217 mutant had an average of 80 COs ( Table 1 ) ; these were classified as “high-functioning ( HI ) ” . The remaining tetrad from spo11-217 and the two tetrads from spo11-32 sustained an average of 55 COs ( Table 1 ) ; these were grouped together and classified as “low-functioning ( LO ) ” . Comparison of recombination in these tetrads with the genetic analysis on chromosome 3 reveals a large bias from selecting four-spore viable tetrads from mutants with low spore viability . spo11-32 strains sporulated at 31 . 5°C show only 26% of wild-type crossing over in spo13 dyads compared to 60% of wild-type COs from the two sequenced four-spore viable tetrads . It seems likely that this difference reflects the fluctuation of DSB activity on a cell-by-cell basis and that selecting for four-spore-viable tetrads constrains the range of DSBs in the meioses that we can recover . We assume that the reason a given mutant acts differently in individual tetrads is due to a difference in DSB number , and thus we categorized these tetrads based on numbers of COs ( rather than genotype ) . Deep sequencing of wild-type tetrads revealed recombination types that included single and multiple COs and NCOs with and without continuous tracts , as well as multi-strand events [6] . Examination of the recombination landscape in the five SPO11 hypomorphic tetrads revealed a similar proportion of recombination types as observed in WT . The sole exception is a disproportionate reduction of NCOs ( Table 1 ) as will be discussed below . We note that the proportions of multi-strand COs relative to the total COs were found to be the same for spo11 hypomorphs and WT ( Table 1 ) . This suggests that events classified as complex events are likely to arise from a single DSB rather than via multiple DSB events , since reducing the number of DSBs did not decrease the proportion of the multi-strand COs . CO interference is the nonrandom spacing of COs that is a feature of the normal distribution of COs but not NCOs . Through sequencing analysis we obtained the positions of COs genome-wide and we then examined interference by fitting a gamma distribution to intercrossover distances [19] obtained from grouping all five spo11 tetrads . This grouping was necessary to obtain sufficient numbers of intercrossover distances . Gamma values obtained from the distribution reflect the strength of CO interference . Consistent with previous genetic analysis of interference that showed no change in interference for spo11 mutant alleles [15] , our genome-wide analysis detected wild-type levels of interference ( gamma = 2 . 0; gamma = 1 . 0 represents no interference and 1 . 8 is wild-type interference ) [19] . Moreover with the reduction of DSBs , it is likely that some chromosomes will fail to experience a CO . From the five sequenced hypomorphic tetrads we found five instances where small or relatively large chromosomes ( chromosomes 1 , 8 , 8 , 13 , 14 ) lacked a crossover ( E0s ) . On these 5 E0 chromosomes no other events ( NCO's ) were observed . Successful detection of NCOs is subject to the density of markers as well as tract length ( e . g . , short tracts have a reduced probability of converting a marker and thus are less likely to be detected ) . As shown in Table 1 , the total number of detectable events decreased in the mutants , but NCOs decreased to a greater extent than COs . As a result , COs represent a somewhat greater fraction of total events in the mutants ( an average of 77 . 1% in the LO tetrads versus an average of 66 . 7% in the wild-type tetrads; P = 0 . 04 , two sample t-test ) , suggesting a relative increase in COs at the expense of NCOs . However , similar to our previous analysis using prototroph formation , this conclusion assumes that the probability of detecting a NCO by sequencing is constant across the spectrum of mutants . Gene conversion tracts result from the repair of heteroduplex or double-stranded gaps formed during DSB repair . In order to assess gene conversion tract lengths in spo11 hypomorphs , we inspected all CO and NCO events for regions of gene conversions , i . e . , regions of 3∶1 or 1∶3 segregation of markers flanked by regions of 2∶2 segregation . The distance between the last marker showing 3∶1 segregation and the closest marker showing 2∶2 segregation is highly variable due to the differences in the density of markers within different regions along the chromosomes , leaving the actual tract length ambiguous . For example , a 3∶1 tract that spans only 1 kb may have the next 2∶2 markers at 10 kb on one side and 1 kb on the other . Using the common midpoint method , this tract would be scored as 6 . 5 kb ( 1 kb minimum plus half of each maximum tracts ) in length . Even a few of these might skew our tract length analysis . In order to better define tract lengths , we developed an algorithm , called Tract-Seq , that estimates tract length using a Markov model for tract growth combined with Monte Carlo simulations ( Materials and Methods , Figure 4A ) . Figure 4B shows the distribution of wild-type tract lengths calculated by Tract-Seq in comparison to the distribution calculated by taking the midpoints . We generated similar graphs of “Tract-Seq” lengths for HI and LO ( Figure 4C ) . Average tract lengths were determined from the log-normal fits for the distribution of CO-associated tracts and NCO tracts for WT , HI and LO tetrads ( Figure 4D and E ) . CO-associated tracts increase in length as the Spo11 activity decreases; the means increased by 56 bp for HI tetrads and 316 bp for LO tetrads , compared to WT ( Table 2 ) . However , CO-associated tract lengths for LO tetrads , but not HI tetrads , compared to WT are significantly longer ( p = 0 . 05 for LO vs . WT and p = 0 . 28 for HI vs . WT ) . In contrast , NCO tracts are smaller in context of decreased Spo11 activity; the mean length of tracts decreased by 134 for HI and 292 bp for LO ( Table 2 ) . Again , the difference in NCO tract lengths relative to WT is significant for LO tetrads , but not for HI tetrads ( p = 0 . 02 for LO vs . WT and p = 0 . 17 for HI vs . WT ) . For both CO- and NCO-associated tracts , the range of possible tract lengths increases as Spo11 activity decreases . The decrease in NCO tract length associated with decreasing Spo11 activity makes it possible that the abundance of NCOs relative to COs in the spo11 mutants was underestimated by sequencing analysis . This could occur if the decrease in NCOs is not an actual decrease in the number of NCOs , but a decrease in the ability to detect NCOs given their shorter length . To address this issue , we calculated how many NCOs would be missed due to detection issues ( Materials and Methods ) . For LO , 6 . 3+/−0 . 2% are missed . Thus as Spo11 activity declines , the ability to assess the true number of NCOs becomes more difficult and could influence calculations such as the NCO/CO ratio , though marginally . Besides detection , alterations in tract length could have unexpected consequences on measurements of DNA repair . Tract length changes of NCOs and COs in opposite directions , might affect measurements of prototroph formation or the association of a reciprocal recombination with gene conversion . In order to explore whether tract length changes in the spo11 hypomorphs could account for the discrepancy between prototroph levels and crossing over , we tested if we could obtain the observed ARG4 prototroph frequencies ( Figure 3A ) by computationally distributing tract lengths around the ARG4 hotspot for each of the three levels of Spo11 activity . For this analysis , we took advantage of the detailed information available regarding the location and usage of meiotic DSB sites near ARG4 [38] . Although these data were generated in the SK1 strain , we reasoned that the distribution of DSBs describes most genetic backgrounds since available comparisons with a hybrid strain used for microarrays and sequencing ( YJM789xS96 ) have so far been in general agreement [7] , [17] , [19] . Information regarding tract length distribution and the relative abundance of NCOs and COs was used to generate sets of tracts for WT , HI and LO tetrads . Relative DSB frequencies at three hotspots in the vicinity of ARG4 were taken from a DSB hotspot map derived from Spo11-oligos and were used as a template to position these tracts [38] ( Figure 5A ) . We assumed that a tract initiating at one of the three hotspots and ending within the region between the two ARG4 heteroalleles could generate a prototroph . We found that as Spo11 activity declined , projected prototrophs declined at an increased rate , comparable to the genetic results ( Figure 5B , red dots ) . We next asked if our tract length distributions could recapitulate the change in the fraction of Arg prototrophs associated with a CO that we observed genetically ( Figure 3C ) . By determining which of the prototroph-forming tracts were associated with a CO and which were due to NCOs ( Materials and Methods ) a trend similar to that seen for the genetic results emerged , with increasing contributions from the COs as Spo11 activity declined ( Figure 5C , red dots ) . The percentage of prototrophs associated with a CO increased from 54% in WT , to 63% in the HI tetrads , and 68% in LO tetrads ( Figure 5C ) . In Figure 5F , the proportion of NCOs and COs contributing to prototrophy as Spo11 activity decreases was obtained from the sequencing data ( Figure S3A ) . Figure S3A indicates that as Spo11 activity decreases , COs represent a greater proportion of recombination repair , an observation consistent with the presence of CO homeostasis . However , is CO homeostasis necessary to attain the good fit to the genetic CO association data observed in Figure 5C ? To address this question , we simulated tract lengths for our mutants ( HI and LO ) and generated values for CO association assuming no CO homeostasis , that is , we used the wild-type proportion of COs even when Spo11 activity declined . In the case of no CO homeostasis , the calculated values for CO association ( Figure 5C , blue circles ) , did not match the genetic data , unlike the good fit achieved when CO homeostasis is incorporated ( Figure 5C , red circles ) . We also looked at the effects of varying other parameters independently Figure S2 . Can our experimentally generated genetic data be used to understand the relationship between tract lengths of NCOs and COs as Spo11 activity decreases to extremely low levels ? The HI and LO tetrads examined above represent a limited range of Spo11 activity . It was not possible to obtain the sequence data needed to measure tract lengths for severe mutants because they do not produce tetrads with four viable spores . To circumvent this biological limitation , we performed simulations with various models of how tract length would change , extrapolating from the data of WT , HI and LO tetrads ( Figure S3B–F ) . We then determined which of our simulations best describes our genetic observations . First , we computed the number of NCOs as a fraction of total events in hypothetical mutants having 30 , 10 and 3 COs , assuming a linear trend of the CO fraction as seen in our experimentally determined data ( Material and Methods , Figure S3A ) . Second , we determined the mean tract length of log-normal CO and NCO distributions for these hypothetical mutants applying four different regression models of tract length behavior ( Figure S3B–F , Materials and Methods ) . We asked whether these models would recapitulate prototroph levels as a function of CO levels . All models tested were consistent with the aforementioned trend that , as COs declined , the frequency of prototrophs decreased at an increased rate , to as much as 7-fold lower than the decrease in COs ( Figure 5D , data not shown ) . We then asked which model would best recapitulate the genetic data with regard to the fraction of Arg prototrophs associated with a CO . Of all four regression models , only the inverse model for the mean CO tract distributions that predicts an inverse relationship between tract lengths and COs recapitulated the genetic data well ( 70% , 67% and 45% for simulated vs . 68% , 70% and 59% for genetic , for mutants undergoing 30 , 10 and 3 COs , respectively ) ( Figure 5E ) . Moreover , although all models gave reasonable fits for Arg prototrophy frequency , the inverse model for mean CO tract distributions was the overall best ( 16% , 4% and 0 . 6% values vs . 12% , 1% and 0 . 5% genetic values , for mutants undergoing 30 , 10 and 3 COs , respectively ) ( Figure 5D ) . Figure S3G shows that the inverse model for the CO tract length distribution predicts a severe increase in CO associated tract lengths in the cases where Spo11 activity is greatly diminished . This large increase in tract length in our model could explain why prototroph frequencies decline faster than CO frequencies in the measured genetic data as Spo11 activity declined . This suggests that as Spo11 activity decreases , gene conversion tracts associated with COs become too long to produce prototrophs , and NCO tracts become too short to produce prototrophs .
Tract lengths associated with COs become longer as Spo11 activity declines . Longer tracts may be due to an increase in branch migration and resection [42] . Crossover recombination is thought to occur at SICs , which , like recombination nodules , contain many of the enzymes thought to be necessary for catalyzing the CO [43] , [44] . In spo11 hypomorphs , there are fewer SICs , as estimated by numbers of Zip3 foci [14] . We propose that , in the SPO11 hypomorphs , recombination enzymes , including those for resection and mismatch repair , are distributed between fewer SICs , thus enriching them with a relative abundance of enzymes . Consequently , these SICs may be more active , and generate longer resections . This proposed increase in enzymatic ability in hypomorphs with low Spo11 activity might increase in a nonlinear way , reflecting the predicted increases in tract lengths . A similar explanation was given to explain the long resection products of a single VDE cut in a spo11 null background [45] , [46] . Efficient DSB formation at a site in an otherwise DSB-deficient strain ( spo11-null ) produced longer resected tracts and generated fewer prototrophs than expected , and this phenomenon was apparently due to reduced levels of DSBs in the cell ( and not synapsis defects ) [45] . Similarly , Malkova et al . [47] , [48] saw increased GC tracts from HO-induced DSBs in strain backgrounds with no meiotic DSBs . In contrast to GC tracts associated with COs , NCO tract lengths are shorter in cells with diminished Spo11 activity . Most NCOs are thought to occur by a different mechanism than COs which may help to explain this altered outcome . While COs go through a double Holliday junction intermediate , NCOs are thought , for the most part , to undergo a single strand invasion or SDSA ( Synthesis-Dependent Strand Annealing ) [16] . SDSA events are likely independent of SICs , since the number of these events is not reduced in mutants lacking SIC components [49] , [50] . We propose that the reduction in NCO tract length is due to SDSAs forming in the absence of stably associated homologs . That is , the Spo11 hypomorphs have reduced levels of homolog pairing and SC ( [14] , this paper ) , which may otherwise stabilize the interaction of non-sister chromatids . We propose that SDSA recombination is sensitive to this lack of reinforced homolog alignment . In the absence of stable associations , the movement of chromosomes in the prophase nucleus [51] could interrupt the SDSA event prematurely , leading to even shorter gene conversion tracts . Nuclei with less Spo11 activity and fewer paired chromosomes would have the greatest obstruction to SDSA events . Our data raise important considerations regarding the characterization of CO homeostasis . First , sequencing of spo11 hypomorphic tetrads revealed an overall reduction in recombination events , with a stronger reduction in NCOs , supporting the notion of CO homeostasis . However , since NCO tracts in strains with limiting DSBs are shorter , these data may reflect an inability to detect NCOs due to their lesser likelihood to span a heterologous SNP . We have estimated the loss of NCOs in the mutants by positioning tracts on the genome taking into account the actual SNP density map . In the case of our most severe hypothetical mutant ( 3CO ) up to 16% of the NCOs become undetectable yet this apparent loss of NCOs only marginally affects measures of CO homeostasis . An alternative explanation for the reduction of NCOs , particularly in severe mutants where homolog pairing is severely compromised , may be that some DSBs fail to find their homologous partner and are instead repaired using the sister chromatid [52] . Second , the assay used to measure CO homeostasis , flanking marker exchange of prototrophs , assumes that a tract emanating from a CO or an NCO would have the same probability of forming a prototroph . However , since tract lengths change in unexpected ways in both NCOs and COs , and thus alter their abilities to form prototrophs , apparent flanking marker exchange cannot be assumed to be equal , particularly in severe hypomorphs . Note that different pairs of heteroalleles and different loci will elicit unique outputs . Using a novel color system to study CO homeostasis in tetrads , Thacker et al . [18] were unable to detect CO homeostasis in a flanking marker assay in the spo11-HA mutant which has little impact on DSB levels . The authors argued that tract-length alterations might obscure NCO/CO ratios . Consistent with this , if we apply our observed tract length changes to the GFP alleles , we would expect a decrease of GFP+ recombinants associated with COs and an increase in GFP+ recombinants unassociated with a CO , since the heteroalleles are so close in this assay . So although our measurements of tract length in the HI mutants may not be significant , it seems reasonable to suggest that tract lengths are changed in modestly defective spo11 alleles in both budding and fission yeast ( see above ) . Consistent with the operation of CO homeostasis , we were only able to recapitulate the CO/NCO genetic data when applying the experimentally observed changes in numbers of CO and NCOs . Thus our data support the accumulating evidence that as Spo11 activity decreases , the fraction of recombination events that become COs increases . Our observation that two severe spo11 alleles can be rescued for SC formation , recombination and spore viability by the ndt80 mutant suggests that holding cells in prophase allows further DSB formation . This notion , that DSB formation may continue in cells held at mid prophase , has been suggested previously [53] , based on limited increases in prototroph formation of arrested cells . Further support came from the observations of Allers and Lichten [16] in which they report an accumulation of joint molecules during Ndt80 arrest . However , we find that WT diploids arrested at the Ndt80 arrest point fail to show genetic evidence of the additional DSBs in spores ( i . e . , no increase in heteroallelic recombination or crossing over , Figure 2 ) . If continued DSB activity is the result of Ndt80 arrest , then why is the genetic consequence strongly observed in certain spo11 hypomorphs , but only minimally in WT ? Two scenarios seem likely . In the first , DSBs are unregulated ( occurring in all backgrounds ) , but the repair outcome is dependent on the state of synapsis . In synapsed chromatin , the repair would favor the sister ( SC is inhibitory to interhomolog recombination ) and in unsynapsed chromatin , interhomolog repair would be favored . For poorly synapsed mutants , Ndt80 arrest would result in a disproportionate increase in observable recombination ( and consequently increased synapsis ) . This model , where stably paired and/or synapsed chromatin is protected in cis from additional interhomolog repair , has been previously suggested to account for observations in worms [54] . A second possibility has been suggested for mouse meiosis , where DSB formation may be limited within synapsed chromatin [55] . If DSB activity , rather than repair , is modulated , such that unsynapsed chromatin receives more DSBs than already synapsed chromatin , and the DSBs are repaired normally ( in favor of inter-homolog repair ) , this would also disproportionately affect meioses with poor synapsis . In either case , how DSBs and their repair may be controlled gives insight into how the meiotic cell may ensure higher levels of pairing , SC formation and recombination in meiotically compromised cells .
Mutants of SPO11 were generated by a degenerate PCR based method . First , the endogenous SPO11 gene from our strain background ( BR1919-8B ) was cloned and sequenced , and it was found to encode six altered amino acid changes compared to that reported for S288C ( SGD ) , although it had the identical amino acid sequence to SK1 , another strain often used to study meiosis . The sequence of the SPO11 gene contains a string of 12 As which can corrupt the integrity of its sequence during PCR amplification . To alleviate this problem , two As in the middle of the run were silently substituted with Gs ( A225G , A228G ) . A centromere plasmid containing this functionally wild-type SPO11 gene and URA3 ( pRS316:SPO11 ) was subjected to degenerate PCR amplification ( GeneMorphII EZClone Domain Mutagenesis kit from Stratagene ) . Three independently mutagenized plasmid libraries were prepared and transformed into a spo11-null diploid yeast strain ( BR4590 ) for screening ( Table S6 ) . In order to identify non-null alleles of SPO11 , assays for both prototrophs at the LEU2 and THR1 loci , and spore viability ( CanR CyhR ) were incorporated into the diploid yeast strain used for the screen ( see Table S6 ) . Using this strain , null mutants , after sporulation would form no or very few papillae on either the leucine or threonine omission medium or the drug-containing medium after sporulation . Both cold-sensitive and heat sensitive alleles were isolated by screening replicas sporulated at different temperatures ( 22° and 30°C ) . Over 26 , 000 transformants were screened from the three independent mutagenized libraries . 10 . 8% of the screened transformants behaved as null mutants , indicating that the mutagenesis was successful . Those passing a second screening ( 52 mutants ) were chosen for plasmid retrieval , sequencing and retesting by retransformation into BR4590 . The mutations were distributed throughout the gene ( Figure S1 ) . Six mutants were isolated more than once and at least five of these were independent . Nine of the mutants encoded a single amino acid change , but we chose to include eight alleles with two amino acid changes and two with three changes in our analysis . The sequenced alleles were subcloned into integrating vectors ( pRS306 ) so that they could be transformed stably into yeast by pop in/pop out and studied as homozygotes . Crosses to generate the various strains utilized a marked SPO11::KAN ( or SPO11::HYG or SPO11::NAT ) allele to monitor the unmarked mutants . All strains for subsequent analysis are isogenic to BR1919-8B [3] and any markers were changed by transformation . Isogenic derivatives were used for the subsequent analysis ( Figure 1A , BR5348 and it's spo11-m derivatives Table S1 , S6 ) . For nineteen of the new mutants and one previously described ( D290A [28] ) , arginine prototroph frequencies and map distances for three intervals on chromosome 3 were obtained from cultures sporulated at 18 and 22 , and 31 . 5°C . Prototroph frequencies were averaged from at least three cultures and premeiotic frequencies were determined and subtracted . Map distance measurements in dyad spores are complicated by occasional reductional segregation and the fact that recessive phenotypes can be masked by the dominant allele . Map distances for chromosome 3 markers were determined by first sorting out all the phenotypic classes . Although several classes of events are hidden by heterozygosities in one spore , their sister spore would be scorable . Equal numbers of scorable genotypes were then subtracted from the hidden classes and added to the appropriate recombinant class . Then CO's are summed for each interval to provide map distances . In a subset of mutants ( spo11-32 , 179 and 217 ) and WT , LEU2 heteroalleles were added to the strains to measure prototrophs at a second locus . The leu2-3 , 112 allele already exists in the strain background . To create a second allele , a haploid was first transformed to Leu+ and subsequently transformed to leu2-cla by pop-in pop-out transformation ( R477: pBR322 with EcoRI/XhoI fragment of LEU2 where the ClaI site of LEU2 was filled in with Klenow ) . HIS4 and iHYG flank the LEU2 gene . Strains for the sequence analysis were heterozygous for the spo11 alleles . The YJM789 strain is deleted for spo11 and constructed by transformation of ura3::NAT derivatives of the two haploids , S96 and YJM789 [37] , with pRS306-spo11-32 and pRS306-spo11-217 ( and subsequent popout ) . Zygotes were isolated and sporulated at various temperatures . Tetrads were used from the spo11-217 strain from dissections at 31 . 5°C ( spore viability was 81%; 78/96 ) and 24°C ( spore viability was 11%; 65/576 ) and from the spo11-32 diploid , tetrads were used from dissections at 30°C ( spore viability 10%; 58/576 with only four 4-spore viable tetrads ) . Wild-type hybrid strains were examined for temperature effects by sporulation at 22°C , 26°C and 30°C . Wild-type sequences are stored at the National Center for Biotechnology Information Sequence Read Archive ( Bioproject accession number SRP028549 ) . Output files from Recombine , CrossOver and TractSeq are available from the Dryad Digital Depository: http://doi . org/10 . 5061/dryad . 53t4c . The fraction of Arg prototrophs resulting from a CO was determined from analysis of random dyad spores . Flanking markers , THR1 ( 4 cM distal to arg4-Nsp ) and iTRP1 ( a marker placed near the centromere , 16 cM proximal to arg4-Bgl ) were scored for nonparental segregation among Arg prototrophs . Since the most frequently converted allele is cis to the Trp+ marker , most of the Arg+ spores are Trp+ and the reciprocal recombinants are Thr− . The number of Trp+ Thr− recombinants were doubled to account for the inability to see the reciprocal event . A similar assay for CO association was carried out on chromosome 3 , utilizing LEU2 heteroalleles and using the hygromycin resistance marker at CEN3 and the distal marker , HIS4 as flanking markers ( Figure 1A ) . The configuration of the LEU2 alleles and the flanking markers is best suited for detecting recombinants ( leu2-cla converts ∼80% of the time and thus is flanked by both HygR and HIS4 ) Table S4 . Plugs were prepared from meiotic cultures and subjected to CHEF gel analysis [29] and Southern hybridization with a probe from chromosome 3 sequences ( THR4 region ) prepared from random priming kit ( Rediprime II , GE Healthcare ) and analyzed on a Storm phosphorimager ( GE Healthcare ) . The “fraction recombinant bands” was estimated by summing twice the intensity of the trimer band plus the dimer band over the total intensity of the three bands . The averages of five experiments are presented . For the DSB analysis , plugs were made , ran , and hybridized as above , but from sae2Δ versions of the mutants . Three gels were averaged . Meiotic chromosomes were spread according to Rockmill [56] . Cultures sporulating at 31 . 5°C were spread after 15 hours , those sporulating at 22°C were spread at 24 hours and those sporulating at 18°C were spread at 30 hours to maximize those cells with the most SC . Since sporulation in this strain background is not synchronous , Red1 staining was used to identify nuclei at a similar stage , mid prophase I . Red1 accumulates on the chromosomes from early in prophase and culminates at the stages where the SC formation is most abundant , and is lost by Meiosis I . SC is scored by visualizing the Zip1 protein by immunofluorescence . spo11 mutants often form an aggregate of Zip1 and other proteins called a polycomplex ( PC ) . In our strain background WT does not normally form PCs . SC was scored among Red1 positive nuclei as follows: One or no Zip1 lines ( the PC can appear linear ) is “no SC”; Two or more Zip1 lines but fewer than complete SC is “some SC”; and Zip1 lines encompassing the entire nuclear region is “full SC” . Raw data is presented in Table S2 . DNA from four-spore viable tetrads was purified and processed for Illumina high-throughput sequencing . Illumina sequencing libraries were generated using adapters for multiplexing , as described [57] . In general , four barcoded libraries , one from each spore of an individual tetrad , were mixed in equimolar ratios and processed on an Illumina Genome Analyzer II . One tetrad was analyzed using a two-plex strategy . Each sequence read started with a 4-bp index followed by 32 bp from the sample . Raw sequencing data were first processed by Illumina's Casava pipeline . After barcode parsing , the remaining bases were aligned against the S96 and YJM789 genomes using ReadAligner [6] . Barcode sorting , alignment to reference genomes , genotyping and detection of recombination events were performed using a suite of programs included in ReCombine [6] . A pillar of our analysis rests on a stringent and accurate estimation of gene conversion tracts for CO's and NCO's among various tetrads . Again the ReCombine package was employed , with a few modifications . First , we examined marker calls consisting of SNPs and indels to ensure the use of high-quality heterologies and to remove markers that might be misannotated and thus assigned to the wrong parent genome . A “confusion matrix” was created by comparing quality scores against S96 and YJM789 genomes from sequenced S96 and YJM789 parental strains , an important metric during genotype calling . For example , from a sequenced S96 parent , if a SNP had a higher quality score for YJM789 than S96 or equal scores , this heterology would be classified incorrectly as an YJM SNP . This ambiguity would lead to false positives and incorrectly called events and tracts . The estimated error rate was 0 . 3% for SNPs and 29% for indels after a first-pass analysis; these markers were excluded . Nevertheless , some single marker events encompassed an indel with mixed reads that passed our first-pass thresholds . Sanger sequencing of genomic DNA from spores failed to validate 13 out of 14 such events . We limited our analysis to SNP marker calls given the higher genotyping errors found in indels . Second , marker calls were made on a tetrad basis and on individual spores to increase the precision of tract length predictions . The ReCombine package determines events and tracts considering all four spores [6] . It excludes from analysis high-confidence heterologies that could not be called in any one of four spores in a tetrad , often due to reduced local sequencing depth ( no sequence read spanning the SNP ) . Using as a template the recombination map generated by ReCombine , we confirmed these events in a separate single spore analysis . This approach better defined tract lengths , by extending tract lengths , adding better boundaries to existing tracts and increasing the number of heterologies defining a given event . Moreover , our strategy uncovered a handful of recombination events that were missed by tetrad-based analysis . Overall , tetrad-based and individual spore analyses gave very similar results . Tract length boundaries were predicted using the individual spore procedure described above . The minimal tract length was defined by markers that are certain to be part of the gene conversion event , while the maximal tract length was defined by the next 2∶2 segregating flanking markers , before which the gene conversion tract must end . The difference between the maximal and minimal tract lengths is often large , leaving the actual tract length ambiguous . Rather than using a commonly used measure , such as the midpoint , we have developed an algorithm to better estimate the tract lengths , taking into account biologically relevant information such as enzymatic processivity and marker distribution asymmetry ( Figure 4A ) . Processive enzymes have a probability “p” of moving to the next base , and a probability “1-p” of falling off , following a geometric distribution ( or exponential law ) . Statistical approaches have been able to model gene conversion tracts from genetic data using a maximum likelihood approach , considering ressection as a succession of Bernoulli trials [58] . Assuming bidirectionality from an initiation site , tract lengths would then follow a log-normal distribution [58] , [59] . In Figure S4 we show the results using unidirectional compared to bidirectional tract formation . From gene conversion data obtained by genetic means , enzymatic processivity parameter “p” has been estimated to be around 0 . 999 in flies , yeasts and mice , from a meta-analysis [59] . Sequencing data differ from genetic assays in the measurements of tract lengths . In genetic analyses , an initiation site , i . e . the position of a DSB hotspot , is often known , and gene conversion gradients are usually measured from one side , establishing tract lengths that may only reveal half the length . Whereas a conversion tract determined by sequencing consists of a central region of certainty ( minimal tract length ) flanked by regions of uncertainty ( maximal tract length ) . We developed a novel analytical method , Tract-Seq , to estimate more accurately the lengths of gene conversion tract , or any marker-based estimation of enzymatic processivity from sequencing data , using Monte Carlo simulations ( Figure 4A ) . For each iteration , the bidirectional enzymatic complex falls from the end of the minimal interval , using an exponential law decreasing away from the minimal tract on both sides of the interval . These are termed endpoints . Tracts extending past the maximal gene conversion interval were not considered as were those not extending past the minimal gene conversion interval . We assumed a similar probability “p” as the enzymatic processivity ( 0 . 999 ) [59] . Using p values of 0 . 9970 to 0 . 9993 , showed indistinguishable trends in tract lengths , suggesting these trends are robust to slight perturbations of enzyme processivity . The estimated tract length was then represented by the sum of the following distances: endpoint 1 to minimal tract ( start ) + minimal tract + minimal tract ( end ) to endpoint 2 . After a minimum of 10 , 000 iterations , the median length of the estimated tract was used for distribution fitting . Statistical analysis of the median of tracts was performed using a non-parametric resampling test . For events that span a single marker , a minimal interval cannot be determined with this method . This is often the case in regions with low marker density . As an example , a maximal interval might be 20 kb while a minimal interval of 1 bp . In such cases , using the maximum likelihood estimation from geometric distribution with processivity “p” described above to fine-tune estimated tract lengths , intervals of similar length would be generated . Since these point events are common ( ∼15% ) , it would result in biases during distribution fitting . To circumvent this problem , we made an estimation of tract length for tracts defined by a single SNP . From the linear relationship between maximal tract lengths and tract lengths defined by multiple SNPs , we determine an approximation for the tract length for a single SNP tract based on its maximal tract length . To ensure a better overall fit , tract lengths from COs and single tract NCOs were grouped by their levels of Spo11 activity: WT , HI and LO . Tracts from complex events , such as NCOs on two chromatids and NCOs associated with a CO , display longer tract lengths than regular COs and NCOs , and these types were treated separately ( and SPO11 genotypes were pooled ) . Gene conversion tract lengths can be described by a log-normal distribution [58] , [59] ( see Text S1 ) . Log-normal curves were fitted by a Bayesian analysis with Markov Chain Monte Carlo ( MCMC ) , using the R package – rjags ( JAGS , http://mcmc-jags . sourceforge . net/ ) to determine the mean and standard deviation of log normal distributions . We tested the goodness of fit using Kolmogorov-Smirnov tests and could not reject any fitted curves from tract length data points ( P>0 . 05 in all cases ) . Significance in the differences of the means was tested using re-sampling by comparing actual means along with the corresponding error on the means obtained from MCMC iterations ( sdmean ) , correcting these sdmean for sample size differences ( i . e . different numbers of tract lengths ( data points ) used in the distribution fitting process ) . To estimate the percentage of NCOs that would be missed because of shorter conversions tracts , we distributed onto our SNP map , a wild-type number of NCO tracts that were sampled from the fitted tract length distributions for WT , HI and LO . The percentage of NCO tracts that span at least one SNP was recorded . The difference between the percentage detected in WT vs . HI and WT vs . LO estimates the change in NCO numbers due to detection issues . Gene conversion tracts were generated from fitted CO and NCO log normal distributions for WT , HI and LO . Similarly , tracts were created for severely hypomorphic SPO11 mutants from log normal curves using inferred parameters , as described above . We assume that genome-wide averages hold for any individual site . Tracts originated from one of three DSB sites around the ARG4 locus , in proportions that parallel the frequency of DSBs found by Spo11 oligo sequencing [38] . A tract originating from one initiation site and of length falling between the Bgl and Nsp heteroalleles is classified as prototroph-forming ( Figure 4A ) . DSB site 1 , located downstream of ARG4 , is the initiation site for 6% of simulated tracts , which form prototroph if they are 1544 to 2818 bp in length . DSB sites 2 and 3 , upstream of ARG4 , accounts for 53% and 41% of initiations [38] and tracts emanating from these sites can generate a prototroph if they are comprised between 204 and 1478 bp for DSB site 2 , or between 2354 and 3628 bp for DSB site 3 . Monte Carlo simulations were performed to record the average number of prototroph-forming events and its standard deviation for different classes of events emanating from one of three DSB sites . For each of 1 , 000 iterations , more than a million random tracts were generated from various CO and NCO log normal distributions , totaling more than a billion tracts considered per genotype . The number of prototroph-forming tracts was then adjusted to follow the number of various types of meiotic recombination events for SPO11 hypomorphs . The percentage of prototroph-forming tracts from NCOs and prototroph-forming tracts originating from a CO event were calculated . To determine the individual effect of the parameters used , we varied each parameter separately: biased usage of ARG4 hotspots , NCO/CO ratio and tract length variations ( Figure S2 ) . Hypothetical mutants with 30 , 10 and 3 COs were designated as highly hypomorphic , ( for which four-spore viable tetrads could not be experimentally obtained ) . Number of COs with tracts was estimated at 71% from pooled WT , HI and LO tetrads . Assuming a linear decrease in the proportion of NCOs as Spo11 activity diminishes ( CO homeostasis ) , the number of total NCO's was estimated at 5 . 5 , 1 . 2 and 0 . 3 for mutants with 30 , 10 and 3 CO's respectively , using a linear regression for the fraction of CO's in total events from tetrad sequencing data . To determine the fraction of “complex events” ( i . e . , NCOs that appear as two COs and GC associated with a CO on a chromatid not involving the CO ) , the average proportion across WT , HI and LO was used . Complex NCO events are 14% of total NCO events and each event consists of two distinct tracts . Similarly , 7% of total CO's include a conversion tract on a third chromatid . In analyses involving the absence of homeostasis , the number of NCO events was kept constant at the same fraction of total events as in WT . Means and standard deviations of log normal CO and NCO distributions for simulated mutants were first inferred using linear models lm ( mean ∼ #CO ) and lm ( sd ∼ #CO ) . Given that the CO association at ARG4 from in silico analysis did not follow the trend seen in genetic assays due to the overproduction of prototrophs from CO tracts , we performed non-linear regressions to increase the means of CO curves and obtained overall better fit compared to a simple linear regression . Specifically , the following non-linear models were used during extrapolation: lm ( mean ∼ √#CO ) , lm ( mean ∼ log ( #CO ) ) and lm ( mean ∼ 1/#CO ) . The latter one provided the best fit to genetic data for CO association and prototroph formation frequency at ARG4 .
|
Most eukaryotes depend on the meiotic division to segregate each pair of chromosomes properly into their gametes . Chromosome segregation mistakes happening during meiosis are responsible for most miscarriages as well as many diseases such as Down's and Kleinfelter's syndromes in humans . Proper chromosome segregation during meiosis depends on efficient and regulated recombination events that link homologous chromosomes prior to the first meiotic division . These linkages are initiated at double-stranded breaks ( DSBs ) in chromosomal DNA by Spo11 and associated proteins . We isolated a valuable new set of SPO11 alleles in yeast with a wide range of Spo11 activity . Genetic analysis and high throughput sequencing of tetrads from these mutants has revealed unexpected features of meiotic recombination . First , Spo11 DSBs likely continue to form throughout a pachytene arrest in cells compromised for Spo11 activity . Second , the number of recombination initiation events in a given meiosis influences the repair outcome of those events . In addition , our results provide support for crossover homeostasis – a phenomenon in which crossovers are disproportionately maintained over other types of repair in the face of a decrease in DSBs .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
High Throughput Sequencing Reveals Alterations in the Recombination Signatures with Diminishing Spo11 Activity
|
The evolution of drug resistant bacteria is a severe public health problem , both in hospitals and in the community . Currently , some countries aim at concentrating highly specialized services in large hospitals in order to improve patient outcomes . Emergent resistant strains often originate in health care facilities , but it is unknown to what extent hospital size affects resistance evolution and the resulting spillover of hospital-associated pathogens to the community . We used two published datasets from the US and Ireland to investigate the effects of hospital size and controlled for several confounders such as antimicrobial usage , sampling frequency , mortality , disinfection and length of stay . The proportion of patients acquiring both sensitive and resistant infections in a hospital strongly correlated with hospital size . Moreover , we observe the same pattern for both the percentage of resistant infections and the increase of hospital-acquired infections over time . One interpretation of this pattern is that chance effects in small hospitals impede the spread of drug-resistance . To investigate to what extent the size distribution of hospitals can directly affect the prevalence of antibiotic resistance , we use a stochastic epidemiological model describing the spread of drug resistance in a hospital setting as well as the interaction between one or several hospitals and the community . We show that the level of drug resistance typically increases with population size: In small hospitals chance effects cause large fluctuations in pathogen population size or even extinctions , both of which impede the acquisition and spread of drug resistance . Finally , we show that indirect transmission via environmental reservoirs can reduce the effect of hospital size because the slow turnover in the environment can prevent extinction of resistant strains . This implies that reducing environmental transmission is especially important in small hospitals , because such a reduction not only reduces overall transmission but might also facilitate the extinction of resistant strains . Overall , our study shows that the distribution of hospital sizes is a crucial factor for the spread of drug resistance .
The last decades have shown that the introduction of an antibiotic agent is almost inevitably followed by the spread of resistance mutations that jeopardize the beneficial effect of this agent [1] , [2] . Because of this process of bacterial adaptation to antibiotics , maintaining the benefits of antibiotic therapy requires a steady development of new drugs or drug classes . Population biological models may contribute to slowing down the required pace of this “drug treadmill” by identifying the factors that determine the adaptability of bacterial populations to antimicrobial treatment [3] . The epidemic spread of antibiotic resistance can be strongly affected by the structure of the human host population[4] . One of the most important instances of such population structure is the interaction between the hospital and the community[4] . These two settings differ with respect to several parameters that are crucial for the spread of antibiotic resistance . While the hospital environment is characterized by small population sizes , high transmission rates , fast turn over and frequent use of antibiotics , the community exhibits comparatively large population sizes , small transmission rates , slow turn-over rates and infrequent use of antibiotics . Hospitals are often the source of emergent resistant strains [5] , but this spread is not unidirectional , as illustrated by outbreaks of community-acquired MRSA in health care facilities [6] . Since the spread of resistance mutations [7] increases with antibiotic usage [8] , the difference in treatment frequencies may explain why hospitals mostly act as source for resistance mutations and the community acts as a sink . The size distribution of hospitals is an important determinant for the population structure generated through the hospital-community interaction . In small hospitals , bacterial population sizes and frequencies are subject to strong stochastic effects and populations may frequently become extinct . It has been empirically shown that resistance levels tend to be lower in small hospitals [9] , [10] , [11] , [12] , [13] . In principle , this can be due to two reasons: On the one hand , small hospitals might be associated with different types of patients and treatments ( i . e . lower antibiotic usage [14] ) , which select for less resistance . On the other hand , small hospital size by itself might hinder bacterial adaptation and thereby reduce resistance levels . The fact that the analysis in [10] controlled for patient characteristics , suggests that , at least in that case , the impact of hospital size was due to the second “intrinsic” mechanism . In this study we use a simple population biological model to analyze the intrinsic effects of hospital size , i . e . we assess to what extent the stochastic effects resulting from small hospital populations may help in alleviating the burden of antibiotic resistance .
Environmental transmission in the hospital can be added to the above model via two additional , deterministic compartments corresponding to the sensitive and resistant strain: the density of bacteria of strain i in the environment , denoted Ei . Bacteria of strain i colonize the environment with a rate NH , i cE and are cleared from the environment at a constant rate rTurn , E . Bacteria of strain i from the environment can in turn infect susceptible patients with a force of infection βE Ei . . Because the total number of bacteria in the environment is presumably very large and the dynamics of the environmental compartment are not directly affected by the fluctuations in the patient population , we assume that this compartment can be adequately described deterministically .
We used an Irish surveillance dataset published by the Health Service Executive Ireland [19] to investigate to what extent the incidence of infections , especially by resistant strains , correlates with hospital size . This dataset contains information from 53 hospitals about both the total number of new infections with S . aureus as well as infections with MRSA ( all positive blood-cultures were recorded ) . Additionally reported quantities were: length of stay , total inpatient antibiotic usage , injectable inpatient antibiotic usage , the usage of hospital-specific antibiotics , consumption of alcohol hand-gel and the frequency of blood cultures per admission . We found a significant correlation between the number of patient days/year ( which is a proxy for hospital size ) , and the rate of both total ( figure 3A ) and resistant ( figure 3C ) S . aureus acquisitions , as well as the percentage of methicillin-resistant isolates among all S . aureus positive blood cultures ( see figure 3E ) . This correlation remained significant even when controlling for all the above-mentioned variables ( figure 3B , D , F ) . In the minimum adequate model chosen on the basis of the Akaike information criterion ( as implemented in the function step ( lm ( ) ) in R ) , hospital size was the parameter which overall had the most significant impact . Unsurprisingly , the amount of overall antibiotic usage also strongly correlated with both absolute and relative resistance levels . Nevertheless , the model fits ( Figure 3 ) suggest that the impact of hospital size on resistance level is at least of similar magnitude than the impact of antibiotics consumption . If smaller hospitals had lower resistance levels due to extinction events , the same should generally be true for the total incidence of hospital-acquired infections ( although to a lesser degree , since the number of total infections is always higher than the one of resistant infections ) . To test our findings obtained with the Irish dataset , we used surveillance data from the Pennsylvania Health Care Cost Containment Council ( PHC4 ) [20] to investigate in how far infection rates are proportional to hospital sizes . In total , 149 Hospitals reported nosocomial infection rates ( i . e . infections that became symptomatic >48h after admission ) quarterly from 2005 to 2007 . Furthermore , mortality rates , mean length of stay and infection rates by disease type ( e . g . bloodstream infection or pneumonia ) were reported once per year . The median infection rate throughout these 12 quarters strongly correlated with hospital size ( see figure 4A ) . This correlation remained very strongly significant even when accounting for potential confounders such as length of stay , mortality or the number of quarters in which electronic surveillance was used ( see figure 4B ) . In the minimum adequate model , the usage of electronic surveillance also had a significant influence on infection rates ( see figure 3B ) . This is presumably because a higher fraction of infections is reported with electronic surveillance . However , even in this analysis the most significant effect ( i . e . smallest p-value ) on resistance levels was due to hospital size . We observed that infection rates and hospital size were significantly correlated in any given year for the total infection rate as well as for the infection rates of most types of infections ( see table 2 ) . Apart from the average level of infection rates , one would also expect that the temporal increase of infection rates would be smaller with frequent extinction events . During the years 2005–2007 , there was a slight overall increase in infection rates ( see figure 4C ) . For each hospital , we fitted the total rate of acquiring infection with a linear model with time and presence of electronic surveillance as explanatory variables . If the change over the three years was not significantly correlated to time , we set this change to zero . Also the change in infection rates was significantly correlated to the logarithm of the hospital size ( see figure 4D ) . We mainly consider two simple settings ( see Figure 2 ) , in which hospitals of equal size are linked to a community , which is either panmictic or strongly subdivided . In the case of a subdivided community , we allow for migration between the sub-communities ( migration rate between 1%/year and 20%/year ) . These two settings can be considered as models of the population structure in urban and rural environments respectively . In either case we assume that the number of hospital beds per inhabitant ( i . e . the fraction between the total hospital size and the community size ) is constant . Specifically , we consider a community of 3*105 individuals with 1000 hospital beds . We find that , if drug use is high in the hospital and low in the community , the level of resistance increases with hospital size ( Figure 5 ) . This effect is even more pronounced if the community is subdivided as well . However , substructure in the community only seems to have a minor impact compared to the population structure of hospitals ( Figure 5 ) . This makes sense intuitively as population sizes in the community are much larger and hence stochastic effects are comparatively weak . The pattern in figure 5 represents the typical situation for a nosocomial pathogen where treatment rates are high in the hospital but low in the community . In the following we generalize this pattern to a broad range of treatment frequencies . We first describe the case of a panmictic community , which as Figure 5 suggests represents the conservative scenario concerning the effect of population size , and then consider the additional effects conferred by substructures in the community .
It has been argued that concentrating highly specialized services in large hospitals will both improve patient outcomes as well as reduce health care costs [24] , [25] , [26] , [27] . Yet , larger hospitals may have their own disadvantages by facilitating the spread of infectious diseases and resistance genes [10] . We used two surveillance datasets from the US [20] and Ireland [19] to illustrate how the incidence of infections and resistance depends on hospital size . In line with the results of Bhavnani et al . [10] , we found that the rates of acquiring both resistant infections and infections in general strongly and significantly correlated with patient-days/year ( hospital size ) . Furthermore , the increase of total infections over time and the ratio of resistant and drug-sensitive infections also significantly correlated with hospital size . Of course , different patient populations in small and large hospitals could also explain such a pattern . Antibiotic consumption and morbidity ( and thereby susceptibility to infections ) might be much higher in larger hospitals caring for more severely ill patients . However , this would not explain the steeper rise of infections in large hospitals . Additionally , the correlations between hospital size and incidence levels persisted even when taking the following potentially confounding factors into account: mortality , length of stay , antibiotic usage , usage of injectable or hospital-specific antibiotics , sampling frequency , or the usage of disinfectants . Both datasets analyzed have both problems and advantages . In the dataset from Ireland , for example , only S . aureus was considered , all positive blood cultures were recorded ( even those which might have been acquired outside the given hospital ) and there are no data on mortality . Conversely , the dataset from the US does not include the drug-sensitivity of the infections , drug usage or disinfection . However , these drawbacks do not overlap , and our results are coherent for both datasets . This makes it somewhat less likely that such confounders are the only reason for the correlation of resistance with hospital size . Furthermore , hospital size had a more significant impact than antibiotic usage on both relative prevalence and absolute incidence rates of methicillin-resistant S . aureus . Reducing antibiotic consumption is a common recommendation for curbing the spread of resistance . The analyzed data suggest that a reduction in hospital size might therefore be a similarly successful intervention . We used a stochastic epidemiological model in order to describe both the spread of drug resistance in a hospital setting as well as the interaction between one or several hospitals and the community . Like all theoretical models , this is a very simplified description of complex interactions . For example , the epidemiological fitness costs of resistance are inherently difficult to determine . We used estimates of the epidemiological costs of resistance in tuberculosis [28] and of glycopeptide resistance in farm animals [29] as well as in vitro data [30] as guidelines , because no estimates on the transmissibility of resistant nosocomial infections exist , to our knowledge . Additionally , the model neglects co-infection , whereas in reality patients may harbor both resistant and sensitive strains , such that sufficiently long after discontinuation of antibiotic therapy the sensitive strains could dominate again . This effect can , however , be captured in our model by increasing the back-mutation rate ( i . e . the rate with which a patient that is predominantly infected with the resistant strain becomes dominated by the sensitive strain in the absence of therapy ) . We find that increasing this rate substantially increases the magnitude of the effect of hospital size ( results not shown ) . Intuitively , this is the case because with a higher reversion rate the level of resistance in the community ( the “sink” ) declines . Therefore it becomes less likely that the resistant strain is reintroduced from the community once it becomes extinct in a hospital . Thus , the assumption made here leads to rather conservative estimates for the effect of hospital size . In summary , the results from our theoretical model should be regarded as qualitative descriptions , not as quantitative predictions on how much resistance could be reduced if hospitals were smaller . Theoretically , the impact of hospital size on the evolution of antibiotic resistance can be explained by the meta-population dynamics characterized by local extinctions and re-colonizations . According to this interpretation , the beneficial effect of small hospital size is the result of a simple evolutionary mechanism: Selection typically acts in different directions in the hospital and in the community: resistance is selected for in the hospital and selected against in the community . Thus any mechanism , which weakens the effectiveness of selection in the hospital relative to that in community , will lead to an increase in the level of resistance . According to Fisher's fundamental theorem , the effectiveness of selection acting on a trait ( here trait = resistance ) is proportional to the variance in that trait . Small hospital size decreases the variance in resistance through frequent extinctions of the resistance-conferring allele . In other words , if the resistance-conferring allele is extinct in a given hospital , selection ( which would potentially favor this allele ) is completely ineffective because the population consists only of the sensitive strain and selection can only favor one strain over the other if they coexist . In addition to the hospital-community setting described here , many more instances of population structure might cause similar meta-population effects that hinder the epidemic spread of antibiotic resistance mutations . Indeed , we made the same observations when considering inter-ward transfer . Further examples include: household structure , caring facilities , schools etc . Thus the effects described here are likely to extend beyond the hospital-community setting . In this context it should be noted that for a real hospital of a given size , the magnitude of stochastic effects will typically be considerably stronger than what might be expected from a simple population-biological model like the one presented here . The reason for this is that several processes such as population structure , fluctuating population sizes , and variability in transmission rates ( e . g . super-spreaders ) strongly enhance stochasticity . Thus , a real hospital with a given number of patients will typically behave like an idealized hospital , which is much smaller . ( This type of problem is well known in population genetics [31] , and as a consequence populations are often characterized by a so-called effective population size rather than by their census population size ) . Nevertheless , all other things being equal , a small hospital will be subject to stronger stochastic effects than a large hospital and will hence suffer less from antibiotic resistance . In contrast to the typical beneficial effect of small hospital size , we also found that under some circumstances small hospitals may increase the prevalence of antibiotic resistance . This effect occurs if antibiotic usage in the community reaches similar magnitude than in the hospital . Such a setting might be uncommon for nosocomial infections , which are mainly treated in the hospital . However , it might occur for other infections , which occur frequently in both the hospital and the community ( e . g . E . Coli causes many opportunistic infections in the hospital and urinary tract infections in the community ) . Overall , the results of this study suggest the general pattern that strong population subdivision in those compartments , where antibiotic usage is high ( typically: the hospital ) , can substantially reduce the spread of antibiotic resistance . Furthermore , we find that this beneficial effect of small hospital-size is substantially reduced if a substantial fraction of infection events is not acquired directly from other patients ( which exhibit a fast turnover in the hospital ) , but , indirectly , via a slowly decaying environmental compartment . This latter point indicates that reducing such environmental transmission might be especially important in small hospitals as this might not only reduce the force of infection but also the burden of antibiotic resistance by promoting the stochastic extinction of resistant pathogen strains .
|
The increasing spread of bacteria , which are resistant to antibiotics , is a serious threat to clinical care . Currently , several countries aim at concentrating highly specialized services in large hospitals in order to improve patient outcomes . However , empirical studies have shown that resistance levels correlate with hospital size . To illustrate this correlation , we analyze two published datasets from the US and Ireland and controlled for antimicrobial usage , disinfection and length of stay . The proportion of patients acquiring both sensitive and resistant infections in hospitals strongly correlated with hospital size . Moreover , we observe the same pattern for both the percentage of resistant infections and the temporal increase of hospital-acquired infections . To investigate to what extent hospital size can directly affect the prevalence of antibiotic resistance , we use mathematical models describing the epidemic spread of resistance in hospitals and the community . We find that small hospitals typically lead to considerably lower resistance levels than large hospitals . However , this beneficial effect of small hospital size may be reduced if bacteria are transmitted indirectly via the environment . Therefore , reducing environmental transmission might be particularly important in small hospitals . Overall , our findings suggest that the short-term benefits of larger hospitals may come at the price of increasing resistance in the long term .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases",
"infectious",
"diseases/antimicrobials",
"and",
"drug",
"resistance"
] |
2011
|
On Being the Right Size: The Impact of Population Size and Stochastic Effects on the Evolution of Drug Resistance in Hospitals and the Community
|
The β1i , β2i and β5i immunoproteasome subunits have an important role in defining the repertoire of MHC class I-restricted epitopes . However , the impact of combined deficiency of the three immunoproteasome subunits in the development of protective immunity to intracellular pathogens has not been investigated . Here , we demonstrate that immunoproteasomes play a key role in host resistance and genetic vaccination-induced protection against the human pathogen Trypanosoma cruzi ( the causative agent of Chagas disease ) , immunity to which is dependent on CD8+ T cells and IFN-γ ( the classical immunoproteasome inducer ) . We observed that infection with T . cruzi triggers the transcription of immunoproteasome genes , both in mice and humans . Importantly , genetically vaccinated or T . cruzi-infected β1i , β2i and β5i triple knockout ( TKO ) mice presented significantly lower frequencies and numbers of splenic CD8+ effector T cells ( CD8+CD44highCD62Llow ) specific for the previously characterized immunodominant ( VNHRFTLV ) H-2Kb-restricted T . cruzi epitope . Not only the quantity , but also the quality of parasite-specific CD8+ T cell responses was altered in TKO mice . Hence , the frequency of double-positive ( IFN-γ+/TNF+ ) or single-positive ( IFN-γ+ ) cells specific for the H-2Kb-restricted immunodominant as well as subdominant T . cruzi epitopes were higher in WT mice , whereas TNF single-positive cells prevailed among CD8+ T cells from TKO mice . Contrasting with their WT counterparts , TKO animals were also lethally susceptible to T . cruzi challenge , even after an otherwise protective vaccination with DNA and adenoviral vectors . We conclude that the immunoproteasome subunits are key determinants in host resistance to T . cruzi infection by influencing both the magnitude and quality of CD8+ T cell responses .
CD8+ T cells are important mediators of pathogen control during intracellular infections . Sufficient induction of these cells leads to pathogen elimination [1–8] , whereas weak or exacerbated CD8+ T cell stimulation may lead to pathology [9–17] . Therefore , the proper induction of CD8+ T cells must be tightly regulated and may be co-opted in the development of new vaccines against intracellular pathogens [18–22] . Critical in the process of CD8+ T cell induction is the kinetics and efficiency of the provision of MHC class I-restricted epitopes recognized by these lymphocytes , which is linked to the degradation of mature proteins and defective ribosomal products in the cytosol by barrel-shaped structures denoted proteasomes , as recently reviewed [23 , 24] . The catalytic activity of proteasomes is attributed to 3 subunits ( β1 ( Psmb6 ) , β2 ( Psmb7 ) , and β5 ( Psmb5 ) ) located in each of the two inner β rings of the 20S core . In addition , proteasomes featuring the alternative catalytic subunits β1i ( LMP2 or Psmb9 ) , β2i ( MECL1 or Psmb10 ) and β5i ( LMP7 or Psmb8 ) are named immunoproteasomes , which are constitutively expressed in some hematopoietic cells and may be induced by inflammatory stimuli such as IFN-γ , IFN-β or TNF in other cell types ( reviewed in [25] ) . The immunoproteasomes enhance the quantity and diversity of MHC class I-restricted peptides generated and their consequent impact on the magnitude and breath of protective responses of CD8+ T cells against intracellular pathogens has long been studied by several groups . However , the β1i , β2i and β5i proteins have a redundant role , and only partial phenotype is observed in mice lacking a single functional gene encoding one of the immunoproteasome subunits . Only recently a mouse concomitantly defective of all three immunoproteasome genes ( TKO mouse ) was made available , allowing the observation that immunoproteasomes are more relevant for the repertoire of MHC class I-presented peptides than thought before , so much so that WT splenocytes are rejected by TKO mice [26] . However , the participation of immunoproteasomes in the control of infections by most pathogens studied so far was only incremental . Here , we evaluated the role of immunoproteasomes in the CD8+ T cell-mediated immunity , resistance to infection , and protective immunity conferred by genetic vaccination against Trypanosoma cruzi , a human protozoan parasite and causative agent of Chagas disease . Because control of infection by T . cruzi is critically dependent on CD8+ T cells and IFN-γ ( reviewed in [27] ) , we reasoned that the immunoproteasome could be specially relevant for protection in this model . Confirming our hypothesis , we found that CD8+ T cell immune responses to T . cruzi epitopes were remarkably weaker in TKO mice infected with T . cruzi or immunized with an adenoviral vaccine vector expressing an immunodominant parasite antigen . Not only the quantity , but also the quality of parasite-specific CD8+ T cell responses was altered in TKO mice , as indicated by the higher frequency of double-positive ( IFN-γ+/TNF+ ) or single-positive ( IFN-γ+ ) cells specific for immunodominant as well as subdominant T . cruzi epitopes in WT versus TKO mice . Finally , another highly relevant finding was that both naïve and vaccinated TKO mice were extremely susceptible to experimental infection , with most animals succumbing to an otherwise non-lethal challenge . These observations establish that immunoproteasomes play a critical role in the generation of immunogenic peptides and the development of protective T . cruzi-specific CD8+ T lymphocytes .
Because dendritic cells constitutively express immunoproteasomes , we examined whether in vitro-generated bone marrow-derived dendritic cells ( BMDC ) from TKO mice differed from WT BMDC in antigen presentation capacity upon exposure to T . cruzi trypomastigotes or particles of the adenoviral vaccine vector expressing the immunodominant T . cruzi antigen ASP-2 ( AdASP-2 ) [28] . Upon stimulation with parasites or adenovirus , we observed that WT and TKO BMDCs upregulated the costimulatory marker CD86 equally well in vitro , whereas the expression of H-2Kb molecules by TKO BMDC was lower than by their WT counterparts ( Fig 1a and 1b ) . In addition , IL-12 p70 concentrations in the culture supernatants were similar between WT BMDC and TKO BMDC ( Fig 1c ) . BMDC were then co-cultured with purified CD4+ or CD8+ T cells collected from T . cruzi-infected mice . Previously , we have confirmed the cytosolic processing of H-2-restricted epitopes from T . cruzi by incubating these purified T cells in vitro with T . cruzi- or AdASP-2-exposed BMDC deficient in TAP-1 or treated with the proteasome inhibitor epoxomicin ( S1 Fig ) . Consistent with the result showing lower MHC class I expression , TKO BMDC exposed to T . cruzi or AdASP-2 stimulated significantly fewer IFN-γ-producing CD8+ T cells than WT BMDC did , but no difference was observed when BMDC were incubated with synthetic VNHRFTLV peptide corresponding to the immunodominant H-2Kb-restricted epitope from ASP-2 , or the ANYKFTLV and ANYDFTLV peptides that correspond to the respective subdominant epitopes , thus indicating that the reduction in antigen presentation capacity from TKO BMDC is due to the impaired processing of MHC class I-restricted epitopes ( p<0 . 001 , Fig 1d ) . Conversely , we observed that TKO and WT BMDC exposed to T . cruzi or AdASP-2 were equally able to present MHC class II-restricted epitopes in vitro , as measured by their capacity to stimulate similar numbers of IFN-γ-producing CD4+ T cells ( Fig 1e ) . When T . cruzi- or AdASP-2-exposed BMDC were co-cultured with CD4+ or CD8+ T cells isolated from naïve mice , no IFN-γ secretion was detected . These results thus suggested the contribution of immunoproteasomes for the processing of MHC class I-restricted T . cruzi epitopes delivered by the parasite itself or by an adenoviral vaccine vector . To further investigate the participation of immunoproteasomes in the response to T . cruzi infection in vivo , we performed semi-quantitative real-time PCR with primers specific to β1 , β2 , β5 and the respective surrogate β1i , β2i , and β5i genes using cDNA obtained from heart samples of naïve and T . cruzi-infected mice 12 dpi . Additionally , the corresponding human mRNAs were quantified in heart samples from healthy and chronic chagasic patients with cardiomyopathy . These experiments demonstrated that in vivo infection with T . cruzi induces the transcription of immunoproteasome genes in both mice and humans ( Fig 2 ) . Following experimental infection , we evaluated the expression of MHC molecules by splenic antigen-presenting cells ( CD11c+ I-Ab+ CD3- CD19- ) from naïve animals and from mice challenged with T . cruzi twenty days earlier . As expected , T . cruzi infection induced the upregulation of MHC class I in both WT and TKO cells ( p<0 . 001 in both cases ) , suggesting the participation of the conventional immunoproteasome catalytic subunits in the processing of H-2Kb-restricted epitopes . Nevertheless , the expression of H-2Kb molecules on the surface of TKO cells from naïve or infected mice was significantly lower than on their counterpart WT cells ( p<0 . 01 , Fig 3a ) , indicating the contribution of immunoproteasomes to the processing of T . cruzi epitopes . In contrast , the expression of I-Ab molecules on the surface of TKO antigen-presenting cells from naïve or infected mice was similar to their WT counterparts ( Fig 3b ) . Accordingly , naïve and infected TKO mice presented lower numbers of total CD8+ T cells in the spleen in comparison to WT animals , whereas no difference in total CD4+ T cell numbers was observed between WT and TKO mice ( Fig 3c and 3d ) . When results were expressed in cell frequencies , these differences remained exclusive for the CD8+ T cell compartment ( S2 Fig ) . The immune response mediated by CD8+ T cells was evaluated in detail in WT and TKO mice 20 days after infection with T . cruzi . The total numbers of splenic CD8effector cells ( CD8+CD44highCD62Llow ) in infected TKO mice were significantly lower than the corresponding numbers in infected WT animals ( p<0 . 01 , Fig 4b ) . Using pentamer staining , we also measured the numbers of CD8+ T cells specific for the previously characterized immunodominant , H-2Kb-restricted , T . cruzi epitope VNHRFTLV [29 , 30] . Again , these numbers were considerably lower in TKO mice than those observed in infected WT mice ( p<0 . 01 , Fig 4c and 4d ) . We further evaluated the function of CD8+ T cells through their pattern of IFN-γ/TNF production as assessed by ex vivo restimulation of splenocytes with the peptides VNHRFTLV , ANYKFTLV , and ANYDFTLV ( corresponding to T . cruzi H-2Kb-restricted epitopes ) followed by intracellular staining . In presence of the two former peptides , higher numbers of CD8+ T cells from infected WT mice produced IFN-γ and/or TNF in comparison to the infected TKO animals , whereas naïve mice did not respond regardless of their background ( p<0 . 0001 , Fig 4e and 4f ) . We also stimulated splenocytes with the peptides VNYDFTLV and ANYNFTLV , which have similar sequences to the described H-2Kb-restricted epitopes from T . cruzi , in order to test whether they became immunogenic in TKO animals . However , no cytokine response was observed ( Fig 4f ) . These results were similar when expressed in cell frequencies and were further confirmed by estimating the numbers of IFN-γ-producing CD8+ T cells by ELISPOT assay after incubation with each peptide ( S3 Fig ) . The specific response to the subdominant epitope ANYDFTLV elicited by T . cruzi infection was higher in WT mice compared with TKO mice; however , statistical significance was reached only in the ELISPOT assay ( S3 Fig ) . Not only the quantity , but also the quality of specific CD8+ T cell cytokine response was altered in TKO animals . Among the cells that produced any cytokine , the frequencies of double-positive ( IFN-γ+/TNF+ ) or single-positive ( IFN-γ+ ) cells after restimulation with VNHRFTLV peptide were higher in infected WT mice compared with cells from infected TKO mice ( p<0 . 01 , Fig 4g ) , whereas TNF single-positive cells prevailed among CD8+ T cells from TKO mice ( Fig 4g ) . We also compared the frequencies of splenic CD4effector ( CD4+CD44highCD62Llow ) cells in mice infected 20 days earlier , and we also observed that upon infection , CD4+ T cells expressing IFN-γ and/or TNF can be detected without ex vivo restimulation , as previously reported [31] . No difference was observed in the effector phenotype or function of CD4+ T cells between WT and TKO mice infected with T . cruzi ( S4 Fig ) . Based on the experiments above , we concluded that following infection with T . cruzi , the generation of specific CD8+ T cells was severely impaired and profile of cytokine production altered in TKO mice . To test whether the impaired immunity of CD8+ T cells in TKO mice correlates with reduced resistance to infection with T . cruzi , we further estimated the amount of parasite DNA in infected WT and TKO mice . As shown in Fig 5 , in the heart and spleen , the quantity of T . cruzi DNA was significantly higher in TKO mice as compared with WT animals ( p<0 . 0001 ) . Given that CD8+ T cells are critical for immunity against T . cruzi infection , we developed an immunization regimen that successfully vaccinates highly susceptible mice against systemic lethal infection [32–34] . For that purpose , we used recombinant plasmid DNA for priming and human replication-defective recombinant adenovirus type 5 for boost , both vectors expressing ASP-2 of T . cruzi . This vaccination protocol elicited a long-lived protective immune response mediated by CD8+ effector and effector memory T cells [35 , 36] . When we compared the response of CD8+ T cells , we found significantly lower numbers of CD8effector cells in ASP-2-vaccinated TKO mice than in ASP-2-vaccinated WT animals ( p<0 . 05 , Fig 6b ) . As in the case of T . cruzi-infected mice , the numbers of cells specific for the VNHRFTLV epitope among the splenic CD8+ T cells of ASP-2-vaccinated TKO mice were significantly lower than those observed in ASP-2-vaccinated WT mice ( p<0 . 0001 , Fig 6c and 6d ) . In addition , the numbers of specific CD8+ T cells producing IFN-γ and/or TNF upon ex vivo restimulation with the peptide VNHRFTLV were also lower in the population of splenic CD8+ T cells from ASP-2-vaccinated TKO mice compared with cells from ASP-2-vaccinated WT mice ( p<0 . 0001 , Fig 6e and 6f ) . The quality of the immune response of the CD8+ T cells from ASP-2-vaccinated WT mice in comparison to TKO animals was not as different as observed in T . cruzi-infected mice , and IFN-γ single-positive or IFN-γ/TNF double-positive cells predominated in both WT and TKO animals ( Fig 6g ) . When expressed in cell frequencies , or when the cytokine response was assessed by ELISPOT , similar differences between WT and TKO mice were found ( S5 Fig ) . Moreover , the response of CD4+ T cells to genetic vaccination with Asp-2 was similar between WT and TKO animals , as measured by in vivo incorporation of BrdU in CD44high cells or intracelllular staining of IFN-γ after ex-vivo restimulation with AdASP-2-infected cells ( S6 Fig ) . Overall , we concluded that following genetic immunization or infection with T . cruzi , TKO mice were severely impaired in the generation of CD8+ T cell-mediated immune responses to the VNHRFTLV epitope . The ultimate aim of our study was to determine whether immunoproteasomes are important for resistance against infection with T . cruzi . Because MHC I-restricted CD8+ T cells have been described as important for protective immunity in both naïve and vaccinated mice , as estimated based on parasitemia and mouse survival , we expected that TKO mice would be more susceptible to infection than their WT counterparts . Accordingly , we observed that after challenge , TKO mice vaccinated with the βgal unrelated control or with ASP-2 presented levels of parasitemia that were about one order of magnitude higher than those in WT mice vaccinated with the βgal control ( p<0 . 01 , Fig 7a ) . Notably , the parasitemia detected in TKO mice previously immunized with ASP-2 was indistinguishable from that observed among TKO mice that had received the unrelated βgal-expressing vector . In contrast , and as previously described [32] , the parasitemia observed after challenge of ASP-2-vaccinated WT mice was significantly lower than that measured among βgal-vaccinated control animals ( p<0 . 01 , Fig 7a ) . Not only did TKO mice have higher parasitemia , but most of them also succumbed to an otherwise non-lethal infection . A total of 87 . 5% of TKO mice vaccinated with the βgal control succumbed before 45 days after infection . ASP-2-vaccinated TKO mice survived slightly longer , but still , 85 . 7% of the mice died before day 50 after challenge ( Fig 7b ) . The increase in survival was statistically significant in the groups of WT mice compared with TKO mice ( p<0 . 001 ) . These results support the association between the decrease in the frequency of specific splenic CD8+ T cells and the limited parasite control in TKO mice . Because different strains of T . cruzi may present distinct patterns of infectivity in mice , we also tested whether TKO mice were more susceptible to infection with parasites of the CL strain . Similar to the case of mice infected with parasites of the Y strain , we observed that TKO mice were highly susceptible to infection and unable to control their parasitemia , succumbing before 25 days following infection with an otherwise non-lethal challenge ( S7 Fig ) . An altered CD8+ T cell repertoire has been previously described in TKO mice in comparison to WT animals [26] . This was explained by a different repertoire of immunogenic peptides presented by thymic epithelial cells from TKO mice . To test whether a difference in the T cell repertoire accounted for most of the defective immune response observed in TKO mice , we generated bone marrow chimeras . Irradiated WT mice were reconstituted with bone marrow from either WT ( WT-WT ) or TKO ( TKO-WT ) animals and after 8 weeks these mice were infected with T . cruzi . After 20 days of infection , the response of CD8+ T cells was assessed ( Fig 8a ) . In this set up , bone marrow-derived antigen presenting cells lack immunoproteasomes in TKO-WT chimeras , as indicated by the lower expression of H-2Kb molecules on CD11c+ splenic cells from these animals in comparison to WT-WT mice ( Fig 8b ) . Conversely , epithelial thymic stromal cells are WT in both WT-WT and TKO-WT chimeras , thus allowing similar selection of CD8+ T cells . Similar CD8+ T cell repertoire between WT-WT and TKO-WT chimeras was inferred by the indistinguishable staining of CD8+ T cells with TCR Vβ antibodies panel ( Fig 8c ) . Upon T . cruzi infection , the chimeric TKO-WT animals presented lower numbers of VNHRFTLV-specific CD8+ T cells stained with pentamers ( p<0 . 05 , Fig 8d and 8e ) and lower numbers of cytokine-producing CD8+ T cells specific to VNHRFTLV and ANYKFTLV epitopes ( p<0 . 001 , Fig 8f and 8g ) , although the response to the subdominant epitope ANYDFTLV was comparable between WT-WT and TKO-WT chimeras ( Fig 8f and 8g ) . Once again , the frequency of double-positive ( IFN-γ+/TNF+ ) or single-positive ( IFN-γ+ ) CD8+ T cells after restimulation with VNHRFTLV peptide was higher in infected WT-WT chimeric mice compared with cells from infected TKO-WT mice ( p<0 . 01 , Fig 8h ) . Moreover , WT-WT and TKO-WT chimeras were vaccinated with AdASP-2 and the specific CD8+ T cell response was evaluated after 20 days ( Fig 9a ) . Again , splenic CD11c+ cells from TKO-WT mice presented lower expression of H-2Kb ( Fig 9b ) , whereas the staining of CD8+ T cells with TCR Vβ antibodies panel was similar between TKO-WT and WT-WT animals ( Fig 9c ) . Despite homogenous repertoire of CD8+ T cells , the number and cytokine effector function of VNHRFTLV-specific CD8+ T cells was significantly reduced in AdASP-2-vaccinated TKO-WT mice in comparison to AdASP-2-vaccinated WT-WT mice ( Fig 9d–9h ) . Because even WT-WT chimeric mice became highly susceptible to T . cruzi challenge , succumbing to infection from 22 days after challenge , comparisons between WT-WT and TKO-WT animals in terms of resistance to infection and vaccine-induced protection were not possible . Nonetheless , these experiments sustain the notion that the impaired immunity of CD8+ T cells in T . cruzi-infected or genetically vaccinated animals lacking all immunoproteasome genes could not be solely attributed to an altered T cell repertoire .
The recognition by CD8+ T lymphocytes of antigens displayed on the context of MHC class I molecules is a fundamental requirement for the control of tumors and intracellular infections by viruses , fungi , bacteria and parasites . Proteasomes are of paramount importance in the processing of antigens in the MHC class I pathway . However , the specific contribution of alternative proteasome catalytic subunits β1i , β2i and β5i ( as opposed to the canonical active sites ) to the generation of MHC class I-restricted epitope repertoire was thought to be , at most , incremental until the recent development of a mouse model simultaneously devoid of β1i , β2i and β5i subunits ( TKO mice ) [26] . In that study , major differences in the quantity and quality of epitopes recognized by CD8+ T cells was reported , suggesting that models where only one or two subunits are targeted might underestimate the relevance of immunoproteasomes in antigen processing and presentation . Although it is reasonable to assume that immunoproteasome’s role in epitope abundance and specificity may lead to altered resistance to a pathogen , supportive data to this assumption remain relatively scarce . Here we explored the TKO mouse model to further investigate the role of immunoproteasomes in the development of protective immunity to a pathogen . Trypanosoma cruzi , a neglected human parasite , was of didactical use as a model , since protective immunity against it is highly dependent on CD8+ T cell function and IFN-γ ( the major inducer of immunoproteasomes ) . The TKO mice exhibited a drastically reduced response of CD8+ T cells specific for immunodominant and subdominant epitopes after T . cruzi infection , whereas the response of CD4+ T cells was unaltered . Surprisingly , natural resistance to infection in the absence of immunoproteasomes was seriously compromised and even more unexpectedly the protection induced by genetic vaccination was completely abolished , indicating that the immunoproteasome catalytic subunits , rather than the conventional proteolytical sites , are essential for the processing of T . cruzi epitopes related to protective immunity generated during infection or vaccination . Our observations from chimeric mice are consistent with the hypothesis that the defect in TKO mice is the inability of antigen presenting cells to process protective epitopes to CD8+ T cells , rather than the variation of naïve CD8+ T cell repertoire . Also , we were unable to detect any shift in the immunodominance pattern of TKO mice relatively to WT counterparts . We have previously reported that the induction of CD8+ T cell responses to T . cruzi is remarkably delayed and suboptimal , with most cells expressing high levels of the proapoptotic marker CD95 [16] . Conversely , genetic vaccination with adenoviral vectors induces a rapid , expanded , polyfunctional and highly viable CD8+ T cell response that differentiates into a long-lasting memory pool , as reported for T . cruzi as well as other pathogens [33–36] . Here , we report that the function of T . cruzi-specific CD8+ T cell responses was altered in the absence of immunoproteasomes , as indicated by the higher frequency of TNF single-positive cells in TKO animals as opposed to a prevailing compartment of IFN-γ single-positive or IFN-γ and TNF double-positive cells in immunoproteasome-sufficient mice . This difference , however , was more pronounced during infection than upon genetic vaccination with Asp-2 . It is acknowledged that IFN-γ is of major importance for control of T . cruzi infection and its induction is tightly regulated , whereas TNF , expressed throughout the course of infection , is of secondary importance to parasite control [13 , 27 , 37] . Our results may suggest that in our model the threshold of antigen presentation required to induce TNF production by CD8+ T cells might be lower than the levels of stimulation necessary to induce IFN-γ . Consistently , the data presented here reinforce the notion that AdASP-2 is remarkably more potent than T . cruzi at inducing CD8+ T cell responses , since even in the absence of immunoproteasomes the viral vector was able to stimulate the production of IFN-γ-secreting cells . Therefore , even in the absence of immunoproteasomes , the antigen pool generated by the adenoviral vaccine could be cleaved by conventional proteasomes without severely compromising the induction of polyfuncional CD8+ T cells ( yet lower number of cells are induced ) , in contrast to the observations from experimental infection . In fact , we have previously observed that CD8+ T cells specific to subdominant epitopes from T . cruzi can be induced by adenoviral vaccination , but not by protozoan infection , in line with the idea that the provision of epitopes is enhanced during genetic vaccination [33] . Presumably , the lower expression of MHC class I in TKO mice is inextricably linked to a defect in generating peptides that bind to H-2 molecules . We found that upon infection with T . cruzi or adenovirus expressing ASP-2 , dendritic cells from TKO background were highly impaired at presenting antigen to WT CD8+ T lymphocytes explanted from either infected or vaccinated mice . These results contrasted with the fact that WT and TKO BMDC loaded with VNHRFTLV , ANYKFTLV , or ANYDFTLV peptides were equally able to stimulate T . cruzi-specific CD8+ T cells in vitro . These results suggest that the defects in antigen presentation capacity of TKO cells are due to the compromised processing of epitopes . These findings are consistent with a previous study indicating that in absence of immunoproteasomes the presentation of protein antigens that need to be cleaved is compromised , whereas no defect is observed when the processed epitope is directly expressed by a minigene [26] . Nonetheless , dendritic cells lacking immunoproteasomes still upregulate MHC class I molecules and present epitopes recognized by CD8+ T cells upon infection with T . cruzi or genetic vaccination with Asp-2 , an effect most likely attributed to the processing of antigen by the canonical catalytic subunits of the proteasome . In addition to the poor antigen-presenting function of TKO cells , the immunodominance of epitopes could have been shifted due to the absence of immunoproteasomes , as previously described [38–40] . Although we only tested peptides representing 3 known epitopes and 2 other hypothetical ones , we could not find relevant changes in the immunodominance pattern . Independently of their specificities or hierarchies , the CD8+ T cells activated in TKO mice were always reduced in comparison to WT counterparts . An altered CD8+ T cell repertoire has been previously described in TKO mice in comparison to WT animals [26] . Hence , distinct naïve T cell repertoires could also explain the discrepancies in CD8+ T cell response between TKO and WT mice [26 , 41] . It is assumed that this difference in the repertoire of CD8+ T cells from WT and TKO mice is due to MHC I-mediated presentation of a different repertoire of peptides by epithelial cells from thymus during T cell development . To address this issue we performed experiments using chimeras , where recipient WT mice received bone marrows from either WT or TKO mice . The results obtained from these experiments suggest that it is unlikely that a difference in the T cell repertoire accounted for most of the limited immune response observed in TKO mice . In the same direction , the transfer into TKO mice of WT CD8+ T cells or of the P14 transgenic CD8+ T cell clone specific to GP33 epitope was unable to restore the capacity of TKO animals to optimally respond to LCMV infection [26] . Not only did TKO mice have diminished CD8+ T cell immune responses to parasite-derived epitopes , but they were also no longer resistant to infection with T . cruzi . It is reasonable to hypothesize that the susceptibility to infection and the weak CD8+ T cell-mediated immunity are linked because in the mouse models that we used ( naïve and vaccinated ) , CD8+ T cells have been described as being critically involved in the control of parasitemia and in survival [30 , 32] . Nevertheless , in addition to the low numbers of specific CD8+ T cells , other factor ( s ) may also account for the extreme susceptibility that we observed . For instance , a second likely source of reduced resistance may stem from the effector phase of the immune response . During T . cruzi infection , large amounts of IFN-γ are produced and can be detected in the serum [42] . As immunoproteasomes become the dominant form of proteasomes in IFN-γ-activated cells , their absence may drastically reduce the antigen presentation capacity of target cells infected with T . cruzi . This aspect may further contribute to the impaired efficacy of the few CD8+ T cells generated during the priming phase . It is worth mentioning that the susceptibility of TKO mice occurred in parallel to undetectable changes in presentation of MHC II-restricted epitopes to CD4+ T cells and in the generation of CD4effector cells during infection . These findings are also in agreement with previous observations that CD4+ T cell-mediated immune responses were similar in TKO and WT mice in different infection models [26] . The susceptibility of TKO mice may also be linked to a larger inflammatory reaction and to more severe myocardial tissue damage . Recently , immunoproteasomes have been described as protecting the heart from excessive inflammatory tissue damage due to acute coxsackievirus B3 ( CVB3 ) -induced myocarditis [43] . Because T . cruzi may infect myocardial cells , whether a similar event is associated with susceptibility to infection should be further investigated . This topic may not be as important in the model of infection used in most of our studies , considering that infection with parasites of the Y strain targets the spleen as well as the heart tissues [44] . However , other T . cruzi strains , such as the CL or the Colombian strain , cause severe acute heart muscle injury and deserve further investigation [45] . Previous studies using mice genetically deficient for a single immunoproteasome subunit ( LMP7 ) also described poor induction of CD8+ T cells , but not CD4+ T cells , specific for antigens of Toxoplasma gondii . As in our system , these mice were susceptible to an otherwise non-lethal infection [46] . These results further corroborate the interpretation that immunoproteasomes are critical for the generation of intracellular parasite epitopes used for CD8+ T cell activation . Additionally , in the absence of immunoproteasomes , mouse survival after infection with T . cruzi or T . gondii is compromised , pointing to the immunoproteasomes as key mediators of resistance to intracellular parasite infections . In other words , these protozoan infection models were instrumental at evidencing a specific role of immunoproteasomes for the generation of critical epitopes required for protective immunity . By employing adenoviral vectors , our group developed genetic vaccination strategies against T . cruzi . The immunization regimens confer protection to wild-type mice of different strains , including the C57BL/6 , BALB/c and A/Sn [16 , 28] . From these , the A/Sn mice are the most susceptible and succumb to infection in less than 30 days , even when challenged with as low as 150 parasites of T . cruzi Y strain . Conversely , A/Sn mice survive T . cruzi infection if immunized with AdASP-2 even on the same day of challenge [16] . At least in our animal facility , the C57BL/6 mouse lineage is more resistant to T . cruzi infection and survives challenge with high doses ( 10 , 000 ) of blood forms of T . cruzi . Therefore , the C57BL/6 mouse serves as a model to study different aspects of the CD8+ T cell response related to resistance . Here , we observed that mice devoid of immunoproteasomes in a C57BL/6 background are highly susceptible to T . cruzi challenge . Unexpectedly , vaccination of these mice with AdASP-2 does not have any effect in protecting the mice upon challenge , whereas in normal C57BL/6 mice the parasitemia is reduced in about one order of magnitude . These results thus indicate that the efficacy of our vaccination regimen is extremely dependent on the expression of β1i , β2i and β5i subunits of the immunoproteasome , rather than the conventional proteolytical sites . In accordance with this finding , a recent clinical trial aimed at testing the efficacy of the RTS , S vaccine against malaria identified the upregulation of immunoproteasome genes among protected individuals after challenge [47] . Altogether , these results may point to the induction of immunoproteasome genes as pivotal targets to be considered in the design of successful vaccination strategies aimed at inducing CD8+ T cells . In conclusion , we report that immunoproteasomes , rather than canonical proteasomes , have a potentially underestimated role in inducing protective immunity both in primary infection as well as genetic vaccination against a human pathogen .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Brazilian National Council of Animal Experimentation ( http://www . cobea . org . br/ ) . The protocol was approved by the Committee on the Ethics of Animal Experiments of the Institutional Animal Care and Use Committee at the Federal University of Sao Paulo ( Id # CEP 0426/09 ) . The protocol using human samples was approved by the Institutional Review Board of the University of São Paulo School of Medicine ( Protocol number 739/2005 ) and written informed consent was obtained from the patients . In the case of samples from heart donors , written informed consent was obtained from their families . Myocardial left ventricular free wall heart samples were obtained from end-stage heart failure chronic chagasic patients with cardiomyopathy ( 7 females , 7 males , 15–61 years old ) . Control adult heart tissue from the left ventricular-free wall was obtained from nonfailing donor hearts not used for cardiac transplantation due to size mismatch with available recipients ( males , 17–46 years old ) . Hearts were explanted at the time of heart transplantation at the Heart Institute—InCor , University of São Paulo School of Medicine , São Paulo , SP , Brazil . For mRNA extraction , samples were quickly dissected , and myocardial tissue was frozen in liquid nitrogen and stored at -80°C . Five- to 8-week-old female C57BL/6 mice were purchased from CEDEME ( Federal University of São Paulo ) . TKO mice were generated as described by Kincaid et al . [26] and were bred in our own animal facility . Bloodstream trypomastigotes of the Y or CL strain of T . cruzi were obtained from mice infected 7 or 15 days earlier . The concentration of parasites was estimated and adjusted to 105 parasites/mL . Each mouse was inoculated with 104 trypomastigotes diluted in 0 . 1 mL PBS administered subcutaneously ( s . c . ) at the base of the tail . Parasitemia was assessed on the days indicated in each figure , which involved counting the number of parasites per 5 μL blood . The plasmid pIgSP Cl . 9 and AdASP-2 were generated , grown and purified as described previously [28 , 32] . Control mice were immunized with pcDNA3 and human replication-deficient adenovirus type 5 expressing βgal ( Adβ-gal ) . The mice were inoculated intramuscularly ( i . m . ) with 50 μg plasmid DNA into each tibialis anterioris muscle . A total of 21 days later , these mice received 50 μL of a viral suspension containing 2 X 108 plaque-forming units ( pfu ) of adenovirus via the same locations . Immunological assays were performed on the days indicated in each figure . Synthetic peptides VNHRFTLV , ANYKFTLV , and ANYDFTLV were purchased from GenScript ( Piscataway , NJ ) . The peptide purity was higher than 90% . Peptide identities were confirmed using a Q-Tof Micro equipped with an electrospray ionization source ( Micromass , UK ) . The pentamer H-2Kb-VNHRFTLV was purchased from ProImmune Inc . ( Oxford , UK ) . Progenitor cells were flushed from femurs and cultured in vitro in RPMI 1640 supplemented with 10 mM HEPES , 0 . 2% sodium bicarbonate , 59 mg/L penicillin , 133 mg/L streptomycin , 10% HyClone fetal bovine serum , 2 mM L-glutamine , 1 mM sodium pyruvate , 55 μM 2-mercaptoethanol and 20 ng/mL GM-CSF ( R&D Systems ) at a concentration of 2 X 105 cells/mL . After 4 days in culture , half of the medium volume was replaced with fresh medium . At day 6 , the resulting BMDCs were exposed to tissue-culture trypomastigotes of T . cruzi at a ratio of 3 parasites/cell or to AdASP-2 at a ratio of 50 pfu/cell for an additional 24 h . After 7 days in culture ( and 24 h of T . cruzi or AdASP-2 exposure ) , the BMDCs were employed in in vitro antigen presentation assays . The presence of IL-12p70 in the supernatant of the BMDCs was assessed after 7 days in culture ( and 24 h of T . cruzi or AdASP-2 exposure ) using an ELISA ( BD ) . To perform the in vitro antigen presentation assays , spleen cells from naïve animals or mice infected with T . cruzi 15 days earlier were harvested , and the CD8+ T cell population was isolated through negative selection with a CD8+ T Cell Isolation Kit using MACS beads ( Miltenyi ) according to the manufacturer’s specification , followed by staining with anti-CD8 PerCP ( 53–6 . 7 , BD ) and sorting by FACS ( BD FACSAria II ) . The CD8- fraction obtained after MACS isolation was stained with CD4 PE Cy7 ( GK1 . 5 , BD ) and also sorted by FACS . The isolated lymphocytes were co-cultured with BMDCs previously exposed to T . cruzi or AdASP-2 or left unexposed at a ratio of 1 BMDC to 5 CD8 or CD4 cells for 24 h , and the number of IFN-γ-secreting cells was determined by ELISPOT , as described elsewhere [31] . For flow cytometry analyses , we used mouse splenocytes treated with ACK buffer ( NH4Cl , 0 . 15 M; KHCO3 , 10 mM; Na2-EDTA 0 . 1 mM; pH = 7 . 4 ) for lysing the erythrocytes . Single-cell suspensions were washed in PBS , stained for 10 min at RT with biotinylated MHC I multimer H-2Kb-VNHRFTLV , and stained for 20 min at 4°C with streptavidin-APC and CD8 FITC ( 53–6 . 7 ) . For the analyses of other cell-surface markers , single-cell suspensions from spleens of mice were stained with CD11c APCCy7 ( HL3 ) , CD86 APC ( GL1 ) , CD11b PerCP ( M1/70 ) , CD19 PE ( ID3 ) , CD3 PE ( 17A2 ) , H-2Kb FITC ( AF6-88 . 5 ) , IAb biotinylated ( 25-9-17 ) , streptavidin PECy7 , CD4 PeCy7 ( RM4-5 ) , CD8 Pacific Blue ( 53–6 . 7 ) , CD62L APC ( MEL-14 ) , and CD44 PE ( IM7 ) . Staining of TCR Vβ chains was performed using TCR Vβ Screening Panel Kit ( BD Pharmingen ) . Isotype control antibodies were PE-labeled hamster IgG1k and rat IgG1k , and FITC-labeled IgG2ak . All antibodies and streptavidins were purchased from BD Pharmingen . At least 500 , 000 cells were acquired on a BD FACS Canto II flow cytometer and analyzed with FlowJo 8 . 7 ( Tree Star , Ashland , OR ) . For the intracellular staining ( ICS ) of cytokines ( IFN-γ and TNF ) , splenocytes collected from C57BL/6 or TKO mice were treated with ACK buffer . ICS was performed after in vitro culture of splenocytes in the presence or absence of the peptides indicated in each figure . Cells were washed 3 times in plain RPMI and re-suspended in cell culture medium consisting of RPMI 1640 medium supplemented with 10 mM Hepes , 0 . 2% sodium bicarbonate , 59 mg/L of penicillin , 133 mg/L of streptomycin , 10% Hyclone fetal bovine serum , 2 mM L-glutamine , 1 mM sodium pyruvate , 55 μM 2-mercaptoethanol . The viability of the cells was evaluated using 0 . 2% trypan blue exclusion dye to discriminate between live and dead cells . Cell concentration was adjusted to 5 X 106 cells/mL in cell culture medium containing anti-CD28 ( 2 μg/mL ) , BDGolgiPlug ( 1 μL/mL ) and monensin ( 5 μg/mL ) . In half of the cultures , a final concentration of 10μM of the VNHRFTLV peptide was added . The cells were cultivated in V-bottom 96-well plates ( Corning ) in a final volume of 200 μL in duplicate , at 37°C in a humid environment containing 5% CO2 . After 8h incubation , cells were stained for surface markers with CD4 FITC ( GK1 . 5 ) and CD8 PerCP ( 53–6 . 7 ) on ice for 20 min . To detect IFN-γ and TNF by intracellular staining , cells were then washed twice in buffer containing PBS , 0 . 5% BSA , and 2 mM EDTA , fixed and permeabilized with BD perm/wash buffer . After being washed twice with BD perm/wash buffer , cells were stained for intracellular markers using APC-labeled anti-IFN-γ ( XMG1 . 2 ) and PE-labeled anti-TNF ( MP6-XT22 ) for 20 minutes on ice . Finally , cells were washed twice with BD perm/wash buffer and fixed in 1% PBS-paraformaldehyde . At least 800 , 000 cells were acquired on a BD FACS Canto II flow cytometer and then analyzed with FlowJo . Total RNA was isolated from mice and human myocardial tissues with Trizol ( Invitrogen ) , followed by purification with Quick RNA Miniprep columns and DNAse I treatment ( Zymo Research ) . cDNA was synthesized with High Capacity cDNA Reverse Transcription Kit ( Applied Biosystems ) . To confirm the absence of genomic DNA control samples were used without reverse transcriptase . RT-PCR was performed using Power SYBR PCR Mastermix ( Thermo Scientific ) and StepOne Plus thermo cycler ( Applied Biosystems ) . mRNA levels were normalized to HPRT ( mouse ) and beta-actin ( human ) . Primer sequences are reported elsewhere for murine ( procedure B in [48] ) and human [49] samples . Relative quantification was calculated over naïve ( mouse ) or healthy ( human ) controls using Δ ΔCT method [50] . The protocol for quantification of T . cruzi DNA in heart and spleen samples was performed as described elsewhere [51] . Eight-week-old female C57BL/6 mice were irradiated at 900 Rads . Each irradiated animal received 10 × 10 6 bone marrow cells i . v . isolated from C57BL/6 or TKO female mice . Before transfer , bone marrow cell suspensions were depleted of T cells with CD8a ( Ly-2 ) MicroBeads and CD4 ( L3T4 ) MicroBeads ( Miltenyi Biotec ) . Mice were given medicated water containing sulfamethoxazole ( 200 mg/mL ) and trimethroprim ( 40 mg/mL ) . Experiments were performed 8 weeks after bone marrow transfer . Groups were compared using One Way ANOVA followed by Tukey’s HSD test ( http://faculty . vassar . edu/lowry/VassarStats . html ) . Parasitemia values were log transformed before comparison . The Log-rank ( Mantel-Cox ) test was used to compare mouse survival rates after challenge with T . cruzi ( http://bioinf . wehi . edu . au/software/russell/logrank/ ) . The differences were considered significant when the P value was <0 . 05 .
|
CD8+ t lymphocytes are cells of the immune system that mediate control of intracellular infections by viruses , prokaryote as well as eukaryote pathogens . To confer protection , these lymphocytes need to be elicited by pathogen peptides that are presented in association with MHC class I molecules . The degradation of self and microbial proteins by catalytic domains of the cytosolic proteasome β1 , β2 and β5 subunits is intimately linked to the generation of MHC class I-restricted epitopes , which in turn are important determinants of the kinetics , specificity and efficiency of CD8+ T cell-mediated immunity . Importantly , inflammatory stimuli trigger the expression of the inducible alternative β1i , β2i and β5i subunits that form the immunoproteasomes . The qualitative and quantitative importance of immunoproteasomes in generating CD8+ T cell epitopes has recently been demonstrated in mice that are simultaneously devoid of the β1i , β2i and β5i subunits . In this study , we explored the role of immunoproteasomes in host resistance to Trypanosoma cruzi , a protozoan parasite that causes Chagas disease . We found that β1i , β2i and β5i triply deficient mice have an impaired response of CD8+ T cells and are highly susceptible to primary infection with T . cruzi . We also demonstrated that host resistance induced by a genetic vaccine able to protect normal mice from T . cruzi challenge fails to do so in the immunoproteasome-deficient mice . Our study provides strong evidences that β1i , β2i and β5i immunoproteasome subunits are important determinants of both the magnitude and quality of CD8+ T cell responses as well as immune-mediated host resistance to a human pathogen .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"spleen",
"immunology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"clinical",
"medicine",
"protozoans",
"cytotoxic",
"t",
"cells",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"white",
"blood",
"cells",
"major",
"histocompatibility",
"complex",
"animal",
"cells",
"t",
"cells",
"immune",
"response",
"trypanosoma",
"cruzi",
"cell",
"staining",
"trypanosoma",
"cell",
"biology",
"clinical",
"immunology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2016
|
The Combined Deficiency of Immunoproteasome Subunits Affects Both the Magnitude and Quality of Pathogen- and Genetic Vaccination-Induced CD8+ T Cell Responses to the Human Protozoan Parasite Trypanosoma cruzi
|
Decreased bioavailability of nitric oxide ( NO ) is a major contributor to the pathophysiology of severe falciparum malaria . Tetrahydrobiopterin ( BH4 ) is an enzyme cofactor required for NO synthesis from L-arginine . We hypothesized that systemic levels of BH4 would be decreased in children with cerebral malaria , contributing to low NO bioavailability . In an observational study in Tanzania , we measured urine levels of biopterin in its various redox states ( fully reduced [BH4] and the oxidized metabolites , dihydrobiopterin [BH2] and biopterin [B0] ) in children with uncomplicated malaria ( UM , n = 55 ) , cerebral malaria ( CM , n = 45 ) , non-malaria central nervous system conditions ( NMC , n = 48 ) , and in 111 healthy controls ( HC ) . Median urine BH4 concentration in CM ( 1 . 10 [IQR:0 . 55–2 . 18] μmol/mmol creatinine ) was significantly lower compared to each of the other three groups — UM ( 2 . 10 [IQR:1 . 32–3 . 14];p<0 . 001 ) , NMC ( 1 . 52 [IQR:1 . 01–2 . 71];p = 0 . 002 ) , and HC ( 1 . 60 [IQR:1 . 15–2 . 23];p = 0 . 005 ) . Oxidized biopterins were increased , and the BH4:BH2 ratio markedly decreased in CM . In a multivariate logistic regression model , each Log10-unit decrease in urine BH4 was independently associated with a 3 . 85-fold ( 95% CI:1 . 89–7 . 61 ) increase in odds of CM ( p<0 . 001 ) . Low systemic BH4 levels and increased oxidized biopterins contribute to the low NO bioavailability observed in CM . Adjunctive therapy to regenerate BH4 may have a role in improving NO bioavailability and microvascular perfusion in severe falciparum malaria .
Falciparum malaria causes over 600 , 000 deaths worldwide each year , with approximately 560 , 000 fatal cases annually among children in sub-Saharan Africa [1] . Coma in malaria , cerebral malaria ( CM ) , portends a grave outcome among children infected with Plasmodium falciparum and , despite advances in anti-parasitic drug therapies , still has a 10–20% case fatality rate [2–4] . However , the pathogenesis of CM remains poorly understood [5] . CM pathogenesis studies to date show endothelial dysfunction [6] , endothelial activation [7–10] and cytoadherence of parasitized red blood cell ( pRBC ) to endothelial cells in post-capillary venules , resulting in red blood cell sequestration [11–13] , microvascular congestion and impaired blood flow to tissues [14–16] . Metabolic derangements and cytotoxic mechanisms contributing to the pathogenesis of CM have also been proposed [5 , 17–19] . Low nitric oxide ( NO ) is a key cause and contributor to the microvascular pathophysiology observed in severe malaria [20] . A variety of causes for low bioavailability in malaria have been identified within multiple steps of the NO production pathway [21–23] , from low levels of NO synthase ( NOS ) substrate , arginine [24] , to elevated levels of endogenous NOS inhibitors [25 , 26] . All NOS isoforms require the obligate cofactor tetrahydrobiopterin ( BH4 ) to enzymatically generate NO from L-arginine . The role of this NOS cofactor as a potential contributor to low NO bioavailability in malaria is unknown . In vascular diseases , low BH4 and increased concentrations of its oxidized metabolite , dihydrobiopterin ( BH2 ) , are not only associated with impaired NO synthesis , but also linked to generation of reactive oxygen species within the endothelium [27] . In addition to its role in NO synthesis and endothelial function , BH4 is also an essential cofactor for monooxygenase enzymes required for phenylalanine metabolism ( phenylalanine hydroxylase [PAH] ) as well as biogenic amine neurotransmitter synthesis of catecholamines ( tyrosine hydroxylase ) and serotonin ( tryptophan hydroxylase ) [28 , 29] . ( The reader is directed to references 28 and 29 for reviews of BH4 metabolism , which include diagrams of BH4 synthetic and salvage pathways . ) We have previously reported elevated plasma phenylalanine in children with CM [18] . This finding is likely attributable to impaired activity of hepatic PAH , which regulates plasma phenylalanine levels within a narrow range by controlling the rate of conversion of phenylalanine to tyrosine [30] . We hypothesized that hyperphenylalaninemia is an indicator of systemic BH4 deficiency and that BH4 deficiency would also contribute to impaired NO bioavailability observed in malaria . To address this hypothesis we quantified BH4 and its oxidized metabolites in urine as a measure of systemic biopterin availability [28 , 31–33] in children presenting with CM . Quantifying biopterins in urine requires specific collection methods . With ordinary urine collection and storage , BH4 spontaneously oxidizes to its metabolites , BH2 and , to a lesser extent , fully oxidized biopterin ( B0 ) [34 , 35] . By collecting urine directly into an anti-oxidant/chelator cocktail in the dark followed by immediate freezing and storage at -80°C until analysis , we overcame this potential artifact of ex vivo spontaneous oxidation . This established collection method for biopterin analysis enabled quantification of BH4 in its reduced in vivo state [35] . In this report , we provide quantitative data on urinary excretion of the active NOS cofactor BH4 , as well as BH2 and B0 , in children at the time of presentation with CM . We found decreased levels of BH4 and increased levels of oxidized biopterins in children with CM , indicating a marked reduction in BH4 bioavailability in severe malaria . Based on these findings , we propose a pathogenic mechanism wherein low systemic BH4 contributes to low NO bioavailability and endothelial dysfunction in severe malaria .
From November 2007 to January 2012 we screened 528 children for enrollment into one of four study groups . After exclusions , 66 children with uncomplicated malaria ( UM ) , 52 with cerebral malaria ( CM ) , 53 with non-malaria central nervous system conditions ( NMC ) , and 116 healthy controls ( HC ) were enrolled ( Fig . 1 ) . Baseline characteristics comparing the four groups are shown in Table 1 . Children with CM were significantly older than children from the other groups when compared across all groups and in pairwise comparisons between groups . Two children with CM and one child with NMC had overt renal failure as evidenced by plasma creatinine measurements of 2 . 7 , 5 . 1 , and 2 . 5 mg/dL , respectively . Children with CM had significantly lower peripheral blood parasitemia , but significantly higher plasma P . falciparum histidine-rich protein-2 ( PfHRP-2 ) concentrations compared to children with UM . To assess biopterin status in CM compared to other infectious and non-infectious central nervous system ( CNS ) conditions , we enrolled a comparison group comprised of children presenting to the hospital with NMC for which lumbar puncture was clinically indicated . The NMC group proved to be heterogeneous as expected . Clinical and laboratory investigations indicated that at least 10 of the 53 children with NMC suffered from bacterial or fungal meningitis . No evidence for viral encephalitis , trauma , subarachnoid hemorrhage or metabolic encephalopathy was found . Toxic encephalopathy was possible in one child . Idiopathic seizures and aseptic meningitis were other possible diagnostic categories . Seven children with CM and six children with NMC died while hospitalized . All children with UM recovered . Urine pterin concentrations differed significantly across the four groups ( Table 2 ) . Compared to HC ( n = 111 ) , total biopterin levels ( BH4 + BH2 + B0 ) were increased in UM ( n = 55 ) ( p<0 . 001 ) , CM ( n = 45 ) ( p<0 . 001 ) , and NMC ( n = 48 ) ( p<0 . 001 ) ( Fig . 2 , panel A ) . Urine BH4 concentrations were significantly lower in CM than in each of the other three groups ( p≤0 . 005 for all pairwise comparisons ) ( Fig . 2 , panel B ) . The statistical significance persisted in a linear regression model to control for plasma creatinine ( p = 0 . 021 ) and in a linear regression model to control for age , weight , gender and plasma creatinine ( p = 0 . 038 ) . Using logistic regression , we found no significant co-variation of urine BH4 by presence or absence of pyuria ( OR 1 . 21 [0 . 65–2 . 25]; p = 0 . 56 ) or by presence or absence of bacteriuria ( OR 0 . 70 [0 . 34–1 . 46]; p = 0 . 34 ) . Urine BH2 was significantly higher in CM compared to each of the other three groups ( p<0 . 05 for all comparisons ) . Urine BH4 values differed significantly when compared across UM ( 2 . 10 [IQR 1 . 32–3 . 14] ) , CM survivors ( 1 . 10 [IQR 0 . 48–2 . 25] μmol/mmol creatinine ) and CM fatalities ( 1 . 02 [IQR 0 . 77–2 . 09] μmol/mmol creatinine ) ( p<0 . 001 ) . NOS activity is affected not only by the availability of the cofactor BH4 , but also by the stoichiometric balance of BH4 relative to its major oxidized metabolite BH2 . BH2 competes with BH4 at the NOS binding site and can directly inhibit NOS activity [36] . The BH4:BH2 ratio has been proposed as the critical determinant of NO synthesis by endothelial NOS ( eNOS ) [37] . Accordingly , we analyzed this ratio in the four clinical groups enrolled . The BH4:BH2 ratio was significantly lower in CM compared to each of the other groups ( p<0 . 001 for all comparisons ) ( Fig . 2 , panel C ) . In our a priori analytical plan , we also compared the ratio of reduced to oxidized biopterins ( BH4: BH2+B0 ) and the proportion of total biopterins as BH4 ( BH4/ BH4+ BH2+ B0 ) . Using these alternative renderings to examine reduced to oxidized biopterins , we found the same statistical relationships between CM in pairwise comparison to the other three clinical groups ( Table 2 ) . Compared to children with likely bacterial or cryptococcal meningitis ( n = 10 ) , children with cerebral malaria had lower BH4 than this subset of NMC children ( 1 . 10 [IQR 0 . 55–2 . 18] vs . 1 . 63 [IQR 1 . 39–2 . 90] μmol/mmol creatinine; p = 0 . 06 ) , significantly higher BH2 than children in this subset ( 2 . 24 [IQR 1 . 55–2 . 89] vs . 1 . 50 [IQR 0 . 93–2 . 27] μmol/mmol creatinine; p = 0 . 05 ) , and a significantly lower BH4:BH2 ratio than children with likely bacterial or cryptococcal meningitis ( 0 . 45 [0 . 21–1 . 13] vs . 1 . 35 [IQR 0 . 70–3 . 21]; p = 0 . 002 ) . Collectively , these measurements of urine biopterins demonstrate deficiency in systemic BH4 in CM . Guanosine triphosphate cyclohydrolase I ( GTPCH-1 ) is the rate-limiting enzyme in the first step for BH4 de novo synthesis . Mononuclear phagocytes activated by pro-inflammatory cytokines ( e . g . , INF-γ , TNF-α ) show enhanced transcription of the GTPCH-1 gene , GCH1 [38] . However , the activity of the second enzyme for biopterin synthesis , 6-pyruvoyl tetrahydropterin synthase ( PTPS ) , is constitutively very low in these cells and is unresponsive to cytokine-induced transcription activation [28 , 39 , 40] . The result is shunting of the GTPCH-1 product , 7 , 8 dihydroneopterin triphosphate , to neopterin end products—dihydrobiopterin ( NH2 ) and its oxidized metabolite , neopterin ( N0 ) [41] . These neopterin end products accumulate within mononuclear phagocytes and then exit the cells to be excreted in urine . Although the biological function of the neopterins remains obscure , elevated total urine neopterin ( NH2 + N0 ) is established as a marker of the cell-mediated immune response in a variety of inflammatory states [42] . NH2 measurement relies upon oxidation of NH2 to the naturally fluorescent N0 in vitro for quantification [43] . We took advantage of our collection method for preserving urine neopterins in their in vivo reduced ( NH2 ) and oxidized ( N0 ) states to quantify them in CM . In doing so , we sought to determine whether the neopterins , like the biopterins , have a redox ratio skewed towards oxidation in children with CM . Concentrations of urine neopterin metabolites differed significantly across the four groups ( Table 2 ) . Compared to HC , total neopterin ( NH2 + N0 ) levels were increased in UM , CM and NMC ( p<0 . 001 for all comparisons ) ( Fig . 3 , panel A ) . Total neopterin levels did not differ significantly in UM compared to CM ( p = 0 . 113 ) . The ratio of reduced to oxidized neopterins ( NH2:N0 ) was significantly decreased in CM compared to UM ( p = 0 . 001 ) and HC ( p = 0 . 017 ) . This ratio did not statistically differ between CM and NMC ( p = 0 . 069 ) ( Fig . 3 , panel B ) . The reduced to oxidized ratio for neopterins correlated with the reduced to oxidized ratio for biopterins ( BH4: BH2+B0 ) ( r = 0 . 28 , p<0 . 001 ) and oxidized neopterin ( N0 ) correlated with oxidized biopterin ( BH2 ) ( r = 0 . 63 , p<0 . 001 ) . We conclude from these measurements that while total urine neopterin levels did not differ in CM compared to UM , quantifying neopterins in their reduced to oxidized states may indicate oxidative stress in CM compared to UM . Angiopoietin-2 ( Ang-2 ) is a ligand released from endothelial Weibel-Palade bodies that acts as an autocrine mediator of endothelial activation [44] , including up-regulation of intercellular adhesion molecule ( ICAM ) receptors on endothelial cells [45] . Ang-2 plasma levels correlate with clinical severity in pediatric sepsis [10 , 46] and are independently associated with mortality in severe malaria [10 , 46 , 47] . NO is a known inhibitor of endothelial Weibel-Palade bodies exocytosis [48] . In vitro endothelial cell experiments demonstrate biopterin redox status impacts endothelial dysfunction [37] , but it is unknown whether whole body stores of biopterins correlate with in vivo circulating NO-dependent mediators of endothelial activation . To assess for associations between systemic biopterin status and endothelial activation we measured Ang-2 in a subset of randomly selected children enrolled as HC ( n = 9 ) , UM ( n = 10 ) , CM ( n = 16 ) , and NMC ( n = 9 ) . Plasma Ang-2 concentration differed significantly across the four groups ( Fig . 4 ) —HC ( 815 [IQR 345–1264] pg/mL ) , UM ( 1882 [IQR 1035–4693] pg/mL ) , CM ( 3445 ( IQR 2014–7534 pg/mL ) , and NMC ( 1149 [691–1616] pg/mL ) ( p<0 . 001 ) . Pairwise comparisons among groups showed that median plasma Ang-2 levels were higher in CM compared to HC ( p<0 . 001 ) , UM ( p = 0 . 082 ) and NMC ( p = 0 . 006 ) . Correlations between plasma Ang-2 and urine biopterin metabolites were assessed within each clinical group and no significant correlations were noted . A correlation scatterplot of plasma Ang-2 and urine BH2 values does visually demonstrate a non-significant correlation ( S1 Fig . ) , indicating that the small sample size may have limited our ability to establish a significant correlation . We conclude that Ang-2 is significantly elevated in severe pediatric falciparum malaria , but we were unable to demonstrate a correlation between Ang-2 and urine biopterin metabolites , possibly due to small sample size . To determine if the low urine levels of BH4 observed in CM could in part be explained by decreased GCH1 expression for GTPCH-1 , the rate-limiting enzyme in pterin biosynthesis [29] , we measured GCH1 mRNA in peripheral blood mononuclear cells ( PBMC ) isolated from a randomly selected sub-set of children enrolled as HC ( n = 5 ) , UM ( n = 8 ) and CM ( n = 9 ) . GCH1 mRNA transcription was at least as high in each malaria group [UM ( 1 . 35 [IQR 1 . 11–1 . 97] 2-ddCt ) and CM ( 1 . 39 [IQR 0 . 88–1 . 77] 2-ddCt ) ] relative to controls ( 0 . 82 [IQR 0 . 60–0 . 84] 2-ddCt ) ( p = 0 . 082 ) . We also measured GCH1 mRNA in children enrolled into a separate cohort using the same inclusion/exclusion criteria as HC ( n = 33 ) and as CM ( n = 9 ) . Combining the measurements from both cohorts [HC n = 38 and CM n = 18] , we found no significant difference in GCH1 mRNA between these two groups . We conclude that PBMC GCH1 mRNA and pterin synthesis is not decreased in children with malaria and does not account for the lower BH4 . We measured plasma phenylalanine in all 4 clinical groups to confirm our previous observation of hyperphenylalaninemia in clinical malaria [18] and to assess the relationship between systemic BH4 status and plasma phenylalanine concentration . Hyperphenylalaninemia ( upper limit of normal in children age > 12 months , 80uM; upper limit of normal in children age < 12 months , 100 uM ) was found in 4 ( 3 . 7% ) of 109 HC , 44 ( 72 . 1% ) of 61 UM , 40 ( 80% ) of 50 CM , and 23 ( 45 . 1% ) of 51 NMC . Plasma phenylalanine concentrations differed significantly across all groups ( Table 2 ) . This association remained after using linear regression to control for age , weight , and fasting duration . Using the same linear regression model , plasma phenylalanine concentrations differed significantly in CM compared to UM ( p = 0 . 04 ) , and across UM , CM survivors and CM fatalities ( p = 0 . 015 ) . Plasma phenylalanine concentration correlated with urine BH2 concentration among HC ( r = 0 . 29; p = 0 . 002 ) , UM ( r = 0 . 43; p = 0 . 001 ) , and NMC ( r = 0 . 32; p = 0 . 03 ) , but among CM the correlation was non-significant ( r = 0 . 22; p = 0 . 15 ) . Similarly , when analyzing UM and CM children jointly , there was a significant correlation between plasma phenylalanine and BH2 was observed ( r = 0 . 38; p<0 . 001 ) , but we when employed a partial correlation test to control for malaria disease severity , this correlation was non-significant ( r = 0 . 17; p = 0 . 09 ) . Plasma phenylalanine did not correlate with BH4 or the BH4/BH2 ratio . Despite a non-significant correlation among CM , overall we found a moderate correlation between plasma phenylalanine and urine BH2 levels , a result that is consistent with the body of literature demonstrating the ability of BH2 to inhibit PAH enzymatic conversion of phenylalanine to tyrosine [28] . In bivariate logistic regression among all malaria patients ( Table 3 ) , a unit decrease in Log10-urine BH4 and Log10-urine BH4/BH2 ratio was associated with 2 . 94-fold ( 95% CI 1 . 65–5 . 23 ) increase and 3 . 29-fold ( 95% CI 1 . 87–5 . 76 ) increase , respectively , in the odds of CM ( p<0 . 001 ) . A unit increase in log10-urine BH2 was associated with a 1 . 91-fold increase ( 95% CI 0 . 99–3 . 67 ) in the odds of CM ( p = 0 . 04 ) . In the backward stepwise multivariate logistic regression model for risk of CM , variables were retained in the model if their p-value was < 0 . 2 . The final model included urine BH4 , urine BH2 , P . falciparum histidine-rich protein-2 ( PfHRP-2 ) , respiratory rate and weight ( Table 3 ) . Urine BH4 and urine BH2 levels were independently associated with odds of CM: a unit decrease in Log10-urine BH4 was associated with 3 . 85-fold ( 95% CI 1 . 89–7 . 61 ) increase in the odds of CM ( p<0 . 001 ) . Note that as Table 3 models risk of CM for every unit increase in a given variable , the inverse is displayed—i . e . , a unit increase in Log10-urine BH4 was associated with 0 . 26-fold ( 95% CI 0 . 13–0 . 53 ) decrease in odds of CM [p<0 . 001] . A unit increase in Log10-urine BH2 was associated with a 2 . 86-fold ( 95% CI 1 . 01–8 . 07 ) increase in the odds of CM ( p = 0 . 047 ) . The urine BH4:BH2 ratio is not included in the logistic regression model due to co-linearity with BH4 . When the multivariate model included the BH4:BH2 ratio in lieu of urine BH4 , the odds ratio of CM was comparable to the odds ratio of CM seen with urine BH4: a unit decrease in Log10-urine BH4:BH2 ratio was associated with a 3 . 63-fold ( 95% CI 1 . 95–6 . 72 ) increase in the odds of CM ( p<0 . 001 ) .
Systemic BH4 concentrations are significantly lower in children with CM compared to HC , UM , and NMC . BH4 concentrations were likewise significantly lower in fatal CM when compared across the spectrum of UM and CM survivors , and a decrease in urine BH4 was independently associated with a 3 . 85-fold increase in odds of CM . While two prior studies reported conflicting results on BH4 levels in cerebrospinal fluid [17 , 49] , our study is the first published analysis of systemic biopterin levels in severe human malaria . Our findings have important implications regarding the underlying reasons for the decreased NO bioavailability and microvascular dysfunction observed in severe malaria [16 , 19 , 21] . These data also represent a novel link between the extensive vascular medicine research regarding the effects of BH4 and BH2 on the endothelium [27] and the vasculopathic mechanisms of severe malaria [16] . Our prior studies have demonstrated that NO is important in protecting against severe malaria , including CM , and have demonstrated several mechanisms by which NO bioavailability is decreased in malaria . These mechanisms include: low overall NO production [21]; low NOS2 protein levels in peripheral blood mononuclear cells [21]; low NOS activity in PBMC [22]; NOS2 single nucleotide polymorphisms that modulate NO production [23]; increased cell-free hemoglobin , an NO quencher [50]; decreased NOS substrate arginine [6 , 24]; and increased asymmetric dimethyl arginine ( ADMA ) [25 , 26] , an endogenous inhibitor of NOS activity [51] . Our present study documents another potential mechanism for low NO production in severe malaria—diminished levels of NOS cofactor , BH4 . Our prior studies of NO in malaria have focused on NOS2 ( inducible NOS ) , which is expressed in multiple tissues , including brain and endothelium . Our current findings apply not only to NOS2 , but to all three NOS isoforms ( endothelial , inducible and neuronal ) , as BH4 is an obligate co-factor for all three NOS isoforms and deficiency of BH4 can result in diminished NOS activity and cause uncoupling in all NOS isoforms . We observed that total urine biopterins ( which reflect total body biopterins ) were increased in mild and severe malaria . Expression of PBMC mRNA for CGH1 , the rate-limiting enzyme in biopterin synthesis , was at least as high in malaria as in HC . While pterin synthesis in human leukocytes may not reflect biopterin bioavailability in other tissues ( e . g . , the endothelium ) , this finding is consistent with our observation of increased total urine biopterins in malaria . Beyond the absolute totals , we think the more significant observation is the markedly skewed distribution of biopterin species in CM cases—we found that nearly 2/3 of biopterins were oxidized to inactive forms , a finding not previously reported . Tanzanian children with CM had significantly lower urine BH4 concentrations and a lower BH4:BH2 ratio compared to HC , UM and NMC . Quantification of the different redox states of biopterins in malaria—fully reduced , BH4 , and its oxidized metabolites , BH2 and B0—may be a novel method for measuring oxidative stress in severe malaria and possibly other critical illnesses . The same rationale applies to the NH2:N0 ratio . Our finding of increased levels of neopterin in UM and CM is consistent with prior reports of neopterin concentrations in malaria [52 , 53] . While the pathophysiologic role of neopterin remains unclear , the ratio of reduced to oxidized neopterin ( NH2:N0 ) might represent a novel parameter to quantify mononuclear phagocyte redox state . The decrease in both BH4:BH2 and NH2:N0 ratios indicates increased systemic oxidative stress in CM . As a NOS co-factor , BH4 performs both structural and biochemical functions . It stabilizes the NOS homodimer , thereby enabling NOS catalytic function , and it donates an electron to form a transient BH4∙+ radical which is required for oxidation of L-arginine to L-citrulline [27] . However , in the absence of BH4 , NOS catalyzes a reaction in which NADPH-derived electrons reduce heme iron thereby binding and activating oxygen [54] . Without BH4-dependent L-arginine oxidation , activated oxygen is released from the heme catalytic center of NOS as superoxide . This generation of superoxide is termed NOS “uncoupling , ” as L-arginine oxidation to L-citrulline is no longer coupled to oxygen activation for NO synthesis and release [27] . In an environment where both superoxide ( uncoupled NOS ) and NO ( coupled NOS ) are generated , both products may react to yield peroxynitrite [55 , 56] . Peroxynitrite and superoxide induce further uncoupling by oxidizing available BH4 to BH2 [57] , thereby decreasing the availability of BH4 for NO synthesis . This uncoupling is compounded by BH2 accumulation , which competes with BH4 at its NOS binding site , further inhibiting NOS-catalyzed NO synthesis [37] . Thus a feed-forward mechanism , initiated by low intracellular BH4 levels , ensues and leads to oxidative injury caused by superoxide and peroxynitrite [58] . Experimental models of eNOS activity indicate that the BH4:BH2 ratio , which we found to be significantly decreased in CM compared to all other groups , is the measurement that most strongly correlates with NOS uncoupling [37] . The hyperphenylalaninemia that we previously described in UM and CM [18] was one of the initial observations that led us to test the hypothesis of impaired BH4 in CM , since BH4 is an essential cofactor for the enzymatic hydroxylation of phenylalanine to tyrosine . We confirmed our prior finding of elevated plasma phenylalanine in UM and CM . Our present findings suggest that this elevation in malaria is secondary to decreased systemic concentrations of BH4 and increased concentrations of BH2 . Decreased BH4 and increased BH2 systemic levels in children with CM may be relevant to the central pathogenic mechanisms in severe malaria—endothelial activation and dysfunction and the sequestration of pRBCs , unparasitized RBCs , mononuclear cells , and platelets resulting in microcirculatory congestion of post-capillary venules [14 , 16 , 19 , 59] . Endothelial cells low in intracellular BH4 and high in BH2 favor decreased NO production by NOS , and low NO synthesis is associated with endothelial dysfunction [6] . The effect of decreased NO production due to NOS uncoupling may exacerbate microvascular sequestration in several ways . Low NO states are associated with decreased RBC deformability [60] . In vitro endothelial cell models have shown that increased bioavailability of NO is associated with decreased cytoadherence of P . falciparum-infected RBCs via down-regulation of endothelial receptors for P . falciparum erythrocyte membrane protein [61 , 62] . NO deficiency causes increased vascular tone with decreased vessel diameter and possible flow impedance . Additionally , NOS uncoupling due to low BH4 and high BH2 would also favor generation of peroxynitrite , a highly reactive species thought to exacerbate endothelial cell dysfunction [63] . Given that low BH4 levels may be contributing to endothelial dysfunction and sequestration in severe malaria , adjunctive therapy to reduce BH2 and regenerate BH4 warrants further study [64 , 65] . The need for such studies is also supported by an experimental animal model of cerebral malaria demonstrating that the blunted cerebral arteriolar response to NOS agonists is partially recovered by supplementation with BH4 [66] . While our study has many strengths , including rigorous case definitions for the four clinical groups and rigorous methods for sample collection , processing and analysis , we acknowledge several limitations . The specificity of the WHO case definition for CM may be as low as 77% [13] . The causes and severity of illness in the NMC group were heterogeneous , making comparisons with this group more tenuous . Since we were able to measure plasma creatinine in only 16 UM and 39 CM patients , we did not include plasma creatinine in the final multivariate logistic regression model assessing for an independent association between urine BH4 concentrations and malaria severity . We therefore cannot exclude that the significant association between BH4 and malaria severity is dependent upon renal function . We performed a multivariate logistic regression sub-model for the 48 malaria patients who had urine biopterin and plasma creatinine results . In this sub-model we found that only respiratory rate ( OR 1 . 11 [95% CI 1 . 01–1 . 21]; p = 0 . 04 ) and plasma creatinine ( OR 201 . 42 {95% CI 8 . 15–4978 . 96]; p = 0 . 001 ) were independently associated with odds of cerebral malaria ( BH4 OR 0 . 34 [95% CI 0 . 09–1 . 28;p = 0 . 11 ) . This indicates that plasma creatinine has a statistically significant association with malaria severity , but the wide confidence intervals show a lack of power for more precisely assessing this relationship and its impact on the association between urine BH4 and malaria severity . While the lack of plasma creatinine results is a key limitation of our final multivariate logistic regression model , we note the following in regards to the urine BH4 measurements and renal function: 1 ) only two children ( both with CM ) had clinically significant renal impairment; 2 ) A linear regression model of urine BH4 showed no significant variance by plasma creatinine among the 85 children in whom both a urine BH4 measurement and a plasma creatinine measurement . 3 ) Total urine biopterins were increased in CM patients , which argues against the higher mean plasma creatinine values in CM as accounting for the decreased levels of BH4 in the urine of CM patients . Inclusion of plasma creatinine in the model would have been optimal , but the reasons detailed above , we think exclusion of creatinine from the multivariate logistic regression model is justified and preferred . Additional limitations include the fact that we did not measure NO bioavailability directly and that our findings do not demonstrate a causal link between endothelial cell dysfunction and low systemic BH4 . Unlike our companion study demonstrating decreased systemic BH4 in Indonesian adults with severe falciparum malaria , we do not have measures of endothelial function in this study to compare with the decreased levels of systemic BH4 we have observed in Tanzanian children with CM . Among a subset of the cohort we were able to measure Ang-2 , a mediator of endothelial cell activation that is associated with reduced NO-bioavailability in both severe malaria and sepsis [10 , 67] . The markedly elevated plasma Ang-2 levels and the abnormal urine biopterin metabolite levels observed among children with malaria were not significantly correlated . A small sample size may have limited our ability to show a significant correlation between Ang-2 and BH2 . While we speculate that the highly perturbed BH4: BH2 ratio is due to oxidative stress , we do not have additional purported measures of oxidative stress with which to compare biopterin redox status . An alternative explanation for this the low BH4: BH2 ratio is impaired recycling via DHFR conversion of BH2 to BH4 . In conclusion , we found that BH4 , an essential cofactor for NO production , is low in children with CM , and that low BH4 concentrations are independently associated with disease severity in children with malaria . Low BH4 may contribute to the pathogenesis of severe malaria as it represents an important mechanism of low NO bioavailability . Furthermore , low BH4 together with elevated BH2 likely leads to generation of reactive oxygen species . These potential sequelae of low systemic BH4 likely contribute to endothelial dysfunction and pRBC sequestration—hallmark pathophysiologic features of severe malaria . Interventions that replenish BH4 or redistribute the BH4:BH2 ratio toward reduced BH4 merit investigation as adjunctive therapies to improve outcomes in pediatric severe falciparum malaria .
We conducted a prospective observational study in Dar es Salaam , Tanzania . Children were enrolled from the clinics and the inpatient wards of Amana and Mwanayamala District Hospitals into the following four groups: UM , CM , NMC ( see criteria below ) , and HC . These two district hospitals are separated by 6 . 5 kilometers , and both are located in semi-urban areas of Dar es Salaam . Once enrolled , children with CM or with NMC were transferred to the clinical research unit at the Hubert Kairuki Memorial University Hospital for further evaluation and care . Approval for this study was obtained from the institutional review boards of Hubert Kairuki Memorial University , United Republic of Tanzania National Medical Research Institute , University of Utah , and Duke University . Informed consent was obtained from parents or guardians of all children , and the U . S . Department of Health and Human Services guidelines for human subjects research were followed . Children were 6 months to 6 years in age . The World Health Organization ( WHO ) case definition for CM was used as the inclusion criteria for the CM cohort: any level of P . falciparum parasitemia on peripheral blood smear; unarousable coma ( Blantyre Coma Score ≤ 2 ) not attributable to hypoglycemia ( blood glucose level < 40 mg/dL ) and persisting more than 60 minutes after any convulsion; no other identifiable cause of coma [68] . Inclusion criteria for UM were as follows: a clinical syndrome consistent with malaria and a documented fever ( temperature ≥ 38° C ) or history of fever within 48 hours from time of enrollment; P . falciparum parasitemia > 10 , 000 parasites/μL on Giemsa-stained blood film plus a positive P . falciparum RDT ( Paracheck-Pf; Omega Diagnostics ) ; no other cause of fever identified; no WHO warning signs suggestive of severe disease [68] . These warning signs were the following: inability to suckle , eat and/or drink; excessive vomiting; abnormal respiratory rate or respiratory distress as evidenced by accessory muscle use , suprasternal retractions , or intercostal retractions; recent history of convulsions; altered mental status; inability to sit unaided . Exclusion criteria for both groups with malaria were any of the following: microscopic evidence of mixed infection with any other Plasmodium species; bacterial co-infection as evidenced by septicemia or urinary tract infection; oral or intravenous quinine or oral artemesinin combination therapy initiated > 18 hours prior to enrollment; hemoglobin < 5 mg/dL , as erythrocyte transfusions were not readily available at our study sites . Children with CM were excluded if they had evidence of bacterial meningitis as indicated by isolation of a pathogen from CSF culture or by CSF analysis . Similar aged healthy children and children presenting with NMC were prospectively enrolled as control groups . The healthy children were enrolled from outpatient well-baby clinics at the two district hospitals . Eligible children had no signs or symptoms of active illness , no febrile illness within the previous two weeks , no history or evidence of an active inflammatory condition , and negative P . falciparum RDT ( Paracheck-Pf; Omega Diagnostics ) . Children with NMC were eligible if they had a CNS condition for which a lumbar puncture for CSF analysis was clinically indicated as part of diagnostic evaluation and management ( e . g . suspected meningitis , encephalitis , hemorrhage , trauma , metabolic , toxic , recurrent seizures as cause for altered mental status ) . All children enrolled in this group had a Giemsa-stained blood film negative for Plasmodium and a negative P . falciparum RDT ( Paracheck-Pf; Omega Diagnostics ) . A subset of children in this group were classified as likely bacterial or fungal meningitis based on detection of a microbiologic pathogen in CSF and one of the following findings on CSF analysis: white blood cell count > 6/μL , neutrophils present on cytospin Wright stain prep , or glucose < 2/3 plasma glucose level . At presentation , demographic information , clinical history , and examination were documented using standardized case report forms . Capillary blood samples were obtained for Giemsa-stained thick and thin blood films as well as on-site malaria rapid diagnostic testing . Venous samples for routine laboratory analysis included complete blood count ( Beckman-Coulter Act 10 ) , sodium , potassium , chloride , bicarbonate , blood urea nitrogen , creatinine , and venous blood gas ( i-STAT 1; Abbott Laboratories ) . Urine collected for pterin quantification ( see below ) was also subjected to urine dipstick analysis ( Multistix 10 SG; Siemens Healthcare Diagnostics ) and urine culture . These blood and urine laboratory results were immediately available to the clinician . Blood cultures were obtained in all children with UM and CM to rule out concomitant bacteremia . Lumbar puncture with opening pressure measurement was done in all patients with coma to evaluate for meningitis . CSF analysis included determination of glucose and protein levels , cell count with differential by trained microscopists , and bacterial and fungal cultures . Children with UM and CM received anti-malarial therapy and other supportive care as per standard Tanzanian national protocols at the time of the study ( artemesinin combined oral therapy and intravenous quinine , respectively ) . Treatment was initiated as soon as the diagnosis of malaria was suspected . Children with CM and NMC were re-examined daily until death or hospital discharge . Urine was collected from subjects at the time of enrollment . Strict adherence to the collection procedure was followed in every case . All collection and handling of urine samples was done in the dark or under dim lighting ( collection bags covered with a black , light-impermeable cloth ) to prevent photo-oxidation . For quantification of urine biopterins and neopterins , 5–20 ml urine was collected voluntarily into a sterile urine collection cup or with use of a pediatric bag ( U-bag urine collector , Hollister Pediatric ) affixed to the perineum with adhesive . In both cases , issuing urine was immediately collected into solid dithioerythritol ( DTE , approximately 50 mg/ml urine ) and diethylene triamine penta-acetic acid ( DETAPAC , approximately 5 mg/ml urine ) and mixed well to dissolve the powders as described previously [35] . Immediately after collection , urine samples were placed into an insulated transport cooler charged with large cooling packs preconditioned at -80° C . This resulted in freezing of urine samples within minutes . The samples were then stored in a -80° C freezer and then transported to the USA in a liquid nitrogen dry shipper , and subsequently stored at -80° C until pterin analysis . This procedure exceeded in stringency the conditions required to prevent oxidation of reduced pterins previously described [35 , 69] . Reagent BH4 exposed to this procedure was stable at concentrations in the range measured on clinical urine samples; no detectable oxidation of BH4 measured by our analytic methods occurred . Conversely , exposure of oxidized reagent pterins to this procedure did not result in any measureable reduction . Thus the analysis of our samples revealed quantities of urine pterins as they existed in vivo . Thawed urine samples were mixed and directly subjected to analysis for BH4 , BH2 and B0 and for NH2 and N0 by high-performance liquid chromatography ( HPLC ) using sequential electrochemical and fluorescence detection . The procedure relies on reversed phase HPLC separation of pterins , BH4 , BH4 , NH2 , B0 and N0 in samples of urine . BH4 is oxidized to quinonoid dihydrobiopterin ( q-BH2 ) and then reduced back to BH4 . The current generated on the reduction is monitored and used to determine BH4 concentration using an ESA Coularray electrochemical integration system . BH2 and NH2 are oxidized to B0 and N0 respectively by electrochemistry and then measured by fluorescence using EZ Chrom integration system . B0 and N0 are not affected by electrochemical electrodes and are measured by their natural fluorescence using the EZ Chrom system . Further details of the analytical methods are as reported previously [70] ( see S2 Fig . and see Fig . 2 & 3 in citation 41 for example chromatograms ) . PfHRP-2 , a measure of total parasite biomass , was quantified by ELISA as previously described [6 , 10] using primary and secondary monoclonal antibodies to P . falciparum HRP-2 ( MPFM-55A and MPFG-55P; Immunology Consultant Laboratories ) . Concentrations were derived from standard curves utilizing purified PfHRP-2 kindly provided by D . Sullivan ( Johns Hopkins University School of Public Health ) . Samples with ODs outside the linear part of the curve were repeated at an appropriate dilution until an accurate concentration was determined . The lower limit of detection was 1 . 5 ng/mL . Ang-2 was measured by ELISA ( R&D Systems ) as noted before [10] . PBMC were isolated using standard methods . RNA was extracted from PBMC and subjected to real-time reverse transcriptase-polymerase chain reaction ( RT-PCR ) techniques . We used the equivalent of 100 ng RNA per reaction . Stratagene VILO kit was used for the RT reaction , and the Roche faststart universal probe master kit for PCR ( Roche Applied Science ) . Primer/probe sets were designed by the Roche universal probe library application . Primers were then made by IDT while probes were from the Roche universal probe library ( http://universalprobelibrary . com ) ( Roche Applied Science ) . The quantitative PCR were performed] on an ABI7300 machine . Samples were run in triplicate , or in duplicate if sample amount was limiting . The GAPDH and HPRT1 genes were selected as endogenous control genes . The average Ct of the endogenous controls for each sample was used in the dCt calculation . For individual ddCt values , we used each sample dCt and the averaged healthy control dCt . For group ddCt values , we used the average dCt for each group to calculate ddCt for the group . The data is expressed as 2-ddCt , which is the fold change as compared to the average of the healthy controls . Table 4 displays the relevant sequences for the primers and probes . Blood samples were collected into heparin tubes , mixed and immediately centrifuged to sediment blood cells . Supernatant plasma was removed to freezing tubes and placed into the transport vessel described above , after which they were stored at -80° C until thawing for analysis . Amino acids were quantified by ion-exchange chromatography with detection using spectrophotometry after reaction with ninhydrin . All amino acid analysis was performed at the Biochemical Genetics Section , ARUP Laboratories , University of Utah in collaboration with Dr . Marzia Pasquale . Statistical analysis was performed with STATA software ( version 12 . 0; StataCorp ) . Results are presented as mean with 95% confidence intervals for normally distributed continuous variables or median with interquartile range for variables with non-parametric distribution . For continuous variables with normal distribution , differences across groups were compared by ANOVA with Bonferroni adjustment for multiple comparisons and differences between groups were compared using Student’s t-test . For continuous variables with a non-parametric distribution , differences across groups were compared using Kruskal-Wallis test and differences between groups were compared using the Wilcoxon rank-sum test . Correlation between variables with non-parametric distributions was assessed using Spearman correlation coefficients , and we used the partial correlation test to assess for correlations while controlling for malaria severity . Differences in proportions between groups were assessed with Chi-square test . Multivariate linear regression was used to control for confounding variables that could affect urine BH4 and plasma phenylalanine concentrations . Bivariate logistic regression was used to analyze continuous variables for co-variation by a binomial variable ( e . g . , co-variation of urine BH4 concentration by the presence or absence of pyuria ) . For children with UM and CM , backward stepwise multivariate logistic regression was performed to determine whether phenylalanine or urine biopterin species were independently associated with disease severity . Bivariate logistic regression was performed for available biologically plausible variables known to be associated with severe malaria: hemoglobin , PfHRP-2 , and respiratory rate and bicarbonate , as indicators of metabolic acidosis [3 , 71 , 72] . Variables were included in the multivariate logistic regression model if p < 0 . 20 on bivariate analysis , and were retained in the multivariate model if the p-value generated in the multivariate analysis was <0 . 05 . The multivariate model also controlled for age , sex , weight and duration of fasting . Plasma creatinine was not included in the multivariate model because creatinine results were only available for 14 UM and 34 CM patients with urine biopterin results . In a multivariate logistic regression sub-model of these 48 malaria patients we did include creatinine in order to assess the relationship between estimated renal function and odds of cerebral malaria . Continuous variables with non-parametric distributions were log-transformed to meet normality assumptions for use in the logistic regression models . Goodness-of-fit was assessed by Hosmer-Lemeshow test . A two-sided p-value of < 0 . 05 was employed as the cut-off for statistical significance throughout the manuscript unless otherwise indicated ( e . g . , Bonferoni adjustments when making comparisons across multiple groups ) .
|
Vascular nitric oxide ( NO ) bioavailability is decreased in severe falciparum malaria and associated with microvascular dysfunction , increased activation of the cells lining blood vessels ( endothelial cells ) and increased parasite biomass . Tetrahydrobiopterin ( BH4 ) is an essential cofactor for nitric oxide synthase ( NOS ) enzymatic conversion of L-arginine to NO and L-citrulline . But when BH4 is low , NOS is “uncoupled” and produces superoxide instead of NO . In oxidative conditions , BH4 is oxidized to dihydrobiopterin ( BH2 ) and biopterin ( B0 ) . BH2 competes with remaining BH4 at its NOS binding site , further decreasing NOS-catalyzed NO production . We measured BH4 , BH2 and B0 in the urine of children with coma due to falciparum malaria ( cerebral malaria ) , uncomplicated falciparum malaria , children with non-malaria central nervous system conditions and healthy controls . Urine BH4 was significantly decreased and BH2 significantly increased in cerebral malaria compared to uncomplicated malaria , non-malaria central nervous conditions and healthy controls , suggesting increased oxidative stress and insufficient recycling of BH2 back to BH4 . Urine BH4 concentration was independently associated with increased risk of cerebral malaria . Given that safe therapies for regenerating BH4 have been studied in chronic vascular disease , this finding of low BH4 in pediatric cerebral malaria offers a new area of investigation for adjunctive therapies aimed at improving NO bioavailability and , consequently , clinical outcomes in severe falciparum malaria .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Impaired Systemic Tetrahydrobiopterin Bioavailability and Increased Oxidized Biopterins in Pediatric Falciparum Malaria: Association with Disease Severity
|
While the domestication history of Asian rice has been extensively studied , details of the evolution of African rice remain elusive . The inner Niger delta has been suggested as the center of origin but molecular data to support this hypothesis is lacking . Here , we present a comprehensive analysis of the evolutionary and domestication history of African rice . By analyzing whole genome re-sequencing data from 282 individuals of domesticated African rice Oryza glaberrima and its progenitor O . barthii , we hypothesize a non-centric ( i . e . multiregional ) domestication origin for African rice . Our analyses showed genetic structure within O . glaberrima that has a geographical association . Furthermore , we have evidence that the previously hypothesized O . barthii progenitor populations in West Africa have evolutionary signatures similar to domesticated rice and carried causal domestication mutations , suggesting those progenitors were either mislabeled or may actually represent feral wild-domesticated hybrids . Phylogeographic analysis of genes involved in the core domestication process suggests that the origins of causal domestication mutations could be traced to wild progenitors in multiple different locations in West and Central Africa . In addition , measurements of panicle threshability , a key early domestication trait for seed shattering , were consistent with the gene phylogeographic results . We suggest seed non-shattering was selected from multiple genotypes , possibly arising from different geographical regions . Based on our evidence , O . glaberrima was not domesticated from a single centric location but was a result of diffuse process where multiple regions contributed key alleles for different domestication traits .
Domestication of crop species represents a key co-evolutionary transition , in which wild plant species were cultivated by humans and eventually gave rise to new species whose propagation were dependent on human action [1–3] . The evolutionary origin ( s ) of various crop species have been the subject of considerable interest . Studying it has broadened our understanding of the early dynamics associated with crop species origins and divergence , the nature of human/plant interactions , and the genetic basis of domestication . Moreover , an understanding of the evolutionary history of crop species aids genetic mapping approaches , as well as informs plant breeding strategies . Within the genus Oryza , crop domestication has occurred at least twice—once in Asia and separately in Africa . In Asia , the wild rice O . rufipogon was domesticated into the Asian rice O . sativa approximately 9 , 000 years ago [4] . In West Africa , the wild rice O . barthii was independently domesticated into the African rice O . glaberrima about 3 , 000 years ago [4] . Recent archaeological studies have also suggested that a third independent domestication event occurred in South America during pre-Columbian times , but this crop species is no longer cultivated [5] . The domestication history of Asian rice has been extensively studied both from the standpoint of archaeology [6] and genetics [7] . In contrast , much less is known about the domestication of O . glaberrima . Based on the morphology of rice grown in West Africa , the ethnobotanist Portères was the first to postulate an O . glaberrima domestication scenario [8 , 9] , in which the inner Niger delta region in Mali as the center of domestication ( Fig 1 ) . He based this hypothesis on O . glaberrima in this area predominantly having wild rice-like traits ( termed “genetically dominant characteristics” by Portères ) , observing loosely attached spikelets , reddish brown pericarps , and anthocyanic pigmentation . In contrast , O . glaberrima with domesticated rice-like traits ( termed “genetically recessive characteristics” by Portères ) were found in two geographically separated regions: ( i ) the Senegambia region bordering the river Sine to the north and river Casamance to the south , and ( ii ) the mountainous region of Guinea . Portères hypothesized the derived traits observed in O . glaberrima from Senegambia and Guinea were due to those regions being secondary centers of diversification , but the inner Niger delta region remained as the primary center of diversity for African rice . Initial archaeological excavations found ceramic impressions of rice grains in north-east Nigeria dating ~3 , 000 years ago , but first evidence of documented O . glaberrima has been found in the inland Niger delta at Jenne-Jeno , Mail dating ~2 , 000 years ago [10] . A more recent find at an excavation at the lower Niger basin north of Benin found O . glaberrima dating to ~1 , 600 to ~1 , 100 years ago , suggesting domesticated African rice had spread down the Niger river by this time from the inland Niger delta region [11] . Few population genetic studies have attempted to understand the evolutionary history and geographic structure of O . glaberrima . Microsatellite-based analysis showed genetic structure within O . glaberrima [12] , suggesting the phenotypic differences observed by Portères may have stemmed from this population structure . With high-throughput sequencing technology , population genomic analysis indicated O . barthii , the wild progenitor of O . glaberrima , had evidence of population structure as well , dividing it into 5 major genetic groups ( designated as OB-I to OB-V ) [13] . The OB-V group from West Africa was most closely related to O . glaberrima , which caused previous researchers to suggest that this O . barthii group from West Africa was likely to be the direct progenitor of African rice [13] . Genome-wide polymorphism data also indicated that O . glaberrima had a population bottleneck spanning a period of >10 , 000 years , indicating a protracted period of pre-domestication related management during its domestication [14] . A recent study based on simulations detected population expansion after the bottlenecking event , to originate from the inland Niger delta suggesting the origin of O . glaberrima to be in this region [15] . While previous genome-wide variation studies have given valuable insights into the evolutionary history of O . glaberrima , they have not necessarily examined how O . glaberrima was domesticated from O . barthii . This is because the domestication history of a crop is best examined from the pattern of variation observed in genes underlying key domestication phenotypes [2] . Crop domestication accompanies a suite of traits , often called the domestication syndrome [16] , which modified the wild progenitor into a domesticated plant dependent on humans for survival and dispersal [1] . In rice , these traits include the loss of seed shattering [17 , 18] , plant architecture change for erect growth [19 , 20] , closed panicle [21] , reduction of awn length [22 , 23] , seed hull and pericarp color changes [24 , 25] , change in seed dormancy [26] , and change in flowering time [27] . During the domestication process , it is likely that these traits were not selected at the same time and selection would have occurred in subsequent stages . Traits such as loss of seed shattering and plant erect growth would have been among the initial phenotypes humans have selected to distinguish domesticates from their wild progenitors . On the other hand , traits that improved taste and appearance of the crop , or adaptation to the local environment would likely have been favored in later diversification/improvement stages of crop evolution [3 , 28] . Genes involved in the early stage domestication process are key to understanding the domestication process of a crop . Sequence variation from these early stage domestication genes can indicate whether a specific domestication trait had single or multiple causal mutations , revealing whether domestication has a single or multiple origin . The geographic origin and spread of domestication traits can be inferred from sequence variation in domestication loci within contemporary wild and domesticate populations [17 , 29–32] . In Asian rice , for example , genome-wide single nucleotide polymorphisms ( SNPs ) have suggested that each rice subpopulation had independent wild rice populations/species as their progenitors [33–37] , but the domestication genes revealed a single common origin of these loci [35] , suggesting a single de novo domestication model for Asian rice [37–42] . On the other hand , the domestication gene for the non-brittle phenotype ( btr1 and btr2 ) in barley had at least two independent origins [43 , 44] , likely from multiple wild or proto-domesticated individuals [45] . This suggests barley follows a multiple domestication model [46–48] originating from multiple ancestral population [45 , 49 , 50] . To better understand the domestication of O . glaberrima , we have re-sequenced whole genomes of O . glaberrima landraces and its wild progenitor O . barthii from the hypothesized center of origin in the inner Niger delta , the middle and lower Niger basin that includes the countries Niger and Nigeria , and from Central Africa which includes Chad and Cameroon . The latter two regions were not heavily sampled in previous genomic studies . Together with published O . barthii samples from West Africa [13] and O . glaberrima samples from the Senegambia and Guinea region [14] , we conducted a population genomic analysis to examine the domestication history of O . glaberrima . The domestication history were further examined from the evolutionary analysis of genes involved in the early stage domestication process , mainly in the traits involving loss of shattering and erect plant growth . To complement the inferred domestication history , we measured panicle threshability , an important early domestication trait associated with seed non-shattering , from our O . glaberrima samples to further elucidate the domestication history of O . glaberrima . With our data we examine the evolutionary and population relationships between O . glaberrima and O . barthii , the demographic history , and the geographic origin ( s ) of domestication of the African rice O . glaberrima .
We re-sequenced the genomes of 80 O . glaberrima landraces from a geographic region that spanned the inner Niger delta and lower Niger basin region ( S1A Fig ) . Together with 92 O . glaberrima genomes that were previously re-sequenced [14] , which originated mostly from the coastal region ( S1B Fig ) , the 172 O . glaberrima genomes analyzed in this study represent a wide geographical range from West and Central Africa . We also re-sequenced the genomes of 16 O . barthii samples randomly selected from this area , which includes the areas from coastal west Africa , inner Niger delta , and the lower Niger basin ( S1C Fig ) . These were analyzed together with the 94 O . barthii genomes that were previously re-sequenced [13] . The average genome coverage in the data set we gathered for this study was ~16 . 5× for both domesticated and wild African rice samples , and is comparable to the sequencing depth ( ~16 . 1× ) in our previous study . The Wang et al . [13] study sequenced a subset of their samples to a higher depth ( ~19 . 4× ) , although the majority of their samples had relatively low coverage ( ~3 . 9× ) ( see S1 Table for genome coverage of all samples in this study ) . To avoid potential biases in genotyping that arises from differences in genome coverage [51 , 52] , we conducted our population genetic analysis using a complete probabilistic model to account for the uncertainty in genotypes for each individual [53 , 54] . For the subset of our analysis that required genotype information for each sample , we employed SNPs called from individuals with greater than 10x genome-wide coverage . After quality control filtering , we identified a total of 634 , 418 and 1 , 568 , 868 post-filtered SNPs from the non-repetitive regions of the O . glaberrima and O . barthii genomes , respectively . The genetic structure across domesticated and wild African rice was examined by estimating the ancestry proportions for each individual in our dataset . We employed the program NGSadmix [55] , which uses genotype likelihoods from each individual for ancestry estimation and is based on the ADMIXTURE method [56] . Ancestry proportions were estimated by varying the assumed ancestral populations ( K ) from 2 to 9 groups ( S2 Fig ) . With K = 2 NGSadmix divided the data set into O . glaberrima and O . barthii species , with several O . glaberrima samples having varying degrees of O . barthii ancestry ( ranging from 4 . 5 to 40 . 5% ) . Interestingly , there were a number of O . barthii samples that had high proportions of O . glaberrima ancestry . All these wild rice with discernible O . glaberrima admixture corresponded to the designated OB-V O . barthii group and hypothesized progenitor of O . glaberrima from Wang et al . [13] . However , our ancestry analysis suggests this wild O . barthii group could also be a result of either wild-domesticated rice hybridization or mislabeling of O . glaberrima as O . barthii ( see below ) [57] . Increasing K further subdivided O . glaberrima into subpopulations that had a geographical basis ( see S2 Fig for all K and their geographic distribution ) . For instance , K = 3 divided the O . glaberrima into two major subpopulations , first a coastal population that includes the Senegambia and Guinea highland region , and second an inland population that includes the inland Niger delta and lower Niger river basin region ( Fig 2 ) . At K = 5 , there were three major genetic groups within O . glaberrima and two within O . barthii . The two O . barthii genetic groups corresponded to OB-I and OB-II group identified in Wang et al . [13] . For O . glaberrima , the ancestry proportions showed structuring into 3 major geographic regions: coastal , inner Niger delta , and lower Niger basin populations ( Fig 2 ) . At K = 7 , O . glaberrima were divided into 5 genetic groups where the coastal and inner Niger delta population were further divided into northern and southern genetic groups . It is also at K = 7 where O . glaberrima divided into genetic groups that are consistent with Portères observation—that the coastal population is divided into a Senegambia or Guinea highland genetic cluster , while the samples closest to the inner Niger delta forms a unique genetic cluster ( Fig 2 ) . At K = 9 , O . barthii is separated into the three genetic groups OB-I , OB-II , and OB-III that were previously identified from Wang et al . [13] . For O . glaberrima the lower Niger basin population divided into two geographic regions , where the samples closer to the inner Niger delta formed its own genetic cluster ( Fig 2 ) . Importantly , what is noticeable with increasing K is that populations appeared to be separating into smaller , and more highly localized geographical clusters . We then conducted phylogenomic and principal component analysis ( PCA ) to verify our ancestry proportion results . Phylogenomic analysis were conducted using genotype likelihoods to estimate the pairwise genetic distances [58] and build a neighbor-joining tree ( Fig 3A ) . O . glaberrima formed a paraphyletic group relative to several O . barthii individuals . We noticed that O . glaberrima landraces could be divided into 5 phylogenetic groups sharing a common ancestral node . Although the bootstrap support on the five ancestral nodes were weak , the geographic distribution of these 5 phylogenetic groups ( Fig 3B ) were concordant with the geographic distribution of the ancestry components in the O . glaberrima subpopulations identified at K = 7 ( Fig 2 , note at K = 7 O . glaberrima forms five major genetic clusters while O . barthii forms two major genetic clusters ) . The 5 phylogenetic groups clustered into five geographic locations: north and south coastal population , north and south inland Niger delta population , and a lower Niger basin population . The O . glaberrima population genotype likelihoods were also used for principal component analysis ( PCA ) , which visualized the population relationships [54] . For the PCA plot , individuals were color coded according to the grouping status determined from the phylogenomic results ( Fig 3A and Fig 3B ) . When color-coded according to the phylogenomic tree grouping , PCA results showed 5 independent clusters for O . glaberrima ( Fig 3C ) . In addition , the distribution of individuals along the two principal components showed striking similarity with their geographic distribution ( Fig 3B versus Fig 3C ) . Together , our analyses of ancestry components , phylogenomics , and PCA suggest O . glaberrima has a geographically based population structuring with at least 5 subpopulations ( Fig 3A ) . Consistent with the hypothesized Guinea highland and Senegambia populations , the coastal populations were divided into OG-A1 and OG-A2 genetic groups ( collectively the OG-A supergroup ) . The lower Niger basin and central African individuals formed as a single OG-B group . Finally , for the inner Niger Delta region , landraces closest to the delta formed the OG-C1 group while the others formed the OG-C2 group; collectively these represent the OG-C supergroup . We note there are several methods of testing and choosing the most appropriate number of genetic clusters ( K ) for a population sample [56 , 59–62] . However , these statistical tests can be misleading [63 , 64] , often prompting overconfidence in a single K value that may or may not be biologically relevant . Thus , we emphasize our choice of dividing O . glaberrima into five major groups represent the minimum possible grouping based on historical observations [8 , 9] and geography ( Figs 2 and 3 ) . We also find that these 5 groups had significant correlations with the geographical distributions of domestication gene mutations and phenotypes ( see below domestication gene analysis for more detail ) , further suggesting they represent biologically relevant groupings . At K = 7 the majority of the newly sequenced O . barthii from this study belonged to either OB-I or OB-II subpopulations designated by Wang et al . [13] ( S3 Fig ) . The ancestry proportion for the OB-III and OB-IV groups suggested these individuals were an admixed group , with OB-III an admixture of OB-I and OB-II , and the OB-IV group possessing a mix of ancestry from both wild and domesticated rice . Note that at higher K values , OB-III formed its own genetic cluster while OB-IV showed ancestry with large proportions from wild and domesticated rice . Unlike the OB-V group of O . barthii , which also had several individuals of mixed wild and domesticated rice ancestry , the OB-IV group did not phylogenetically cluster with O . glaberrima ( Fig 3 and S3 Fig ) . This suggests that the OB-IV subpopulation may be an evolutionary distinct population , and the ancestry proportions were possibly mis-specified [64] . Hence , we considered individuals that were monophyletic with the OB-I or OB-II subpopulations as the wild O . barthii subpopulation and henceforth designated it as OB-W [= wild] ( Fig 3A ) . O . barthii that were paraphyletic with O . glaberrima were considered as a separate O . barthii group and designated as OB-G [= glaberrima-like] ( Fig 3A ) . Geographically , OB-G was found throughout West Africa but OB-W was found mostly in inland West African countries such as in Mali , Cameroon , and Chad ( S2 Table ) . Before examining the relationships between our 5 inferred genetic clusters for O . glaberrima , we filtered individuals with spurious classification . First , we find that there were 3 O . glaberrima individuals ( IRGC104883 , IRGC105038 , and IRGC75618 ) that did not group with any of the 5 population groups ( Fig 3A white arrow ) , but rather formed as a sister group to all O . glaberrima or sister group to both OG-A and OG-B group . Because they were most closely related to OB-G samples we considered them as OB-G as well . Interestingly , there were also two O . glaberrima individuals ( IRGC103631 and IRGC103638 ) that phylogenetically clustered with O . barthii ( Fig 3A filled arrows ) . Ancestry estimates for the two samples showed high proportions of both O . glaberrima and O . barthii ancestry ( S4 Fig ) . These two O . glaberrima samples were not used in subsequent analyses . Moreover , all O . barthii samples not grouped as OB-W or OB-G , as discussed above , were excluded from downstream analysis . Second , we examined other potentially spuriously grouped individuals by calculating silhouette scores for each individual [65] , which measures similarity with members of its own group compared to members of other groups ( see Materials and Method for details ) . Initially , 174 O . glaberrima samples with greater then 10× coverage were used for genotyping . A multidimensional scaling ( MDS ) plot of the population and their grouping status showed that even before the silhouette score-based filtering , there were clear separation among the OG-A , OG-B , and OG-C groups ( S5A Fig ) . But there were also several individuals whose status was questionable , as they overlapped in coordinate space with other groups . Individuals with negative silhouette scores ( i . e . potential mis-grouping ) or scores lower then 0 . 12 ( i . e . individuals with significant portions of ancestry coming from a different genetic group ) were filtered out ( S5B Fig ) to remove individuals with questionable grouping status and thus specify genetically unique populations [66] . We note some individuals that were filtered from the silhouette score-based method were likely filtered because they are admixed individuals . Omission of those individuals would lead to an underestimation of the recent admixture history of O . glaberrima . Here , our interest is in determining the long-term population histories that shaped each O . glaberrima population; hence , removal of those recently admixed individuals are necessary . This last filtering process resulted in 94 individuals , which we refer to as the core set population ( see S3 Table for list of accessions ) . MDS plot of this core set population showed clear separation among each other ( S5C Fig ) , suggesting these are genetically distinct populations ( S5D Fig ) . This core set population was used to infer the population relationship within O . glaberrima . To determine the population relationships , we also included polymorphism data from the OB-W group individuals with greater then 10× coverage . Because our grouping is based on K = 7 ancestry ( Fig 2 ) , which had two population groups for OB-W , we divided the OB-W group into two ( OB-W1 and OB-W2 ) based on the common ancestor they shared in the phylogenomic tree ( Fig 3A ) . For an outgroup population , polymorphism data from six O . rufipogon individuals with greater then 10× coverage were used [35] . An MDS plot of the nucleotide variation showed clear separation among the three species and separation within species depending on the population grouping status ( S6 Fig ) . Using the core set population , we inferred the population relationships between the five genetic groups of O . glaberrima with Treemix [67] . For the graph rooted with O . rufipogon population , without modeling any migration events the OB-W1 group were sister to all O . glaberrima ( Fig 4 ) . This model without any migration events was able to explain 99 . 4% of the variance , suggesting most of the allele frequency variability in the data can be explained without evoking migration between groups . Nevertheless , residuals from the covariance matrix suggested several population pairs could be more closely related ( population pairs with positive residuals ) compared to the best-fitting tree . Fitting models with 1 , 2 , and 3 migration events brought marginal increase in the variance explained for each migration model ( variance increases as 99 . 8% , 99 . 9% , and 99 . 94% respectively ) . Fitting 1 and 2 migration events suggested an admixture event between a population ancestral to OB-W1 and modern OG-A1 or OG-B ( Fig 4 ) . This suggests an unsampled O . barthii population may have admixed with OG-A1 and/or OG-B population . The first within-O . glaberrima admixture , specifically between OG-A1 and OG-B , was observed in the model fitting 3 migration events ( S7 Fig ) . But the f4 test [68] indicated no significant deviation from non-admixed topology for the tree [[wild rice , OG-B] , [OG-A1 , OG-A2]] , suggesting the 3 migration model is an overfitted model ( see S4 Table for f4 test result ) . Collectively , our analysis suggests the O . glaberrima population could be modeled as a bifurcating tree-like population , with small ancient admixture events from O . barthii genetic groups . Here then , it is tempting to interpret the O . glaberrima population topology , specifically the order of splitting of each genetic group , as the order of the domestication/diversification events . However , we should note that the topology changes with and without modeling migration , and in higher migration models several population pairs ( e . g . based on the residuals between OG-C1 and OB-W2 ) are still not well fitted , while bootstrap support for several internal branches are low . Thus , while this analysis provides an initial framework for depicting population relationships , one should exercise caution in over-interpreting the inferred trees . Previous molecular studies have argued the close genetic affinities of some west African O . barthii ( namely the OB-G group in this study ) to O . glaberrima , as evidence of the former being the progenitor population of African rice [13 , 69] . We thus examined the properties of the OB-G group in relation to OB-W and O . glaberrima . First , we found that the level of population differentiation between OB-G and O . glaberrima was low ( ~ 0 . 06 ) ( Fig 5A ) , almost comparable to the level seen between O . glaberrima genetic groups ( ~0 . 09 ) . In contrast , there is a higher level of differentiation between each O . glaberrima genetic group and OB-W ( Fst ~ 0 . 26 ) . Similarly , OB-G group also had high levels of differentiation to OB-W ( Fst ~ 0 . 21 ) . Second , we examined levels of linkage disequilibrium ( LD ) decay , as wild and domesticated populations have different LD profiles , due to the latter undergoing domestication-related bottlenecks and selective sweeps [70] . In the African rice group , as expected , all O . glaberrima genetic groups had higher levels of LD compared to the OB-W group ( Fig 5B ) . The OB-G group also had high levels of LD that was comparable to those observed in O . glaberrima , although compared to other OG groups , the OB-G group had longer tracts of LD . Finally , genome-wide polymorphism levels for the OB-G group were also comparable between OB-G and O . glaberrima . Specifically , compared to the OB-W group , SNP levels and Tajima’s D [71] were significantly lower in both OB-G and O . glaberrima ( S8 Fig ) . Together , the levels of genetic differentiation , linkage disequilibrium , SNP levels and patterns , all suggest that the OB-G genomes resemble O . glaberrima more than O . barthii . Furthermore , the majority of the OB-G samples carried at least one domestication mutation ( see domestication gene haplotype analysis section for detail ) , calling into question its status as the wild progenitor . In contrast all OB-W individuals do not carry the causal mutation/deletion at known domestication genes . All in all , this suggests the OB-G population is actually O . glaberrima that was mislabeled as O . barthii . It is also possible that this population may represent feral weedy rice [72] , resulting from the hybridization of domesticated and wild African rice; this is certainly consistent with the increased LD structure within OB-G [73] . While the different demographic histories between the source populations can generate an overall negative Tajima’s D for the resulting admixutre population [74] . Together , our results suggest that OB-G may have formed after the domestication event and supports a de-domestication ( endoferality ) origin for that group [57] . To further identify the domestication origin ( s ) of O . glaberrima , we examined the haplotypes for the domestication genes involved in erect plant growth ( PROG1 ) and the non-shattering phenotype ( sh4 and sh1 ) in both wild and domesticated African rice . We took an approach we term functional phylogeography , where we examined the haplotype structure surrounding the domestication gene of interest [29] , inferred a haplotype phylogenetic network , and determined the geographic origin and spread of the functional mutation by comparing the geographic distributions of haplotypes in wild and domesticated African rice in a phylogenetic context . Because we focused on the non-recombining region surrounding a domestication gene , there were only a few sites being analyzed between O . glaberrima and O . barthii . However , we were specifically interested in those few mutations that differ between the domestication gene haplotype and the progenitor gene haplotype , and used those differences to build the haplotype phylogenetic network . The PROG1 gene was first identified as a domestication gene in the Asian rice O . sativa . A mutation in this gene causes the plant to grow erect in both Asian and African rice , increasing growing density and enhancing photosynthesis efficiency for higher grain yields [19 , 20 , 75 , 76] . Our analysis of O . sativa PROG1 gene orthologs in O . glaberrima and O . barthii indicates that this gene is missing only in O . glaberrima ( S5 Table ) . We expanded the analysis to our population dataset , and a sequencing read depth analysis found PROG1 was missing in all O . glaberrima landraces ( Table 1 ) . None of the OB-W individuals had the PROG1 deletion and all but two of the OB-G individuals had the PROG1 deletion . Synteny of the genes immediately surrounding O . sativa PROG1 was maintained in both O . glaberrima and O . barthii , suggesting the PROG1 gene is deleted specifically in O . glaberrima . Because of its importance in early domestication and lack of gene structure in O . glaberrima , we considered PROG1 as a candidate domestication gene in African rice and examined the population genetics of the PROG1 gene in O . glaberrima and O . barthii . We note this is the first candidate domestication gene that has been identified where the causal mutation is fixed in all O . glaberrima population . We first examined whether the PROG1 locus showed evidence for positive selection in O . glaberrima , using genome-wide sliding window analysis of the ratio of polymorphism between the wild OB-W group to all domesticated O . glaberrima ( πw/ πD ) . A domestication-mediated selective sweep would lead to a reduction in nucleotide variation around the target domestication gene , but only within the domesticated group . Because PROG1 is deleted in O . glaberrima , the selection signal will only persist around the candidate deletion region . Spanning 10 kbp of the candidate deletion region , πw/ πD is within the top 1% value , and this is observed regardless of whether the O . glaberrima or O . barthii genome was used as the reference genome in SNP calling ( S9 Fig ) . This is consistent with the PROG1 region having gone through a selective sweep during O . glaberrima domestication . Cubry et al . [15] has also independently found evidence of a selective sweep in the PROG1 region of O . glaberrima , supporting our finding that this region has been a target of domestication-related selection . Polymorphisms surrounding the PROG1 deletion comprised a single unique haplotype segregating across all O . glaberrima samples and most of the OB-G samples ( Fig 6A ) . A haplotype network of a non-recombining 5 kbp region immediately upstream of the deletion showed that all individuals with the deletion belonged to the same major haplotype group , with the dominant haplotype I , as well as peripheral haplotypes III , VII , and VIII ( Fig 6D ) . Maximum-likelihood tree of the region surrounding PROG1 collapsed all O . glaberrima into a single phylogenetic group ( S10A Fig ) , which suggests a single origin for the deletion . We tabulated the geographic distributions of PROG1 haplotypes ( S6 Table ) . PROG1 haplotype VII is the earliest haplotype with the deletion and is found in an OB-G individual from Cameroon . The ancestral non-deleted PROG1 haplotype was carried by haplogroup IV ( Fig 6D ) , which was most closely related to all haplotypes with the PROG1 deletion , and was made up of three OB-W individuals: IRGC103912 ( Tanzania ) , IRGC105988 ( Cameroon ) , and WAB0028882 ( Cameroon ) . The downstream region of the deletion was consistent with what is observed in the upstream region ( S11 Fig ) . Twenty-two polymorphic sites from a non-recombining 7 kbp downstream region show the same OB-W individuals ( IRGC105988 and WAB0028882 ) , both from Cameroon , were the most closely related haplotype to the PROG1 deletion haplotype . Maximum-likelihood trees of both the upstream and downstream regions also showed these two individuals to be the sister group to all O . glaberrima samples ( S10A Fig ) . Together , the geographic distribution of the PROG1 region haplotypes suggest that the PROG1 deletion may have occurred in a wild progenitor closely related to those found in Cameroon , Central Africa , and spread throughout West Africa to the different O . glaberrima genetic groups ( Fig 6G ) . The PROG1 conclusion must be tempered , however , by an acknowledgment that the sample size of ancestral haplotypes is small ( n = 3 ) . Interestingly , a similar observation has been made in O . sativa where all Asian rice subpopulations are monophyletic in the PROG1 region , but genome-wide the different Asian rice variety groups/subspecies do not share immediate common ancestors [35 , 77] . Evidence for a selective sweep around the causal domestication mutation , a C-to-T nonsense mutation at position 25 , 152 , 034 leading to a loss-of-function allele ( Fig 6B arrow ) , has been previously shown [78 , 79] for the sh4 gene ( O . glaberrima chromosome 4:25 , 150 , 788–25 , 152 , 622 ) . The haplotype structure around the sh4 gene showed most of the O . glaberrima landraces carried the causal domestication mutation ( Fig 6B ) . Several individuals within OG-A1 group , including the reference genome , did not carry the causal mutation but still had long tracks of homozygosity at the sh4 locus ( Fig 6B star ) . A four-gamete test [80] of the 4 kbp upstream and 2 kbp downstream region spanning the sh4 gene detected evidence of recombination , within the O . barthii population ( both OB-G and OB-W ) and but not within O . glaberrima . A maximum-likelihood tree of the region surrounding sh4 showed all O . glaberrima populations were divided into two major phylogenetic groups , but with weak bootstrap support ( S10B Fig ) . O . glaberrima individuals without the causal mutation ( Fig 6B star ) formed their own phylogenetic group ( S10B Fig star ) . To determine the origin of the non-shattering trait , we reconstructed the haplotype network of the non-recombining region of the sh4 gene in all O . glaberrima and O . barthii genetic groups ( Fig 6E ) . Majority of the O . glaberrima and OB-G group sh4 haplotypes belonged to haplotypes II , VI , and XIII and they all shared the nonsense mutation . The two main haplotypes II and VI corresponds to the difference observed in the upstream region of the sh4 gene ( Fig 6B ) , with haplotype II arising prior to haplotype VI . The closest haplotype to II was haplotype I , which was separated by two mutations ( position 25 , 146 , 871 and the causal domestication mutation 25 , 152 , 034 ) . We tabulated the geographic distributions of O . glaberrima haplotypes II and VI/XIII , and haplotype I from the O . barthii OB-W group ( Fig 7 ) . The ancestral haplotype I is found in 13 O . barthii individuals ( 4 OB-G group and 9 OB-W group ) , and these individuals originated over a wide geographic region of West Africa that includes both coastal and inland areas ( See S8 Table for full list of members of each haplogroup and their country of origin ) . Of those in OB-W , 2 are from Mali , 2 from Nigeria and 5 are from Cameroon . Among the O . glaberrima that have the sh4 mutation , the older haplotype II is found mostly in Mali , Burkina Faso and also Guinea . Here , we made the assumption that the areas of overlap between the ancestral haplotype ( without the causal mutation ) and the derived haplotype ( with the causal mutation ) is likely the place of origin of the domestication allele . For sh4 , the distribution of haplotype II overlaps with haplotype I in Mali , pointing to Mali as being a likely place of origin for the sh4 nonsense mutation ( Fig 6H ) . The haplotypes VI and XIII thus subsequently evolved from haplotype II , which expanded over a much wider area , particularly in the Senegambia , and also to Nigeria , Cameroon and Chad . It should be noted that the sample size for haplotype I among OB-W is relatively small ( n = 9 ) leading to disjoint geographic ranges for its distribution ( Fig 6H ) . Localizing the origin of the sh4 causal mutation to Mali may be revised as more O . barthii samples are analyzed . However , haplotype II is found at highest frequency in Mali as well ( ~46% , see Fig 7 ) , which provides further support for a Malian origin of the sh4 mutations . Wu et al . [79] had first noticed that several O . glaberrima individuals in the coastal region of West Africa did not have the causal domestication mutation in the sh4 gene ( Fig 6B star ) . Our data shows that all inland O . glaberrima carries the haplotype with the nonsense mutation , and the haplotype without the nonsense mutation was indeed limited to the coastal region , specifically in the OG-A1 genetic group . The haplotype network and neighbor-joining tree suggests these individuals had distinct evolutionary histories for the sh4 gene ( Fig 6E and S10B Fig ) ; they carry haplotype VII which is confined to Guinea . The non-fixed status of the nonsense mutation suggests a role of independent mutation ( s ) in domestication for non-shattering in haplotype VII carriers . Our results showed the causal domestication mutations for the shattering genes sh1 and sh4 were not fixed in several O . glaberrima varities , suggesting their seed non-shattering may be incomplete . Thus we examined the phenotypic consequence of the domestication-related selection process of non-shattering by measuring panicle threshability for O . glaberrima . We measured the degree of non-shattering in 149 O . glaberrima accessions according to the Standard Evaluation System for Rice ( SES ) [83] . We report our measurement of panicle threshability , which is directly related to seed shattering , on a scale of 1 , 3 , 5 , 7 , and 9 , which indicates a percent shattering of less than 1% , 1–5% , 6–15% , 26–50% , and 51–100% respectively ( see S10 Table for each O . glaberrima individuals’ shattering score ) . The geographic distribution of the panicle threshability score showed an east to west gradient , where inland O . glaberrima varieties were more likely to have samples with higher threshability score values ( Fig 8A ) . Specifically , the OG-B and OG-C1 group had a mix of individuals with varying degree of shattering , while the groups closer to the coastal area , namely the OG-A1 , OG-A2 , and OG-C2 group , had predominantly individuals with panicles that were non-shattering ( Fig 8B ) . We compared the shattering scores for each genetic group by conducting Mann-Whitney U test for all pairwise combinations ( S11 Table ) . Results showed significant difference in shattering scores between the coastal and inland genetic groups ( OG-B and OG-C1 vs . OG-A1 , OG-A2 , and OG-C2 ) . Noticeably the threshability scores were consistent with the shattering mutation results ( Table 1 ) . In the coastal region , most individuals ( with the exception of the OG-A1 group and see below for detail ) had both the sh1 and sh4 mutations and were non-shattering . On the other hand , OG-B and OG-C1 were the only groups that were fixed for the sh4 casual domestication mutation while the sh1 gene was wildtype , and many individuals had higher proportions of shattering seeds . This indicates mutations in at least two shattering genes ( in the case of the OG-A2 , OG-B , and OG-C group the genes sh1 and sh4 ) were required for complete non-shattering in O . glaberrima . In the case of OG-B and OG-C1 group , selection for non-shattering was incomplete either because the group represents an ancestral population or is a result from the cultural preference on the degree of seed shattering . Samples closer to the coast and belonging to the groups OG-A1 , OG-A2 , and OG-C2 were predominantly non-shattering rice . Interestingly for the OG-A1 group , the casual domestication mutation was polymorphic in both sh1 and sh4 genes ( Table 1 ) but all varieties in OG-A1 had non-shattering seeds ( Fig 8B ) . There were 27 OG-A1 individuals with the same sh1 and sh4 allelic status ( i . e . sh1 wildtype and sh4 mutant ) as the OG-B and OG-C1 group ( S12 Table ) . However , unlike the inland group , all individuals had non-shattering seeds suggesting there may be a third shattering gene , and/or different mutations in sh1 involved in the non-shattering phenotype . In addition , all seed non-shattering OG-A1 individuals without the casual sh4 mutation ( Fig 6B star ) had the sh1 deletion ( S12 Table ) . This suggests the casual mutations for sh1 and sh4 were independently selected , possibly from different genetic backgrounds . In the end , our panicle threshability results are consistent with the population genetics result of sh1 and sh4 . Specifically , the selection process for non-shattering was either incomplete ( i . e . OG-B and OG-C1 population ) or heterogeneous ( i . e . OG-A and OG-C2 population ) , where two individuals with the same degree of threshability did not share the same casual domestication mutations in their shattering genes ( i . e . OG-A1 population ) . This opens up the possibility that domestication , at least involving seed non-shattering , does not have a single origin in O . glaberrima , but may have occurred in multiple genetic backgrounds and/or geographical regions . Our analysis of whole genome re-sequencing data in the African rice O . glaberrima and its wild ancestor O . barthii provides key insights into the geographic structure and nature of domestication in crop species . Our analysis suggests that O . glaberrima is comprised of at least 5 distinct genetic groups , which are found in different geographic areas in West and Central Africa . We find that many individuals that have been identified as O . barthii ( and which in the past have been thought to be the immediate ancestor of the domesticated crop ) form a distinct genetic group that behaves almost identically to O . glaberrima . These include similarities in LD decay , polymorphism levels , and low genetic differentiation with domesticated African rice . Moreover , several of these O . barthii individuals carried causal mutations in the key domestication genes sh4 , sh1 and PROG1 . Together this suggests these O . barthii individuals , which we collectively refer to as the OB-G group , may represent a feral O . glaberrima or may have been misidentified as the crop species . Portères hypothesized that western inland Africa near the inner Niger delta of Mali as the center of origin for O . glaberrima [8 , 9] , and this has been the commonly accepted domestication model for O . glaberrima [84] . Under this single center of origin model , O . glaberrima from the OG-C1 genetic group ( closest to the inner Niger delta ) would have acquired key domestication mutations before spreading throughout West Africa . Here , we suggest that the domestication of O . glaberrima may be more complex . Phylogeographic analysis of three domestication loci indicates that the causal mutations associated with the origin of O . glaberrima may have arisen in three different areas . Phenotype assay of panicle threshability , a core early plant domestication trait [28] , showed that the selection for seed non-shattering was incomplete in several inland O . glaberrima samples . Within the coastal O . glaberrima samples , almost all individuals have non-shattering seeds but the casual domestication mutations in two key shattering genes ( sh1 and sh4 ) are not fixed . Our results support a view , in which domestication has largely been a long protracted process , often involving thousands of years of transitioning a wild plants into a domesticated state [82 , 85 , 86] . If this indeed happened for O . glaberrima , our study suggests this protracted period of domestication had no clear single center of domestication in African rice . Instead domestication of African rice was likely a diffuse process involving multiple centers [86–88] . In this model , cultivation may have started at a location and proto-domesticates spread across the region with some ( but maybe not all ) domestication alleles . Across the multiple regions , the differing environmental conditions and cultural preferences of the people domesticating this proto-glaberrima resulted in differentiation into distinct genetic groups . Temporal and spatial variation in the domestication genes resulted in causal mutations for domestication traits appearing at different parts of the species range . The genetic and geographic structure in this domesticated species suggests that admixture might have allowed local domestication alleles to spread into other proto-domesticated O . glaberrima genetic groups in different parts of West and Central Africa . This would have facilitated the development of modern domesticated crop species , which contain multiple domestication alleles sourced from different areas . In the end , these gradual changes occurring across multiple regions provided different mutations at key domestication genes , which ultimately spread and came together to form modern O . glaberrima . There has been intense debate on the nature of domestication , and recently ( with particular emphasis on early Fertile Crescent domestication ) discussion on whether this process proceeds in localized ( centric ) vs . a diffuse manner across a wider geographic area ( non-centric ) [82 , 87 , 89] . As we begin to use more population genomic data and whole genome sequences , as well as identify causal mutations associated with key domestication traits , we can begin to study the interplay between geography , population structure and the evolutionary history of specific domestication genes and reconstruct the evolutionary processes that led to the origin and domestication of crop species . Moreover , a functional phylogeographic approach , as demonstrated here , could provide geographic insights into key traits that underlie species characteristics , and may allow us to understand how functional traits originate and spread across a landscape .
O . glaberrima and O . barthii samples were ordered from the International Rice Research Institute and their accession numbers can be found in S1 Table . DNA was extracted from a seedling stage leaf using the Qiagen DNeasy Plant Mini Kit . Extracted DNA from each sample was prepared for Illumina genome sequencing using the Illumina Nextera DNA Library Preparation Kit . Sequencing was done on the Illumina HiSeq 2500 –HighOutput Mode v3 with 2×100 bp read configuration , at the New York University Genomics Core Facility . Raw FASTQ reads are available from NCBI biproject ID PRJNA453903 . Raw FASTQ reads from the study Wang et al . [13] and Meyer et al . [14] were downloaded from the sequence read archive ( SRA ) website with identifiers SRP037996 and SRP071857 respectively . FASTQ reads were preprocessed using BBTools ( https://jgi . doe . gov/data-and-tools/bbtools/ ) bbduk program version 37 . 66 for read quality control and adapter trimming . For bbduk we used the option: minlen = 25 qtrim = rl trimq = 10 ktrim = r k = 25 mink = 11 hdist = 1 tpe tbo; which trimmed reads below a phred score of 10 on both sides of the reads to a minimum length of 25 bps , 3' adapter trimming using a kmer size 25 and using a kmer size of 11 for ends of read , allowing one hamming distance mismatch , trim adapters based on overlapping regions of the paired end reads , and trim reads to equal lengths if one of them was adapter trimmed . FASTQ reads were aligned to the reference O . glaberrima genome downloaded from EnsemblPlants release 36 ( ftp://ftp . ensemblgenomes . org/pub/plants/ ) . Read alignment was done using the program bwa-mem version 0 . 7 . 16a-r1181 [90] . Only the 12 pseudomolecules were used as the reference genome and the unassembled scaffolds were not used . PCR duplicates during the library preparation step were determined computationally and removed using the program picard version 2 . 9 . 0 ( http://broadinstitute . github . io/picard/ ) . Using the BAM files generated from the previous step , genome-wide read depth for each sample was determined using GATK version 3 . 8–0 ( https://software . broadinstitute . org/gatk/ ) . Because of the differing genome coverage between samples generated from different studies , depending on the population genetic method we used different approaches to analyze the polymorphic sites . A complete probabilistic framework without hard-calling genotypes , was implemented to analyze levels of polymorphism ( including estimating θ , Tajima’s D , and FST ) , population relationships ( ancestry proportion estimation and phylogenetic relationship ) , and admixture testing ( ABBA-BABA test ) . For methods that require genotype calls , we analyzed samples that had greater then 10× genome coverage . Details are shown below . We used ANGSD version 0 . 913 [53] and ngsTools [54] which uses genotype likelihoods to analyze the polymorphic sites in a probabilistic framework . ngsTools uses the site frequency spectrum as a prior to calculate allele frequencies per site . To polarize the variants the O . rufipogon genome sequence [36] was used . Using the O . glaberrima genome as the reference , the O . rufipogon genome was aligned using a procedure detailed in Choi et al . [37] . For every O . glaberrima genome sequence position , the aligned O . rufipogon genome sequence was checked , and changed to the O . rufipogon sequence to create an O . rufipogon-ized O . glaberrima genome . Gaps , missing sequence , and repetitive DNA were noted as ‘N’ . For all analysis we required the minimum base and mapping quality score per site to be 30 . We excluded repetitive regions in the reference genome from being analyzed , as read mapping to these regions can be ambiguous and leading to false genotypes . The site frequency spectrum was then estimated using ANGSD with the command: For each genetic group a separate site frequency spectrum was estimated and the options–minInd , -setMinDepth , and–setMaxDepth were changed accordingly . Parameter minInd represent the minimum number of individuals per site to be analyzed , setMinDepth represent minimum total sequencing depth per site to be analyzed , and setMaxDepth represent maximum total sequencing depth per site to be analyzed . We required–minInd to be 80% of the sample size , -setMinDepth to be one-third the average genome-wide coverage , and–setMaxDepth to be 2 . 5 times the average genome-wide coverage . Using the site frequency spectrum , θ was calculated with the command: The θ estimates from the previous command was used to compute sliding window values for Tajima’s θ and D [71] with the command: thetaStat do_stat $Theta–nChr $Indv–win 10000 –step 10000 The option nChr is used for the total number of samples in the group being analyzed . Window size was set as 10 , 000 bp and was incremented in non-overlapping 10 , 000 bp . FST values between pairs of population were also calculated using a probabilistic framework . Initially , we calculated the joint site frequency spectrum ( 2D-SFS ) between the two populations of interest with the command: Each population’s site frequency spectrum estimated from previous step is used to estimate the 2D-SFS . With the 2D-SFS FST values were calculated with the command: FST values were calculated in non-overlapping 10 , 000 bp sliding windows . For the sliding windows calculated for θ , Tajima’s D , and FST values , we required each window to have at least 30% of the sites with data or else the window was discarded from being analyzed . Ancestry proportions were estimated using NGSadmix [55] . Initially , genotype likelihoods were calculated using ANGSD with the command: To reduce the impact of LD would have on the ancestry proportion estimation , we randomly picked a polymorphic site in non-overlapping 50 kbp windows . In addition we made sure that the distance between polymorphic sites were at least 25 kbp apart . We then used NGSadmix to estimate the ancestry proportions for K = 2 to 9 . For each K the analysis was repeated 100 times and the ancestry proportion with the highest log-likelihood was selected to represent that K . Phylogenetic relationships between samples were reconstructed using the genetic distance between individuals . Distances were estimated using genotype posterior probabilities from ANGSD command: Genotype posterior probability was used by NGSdist [58] to estimate genetic distances between individuals , which was then used by FastME ver . 2 . 1 . 5 [91] to reconstruct a neighbor-joining tree . Tree was visualized using the website iTOL ver . 3 . 4 . 3 ( http://itol . embl . de/ ) [92] . Principal component analysis were also conducted using genotype likelihoods . Genotype posterior probabilities from ANGSD command: The genotype posterior probability was then used by the program ngsCovar [54] to conduct the principal component analysis . Since several methods require genotype calls for analysis SNP calling was also performed . Samples with greater than or equal to 10× genome coverage ( GE10 dataset ) was considered to ensure sufficient read coverage for each site at the cost of excluding individuals from genotype calling . These were 174 individuals that belonged to the genetic grouping designated by this study , and full list of individuals can be found in S13 Table . For each sample , genotype calls for each site was conducted using the GATK HaplotypeCaller engine under the option `-ERC GVCF`mode to output as the genomic variant call format ( gVCF ) . The gVCFs from each sample were merged together to conduct a multi-sample joint genotyping using the GATK GenotypeGVCFs engine . Genotypes were divided into SNP or INDEL variants and filtered using the GATK bestpractice hard filter pipeline [93] . For SNP variants we excluded regions that overlapped repetitive regions and variants that were within 5 bps of an INDEL variant . We then used vcftools version 0 . 1 . 15 [94] to select SNPs that had at least 80% of the sites with a genotype call , and exclude SNPs with minor allele frequency <2% to remove potential false positive SNP calls arising from sequencing errors or false genotype calls . The GE10 dataset was used for this analysis , as it requires hard-called genotypes . The O . glaberrima samples were grouped according to the grouping scheme designated in this study ( Fig 3 ) , and any members that were more similar to other grouping then its own were examined by estimating their silhouette scores [65] . Using the program PLINK version 1 . 9 [95] for calculating genetic distances from all pairwise comparisons , silhouette scores were calculated using the formula: s ( i ) =[b ( i ) −a ( i ) ]/max{a ( i ) , b ( i ) } ( 1 ) where i represents an individual , s the silhouette score , a the average genetic distance to members of own group , and b the average genetic distance to members of foreign group . Individuals with negative silhouette scores were filtered out . After filtering , using the remaining individuals the silhouette score based filtering method was iteratively conducted until all individuals had silhouette scores higher then 0 . 1 . To obtain the outgroup nucleotide variants we downloaded raw sequencing data for six O . rufipogon species corresponding to the Or-C and Or-D clade , which were shown to contain the least amount of domesticated Asian rice admixture from feralization [96] . These samples have identifiers W0137 , W1739 , W1807 , W0170 , W0630 , and W2263 with SRA run accession IDs of DRR088674 , ERR224552 , DRR088680 , ERR2245549 , DRR001185 , and DRR088691 . O . rufipogon raw FASTQ reads were aligned to the O . glaberrima reference genome as outlined in our previous steps . GATK HaplotypeCaller engine was used for calling genotypes but the multi-sample joint genotyping step for the six O . rufipogon samples were limited to polymorphic sites that overlapped the SNP positions analyzed in the silhouette score analysis . Population relationships were examined as admixture graphs using Treemix version 1 . 13 [67] . SNP calls from the core set population was used to calculate the allele frequencies for each genetic group . One hundred SNPs were analyzed together as a block to account for the effects of LD between SNPs . The O . rufipogon variation was used as the outgroup and a Treemix model assuming 0–3 migration events were fitted . The four-population test [68] was conducted using the fourpop program from the Treemix package . Genome-wide levels of LD ( r2 ) was estimated with the GE10 dataset and using the program PLINK . LD was calculated for each genetic group separately across a non-overlapping 1Mbp window and between variants that are at most 99 , 999 SNPs apart . LD data was summarized by calculating the mean LD between a pair of SNPs in 1 , 000 bp bins . A LOESS curve fitting was applied for a line of best fit and to visualize the LD decay . We downloaded protein coding sequences for O . sativa , O . glaberrima , and O . barthii from EnsemblPlants release 36 . An all-vs-all reciprocal BLAST hit approach was used to determine orthologs between species and paralogs within species . We used the program Orthofinder ver . 1 . 19 [97] to compare the proteomes between and within species for ortholog assignment . Orthofinder used the program DIAMOND ver . 0 . 8 . 37 [98] for sequence comparisons . Synteny based on the O . sativa sh1 gene ( Ossh1; O . sativa cv . japonica chromosome 3:25197057–25206948 ) indicated orthologs surrounding Ossh1 was found in chromosome 3 of O . barthii and on an unassembled scaffold named Oglab03_unplaced035 in O . glaberrima ( S14 Table ) . The sh1 gene was missing in O . glaberrima suggesting the gene deletion may have led to complex rearrangements that prevented correct assembly of the region in the final genome assembly . Because of this we used the O . barthii genome sequence to align raw reads and call polymorphic sites for downstream analysis . The approximate region of the deletion in the O . barthii genome coordinate was examined by looking at the polymorphic sites , since our quality control filter removed polymorphic sites if it had less than 80% of the individuals with a genotype call . Between the genomic positions at O . barthii chromosome 3 position 23 , 100 , 000–23 , 130 , 000 , no polymorphic sites passed the quality control filter ( S13 Fig ) and contained the gene Obsh1 . Between the region at O . barthii chromosome 7 position 2 , 655 , 000–2 , 675 , 000 there was also no polymorphisms passing the filter and contained the gene ObPROG1 ( S14 Fig ) . Gene deletion was inferred from comparing the read depth of a genic region inside and outside a candidate deletion region . Read depth was measured using bedtools ver . 2 . 25 . 0 [99] genomecov program . Individuals with and without the deletion were determined by comparing the median read coverage of the domestication gene within the candidate deletion region , to a gene that is outside the deletion region . We checked the orthologs to make sure the gene outside the deletion region existed in O . barthii , O . glaberrima , and O . sativa . To determine the sh1 deletion status we examined its read depth and compared it to the O . barthii gene OBART03G27620 that was upstream and outside the candidate deletion region . Ortholog of OBART03G27620 is found in both O . sativa ( Os03g0648500 ) and O . glaberrima ( ORGLA03G0257300 ) . To determine the deletion status of PROG1 gene we examined its read depth and compared to O . barthii gene OBART07G03440 . Ortholog of OBART07G03440 is found in both O . sativa ( Os07g0153400 ) and O . glaberrima ( ORGLA07G0029300 ) . Because some individuals had low genome-wide coverage ( S1 Table ) there is the possibility that some of those individuals had been detected as false positive deletion events . There are two main reasons we believe the deletions are likely to be present even for low coverage individuals . For example for the sh1 deletion , ( i ) all individuals had at least a median coverage of ~1× in the OBART03G27620 gene ( S15 Table ) suggesting read coverage may be low but if the gene is not deleted it is evenly distributed across a gene , and ( ii ) even comparing individuals with and without the sh1 deletion that had a ~1× median coverage in the non-deleted OBART03G27620 gene , there were clear differences in the sh1 gene coverage ( S15 Fig ) where the individuals with the deletion always had a median coverage of zero . Gene names for the non-shattering phenotype have unfortunately varied between different Oryza studies . Genetic studies comparing Asian rice O . sativa cv . Japonica and its wild progenitor O . rufipogon had identified a single dominant allele responsible for non-shattering and named the locus as Sh3 [100 , 101] . The causal gene was later identified on chromosome 4 and was given a new name as sh4 [17] . Studies have used the names Sh3 and sh4 synonymously as the common gene name for the gene with locus ID Os04g0670900 [78] . Lv et al . [81] had found an O . glaberrima specific gene deletion in chromosome 3 that caused a non-shattering phenotype and named this gene as SH3 . SH3 belongs to a YABBY protein family transcription factor . Using the SH3 coding sequence in O . barthii ( ObSH3 ) , which the gene is not deleted , orthologs were found in maize ( B4FY22 ) , barley ( M0YM09 ) , and Brachypodium ( I1GPY5 ) [81] . We discovered this group of proteins belonged to a group identified in Plant Transcription Factor Database ver 4 . 0 [102] under the ID OGMP1394 . The O . sativa gene member of this group was gene ID Os03g0650000 , which has previously been identified as a gene involved in non-shattering [103] . Thus , ObSH3 and Os03g0650000 are orthologs of each other and Os03g0650000 has been named as sh1 . Here , we followed the guideline recommended by Committee on Gene Symbolization Nomenclature and Linkage ( CGSNL ) [104] to designate SH3 from Lv et al . [81] as sh1 to avoid using the overlapping gene name sh3 . To investigate the haplotype structure around the domestication genes we used all individuals from O . glaberrima , OB-G , and OB-W population regardless of the genome coverage . The O . glaberrima and O . barthii genome were used as reference to align the raw reads and call polymorphisms as outlined above . Missing genotypes were then imputed and phased using Beagle version 4 . 1 [105] . We used vcftools to extract polymorphic sites around a region of interest . The region was checked for evidence of recombination using a four-gamete test [80] , to limit the edges connecting haplotypes as mutation distances during the haplotype network reconstruction . To minimize false positive four-gamete test results caused from technical errors such as genotype error and sequencing error , if the observed frequency of the fourth haplotype was below 1% we considered the haplotype an error and did not consider it as evidence of recombination . If a region had evidence of recombination we checked if the recombination was limited to the wild or domesticated African rice . If recombination was only detected in the wild population then we determined the pair of SNPs that failed the four-gamete test . Here , because the four-gamete test did not detect any evidence of recombination in the O . glaberrima population , the fourth haplotype observed in the wild population is only limited to O . barthii and do not provide any information with regard to the direct origin of the O . glaberrima haplotypes . Hence , we removed individuals with the fourth haplotype and estimated the haplotype network of the region . Haplotype network was reconstructed using the R pegas [106] and VcfR [107] package , using the hamming distance between haplotypes to construct a minimum spanning tree . For each domestication gene and its surrounding region , a phylogenetic tree was reconstructed by sampling a single haplotype for each individual . Bootstrap replicated phylogenetic trees were built using RAxML [108] and plotted with iTOL . O . glaberrima landraces were grown during the 2018 dry season at the International Rice Research Institute ( IRRI ) block L4 ( 14°09'34 . 6"N 121°15'42 . 4"E ) experimental field . At maturity , when at least 85% of the grains on a panicle are matured [109 , 110] , panicles were harvested and evaluated for threshability using a established method by IRRI [111] . In brief , a total of 6 plants for each landrace from three plot replicates were sampled . During panicle threshability measurement , each panicle was grasped to apply slight pressure . Grains detached from the panicle and panicles intact with grains were collected . The numbers of grains that detached and remained attached were counted separately to obtain the percentage of shattered grains [83] . Percent shattering were converted to panicle threshability scores according to the Standard Evaluation System for Rice [83] .
|
For many crops it is not clear how they were domesticated from their wild progenitors . Transition from a wild to domesticated state required a series of genetic changes , and studying the evolutionary origin of these domestication-causing mutations are key to understanding the domestication origins of a crop . Moreover , population comparisons provide insight into the relationship between wild and cultivated populations and the evolutionary history of domestication . In this study , we investigated the domestication history of Oryza glaberrima , a rice species that was domesticated in West Africa independent from the Asian rice species O . sativa . Using genome-wide data from a large sample of domesticated and wild African rice samples we did not find evidence that supported the established domestication model for O . glaberrima—a single domestication origin . Rather , our evidence suggests the domestication process for African rice was initiated in multiple regions of West Africa , caused potentially by the local environments and cultivation preference of people . Hence domestication of African rice was a multi-regional process .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] |
[
"biogeography",
"animal",
"types",
"taxonomy",
"ecology",
"and",
"environmental",
"sciences",
"niger",
"domestic",
"animals",
"population",
"genetics",
"geographical",
"locations",
"animals",
"genetic",
"mapping",
"mutation",
"phylogenetics",
"data",
"management",
"rice",
"phylogenetic",
"analysis",
"experimental",
"organism",
"systems",
"nonsense",
"mutation",
"population",
"biology",
"plants",
"zoology",
"africa",
"research",
"and",
"analysis",
"methods",
"geography",
"computer",
"and",
"information",
"sciences",
"grasses",
"animal",
"studies",
"phylogeography",
"evolutionary",
"systematics",
"people",
"and",
"places",
"haplotypes",
"eukaryota",
"plant",
"and",
"algal",
"models",
"heredity",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"organisms"
] |
2019
|
The complex geography of domestication of the African rice Oryza glaberrima
|
Developmental patterning involves the progressive subdivision of tissue into different cell types by invoking different genetic programs . In particular , cell-cell signaling is a universally deployed means of specifying distinct cell fates in adjacent cells . For this mechanism to be effective , it is essential that an asymmetry be established in the signaling and responding capacities of the participating cells . Here we focus on the regulatory mechanisms underlying the role of the neuralized gene and its protein product in establishing and maintaining asymmetry of signaling through the Notch pathway . The context is the classical process of “lateral inhibition” within Drosophila proneural clusters , which is responsible for distinguishing the sensory organ precursor ( SOP ) and non-SOP fates among adjacent cells . We find that neur is directly regulated in proneural clusters by both proneural transcriptional activators and Enhancer of split basic helix-loop-helix repressors ( bHLH-Rs ) , via two separate cis-regulatory modules within the neur locus . We show that this bHLH-R regulation is required to prevent the early , pre-SOP expression of neur from being maintained in a subset of non-SOPs following SOP specification . Lastly , we demonstrate that Neur activity in the SOP is required to inhibit , in a cell non-autonomous manner , both neur expression and Neur function in non-SOPs , thus helping to secure the robust establishment of distinct cell identities within the developing proneural cluster .
The specification of discrete cell identities during metazoan development often requires the establishment of disparate genetic programs in adjacent cells . The Notch signaling pathway is ideally suited to this task , since it mediates direct cell-cell interactions via contact between transmembrane ligands and receptors . Acting in this fashion , it is responsible for distinguishing the gene expression programs of adjacent cells in multiple developmental settings , including boundary formation between neighboring cell populations; binary cell fate specification between daughter cells in a cell lineage; and “lateral inhibition” within a cluster of cells with initially similar fate [1] . If such binary partitioning of cell fate is to function with high fidelity , it ultimately requires the creation of strong disparities in Notch signaling and responding capacity between “sending” and “receiving” cells . In principle , this can be achieved in a number of ways , most obviously via differences in ligand and/or receptor protein levels [2] . In contexts in which such differences are not observed , however , other mechanisms must come into play . One example is the classical process of lateral inhibition within proneural clusters ( PNCs ) in Drosophila . The cells that comprise the mechanosensory bristles of Drosophila are products of serial asymmetric cell divisions , beginning with individual sensory organ precursor cells ( SOPs ) that are specified by Notch signaling within PNCs . PNCs are defined by the expression of basic helix-loop-helix ( bHLH ) transcriptional activators , encoded by the “proneural” genes achaete ( ac ) and scute ( sc ) , that confer upon PNC cells the potential to adopt the SOP fate [3 , 4] . Due in part to their positive auto-regulatory activity , the expression of proneural genes is elevated in cells that will become SOPs . SOPs use Notch signaling to inhibit neighboring PNC cells from becoming SOPs [5 , 6] . Notch receptor on the surface of these “non-SOP” cells is activated by cell-surface ligand on the SOP , resulting in the release of the intracellular domain ( ICD ) of the receptor from the plasma membrane and its translocation to the nucleus . There , the Notch ICD forms a complex with the pathway’s transducing transcription factor Suppressor of Hairless [Su ( H ) ] , converting it from a repressor to an activator and stimulating the expression of a collection of SOP-inhibitory target genes [1] . The Enhancer of split [E ( spl ) ] and Bearded ( Brd ) gene complexes encode two major classes of Notch effectors , the E ( spl ) bHLH transcriptional repressors ( bHLH-Rs ) and the Brd family members ( BFMs ) [7–11] . The bHLH-Rs prevent non-SOPs from becoming SOPs in part by reducing proneural auto-activation [12] , and also by repressing transcription of SOP-specific genes [13] . BFMs function very differently—they bind directly to the E3 ubiquitin ligase Neuralized ( Neur ) , thereby blocking its direct interaction with the ICDs of the Notch ligands Delta ( Dl ) and Serrate ( Ser ) [14 , 15] . Neur expression is strongly upregulated in SOPs , and mono-ubiquitination of ligand ICDs by Neur promotes ligand endocytosis and their ability to activate the Notch receptor [16–18] . Central to the establishment and maintenance of the two distinct PNC cell fates is the emergence of an imbalance in Notch signaling capacity between the SOP and non-SOPs , despite the fact that all PNC cells express both ligand and receptor . Dl has been proposed as a direct target of the proneural proteins in neural precursor ( NP ) cells [19] , which in principle could lead to upregulation of its expression specifically in SOPs . However , NP specification can proceed normally when Dl is uncoupled from proneural regulation [20 , 21] , and similar levels of nascent Dl transcript have been observed in microchaete SOPs and surrounding non-SOPs [22] . By contrast , direct proneural regulation of neur is an attractive alternative , because of the gene’s high SOP-specific expression and its important role in Notch-mediated lateral inhibition . Prior investigations of Neur function in SOP specification have addressed neither the transcriptional regulation of neur nor the specific processes by which functional Neur activity is prevented in non-SOPs . Here we directly address the mechanisms by which Neur contributes to the establishment of unequal signaling capacity between SOPs and non-SOPs . In a previous report [13] , we described the identification and functional activities of neur4D and neur1B , two enhancer modules that drive neur expression in NP cells . In the present study , we investigate the transcription factor inputs and regulatory logic that these modules use to generate the NP specificity . We demonstrate that neur is a direct target of both the proneural proteins and the bHLH-Rs , acting through the neur4D and neur1B enhancers . In particular , we identify a conserved proneural motif type that is capable of binding both the Ac/Sc and Atonal classes of proneural activators , and show that mutation of bHLH-R binding motifs causes expansion of both neur transcript and protein into non-SOP territories . We also provide conclusive evidence of nascent neur transcription in a small subset of PNC cells prior to SOP commitment . This analysis offers for the first time an explicit definition of the “neur group” of PNC cells [23] , and resolves the previous apparent inconsistency of complementary expression patterns for neur and BFMs . Lastly , we demonstrate the consequences of either maintaining neur expression in non-SOPs or blocking Neur activity specifically in the SOP . Together , our work shows that , through its function in promoting Notch signaling from the SOP , neur auto-inhibits , in a cell non-autonomous manner , both its proneural-dependent transcription and the function of its product Neur , and by these mechanisms helps to establish and maintain an SOP/non-SOP dichotomy in signaling capacity .
In the wing imaginal disc , both the accumulation of endogenous neur transcript and the expression of the neur4D-GFP and neur1B-GFP reporter transgenes are dependent upon proneural ac/sc gene activity in trans [13 , 24] . Consistent with direct proneural regulation of neur4D , five Ac/Sc binding motifs fitting the RCAGSTG definition ( which we refer to here as PS ) are found in this module in D . melanogaster . Moreover , four of these five are fully conserved in 11 other Drosophila species , the exception being the P5 site in D . mojavensis and D . virilis , which is changed to RCAGATG , referred to here as PA ( S1A and S2A Figs ) . By contrast , in D . melanogaster neur1B , we find only a single PS motif . This is conserved in 10/12 species , the exceptions being D . persimilis and D . pseudoobscura , in which the motif deviates to the PA form ( S1B and S2B Figs ) . Given the overall strong conservation of the PS motifs in neur4D and neur1B , we sought to assess their functional role in vivo . All PS sites in each enhancer were changed from RCAGSTG to RAAGSGG , a mutation known to abrogate binding of Ac/Da heterodimers [25] . We observed only a slight reduction in GFP expression driven by both neur4D and neur1B ( Fig 1C–1H and 1J–1M; S4B1 , S4B8 , S4B15 and S4B22 Fig ) . This result suggests two possibilities that are not mutually exclusive: First , that direct activation of the neur4D and neur1B modules by proneural factors is mediated by binding sites other than PS motifs; second , that whatever the role of proneural proteins in direct activation of the two modules , other factors are sufficient to drive their activity in SOPs . Though D . melanogaster neur1B includes only a single match to the PS motif definition , outside the D . melanogaster-simulans-sechellia sub-subgroup a second PS motif occurs within this enhancer ( S1B and S2B Figs ) . Interestingly , within the sub-subgroup this motif is changed to the PA variant . A search of D . melanogaster neur1B and orthologous regions in the other species revealed the presence of three additional conserved CAGATG sequences ( S1B and S2B Figs ) . The conservation of multiple PA motifs and the switching of orthologous motifs from PS to PA within both neur4D and neur1B prompted us to ask whether the proneural proteins are capable of binding PA motifs in an electrophoretic mobility shift assay ( EMSA ) . Indeed , we find that Sc/Da heterodimers bind probes containing both the PS and PA motifs in neur1B , but not their corresponding mutant probes ( S6A Fig ) . We next examined the consequences of mutating this expanded group of Ac/Sc-binding motifs , both PS and PA , in the context of the neur4D-GFP and neur1B-GFP reporter transgenes . There is a single PA motif in D . melanogaster neur4D that is not present in D . ananassae , D . mojavensis , D . virilis , and D . grimshawi ( S1A and S2A Figs ) , which may suggest that it is not required . Mutating this sequence in combination with the PS motifs did not result in a further decrease in reporter expression ( S6J , S6Q and S6R Fig ) . In contrast , and consistent with our EMSA data , mutation of both the PS and PA motifs in neur1B-GFP strongly reduced expression in the Ac/Sc-dependent proneural clusters of the wing imaginal disc ( Fig 1O ) , suggesting that Ac/Sc proteins directly activate neur1B through these motifs . Interestingly , this reporter mutant also lost expression in both the ventral radius of the wing imaginal disc and the chordotonal clusters of the leg imaginal discs ( Fig 1O and 1Q ) , territories in which the distantly related proneural protein Atonal ( Ato ) is active [26] . Consistent with this loss of expression , Ato has been reported to bind CAGATG sequences [26 , 27] , suggesting that Ato may also regulate neur1B through these PA motifs . Indeed , we find that Ato/Da heterodimers are capable of binding all PS and PA motifs in neur1B in vitro , but not their mutant versions ( S6A Fig ) . Our results indicate a stark contrast in the requirements for proneural motifs in the activation of the neur4D and neur1B enhancers . Since the proneural motifs in neur4D are not strictly required for its activity , we sought to examine the conservation and functional necessity of other sequence elements within this module , some of which have previously been implicated in SOP-specific expression . In addition to the PS and PA motifs , the neur4D enhancer contains several motifs of at least seven nucleotides that are identical both in sequence and in order in all 12 Drosophila genomes ( Fig 1B; S1A and S2A Figs ) . neur4D contains conserved instances of the SMCα motif [28–30]; the binding motif for the zinc-finger transcription factor Senseless ( Sens ) [31 , 32]; and three other sequences that are fully conserved in all twelve genomes , which we refer to as “mystery blocks” ( MB1 , MB2 , and MB3 ) . By mutational analysis , we examined the functional requirements for these conserved motifs in neur4D , both on their own and in combination with mutation of all the PS motifs . Mutation of the two SMCα motifs alone only slightly reduced the activity of neur4D ( S4A3 , S4A10 , S4A17 and S4A24 Fig ) . Mutating the SMCα motifs plus the PS motifs further reduced reporter gene expression , but failed to eliminate it ( S4B3 , S4B10 , S4B17 and S4B24 Fig ) . Similarly , modest reductions in neur4D activity were observed with mutation of either the Sens motif or MB2 , whereas mutation of MB1 and MB3 each resulted in slightly increased expression ( S4A2 , S4A5–S4A7 , S4A9 , S4A12–S4A14 , S4A16 , S4A19–S4A21 , S4A23 and S4A26–S4A28 Fig ) . We also assayed this series of mutant reporter genes by in situ hybridization with a GFP probe in embryos of various stages ( S5 Fig ) . Similar to the results in larval and pupal tissues , no single motif mutation eliminated reporter expression . However , whenever the PS motifs were also mutated ( S4B Fig ) we observed a consistent qualitative reduction in expression in comparison to the mutation of the motif classes individually . Since none of the motif mutants , whether on their own or in combination with the PS mutations , eliminated neur4D activity , we made a construct in which all the sites contributing weak positive input ( SMCα , Sens , MB2 , and PS ) were mutated . We observed weak GFP expression in wing imaginal discs even for this construct ( S6K Fig ) , suggesting that still other sequences in neur4D play a role in its activation . Furthermore , even the addition of the PA motif mutation to the PS+SMCα+Sens+MB2 mutant failed to yield any further reduction in wing disc expression driven by neur4D ( S6L Fig ) . Thus , the SOP-specific activation of neur4D , in contrast to that of neur1B , appears to be highly complex and require inputs from other , as-yet-unknown factors . While neur4D and neur1B exhibit a striking difference in their schemes for activation in SOPs , both enhancers contain one or more conserved motifs for binding by E ( spl ) bHLH repressor ( bHLH-R ) proteins ( Fig 1; S1 and S2 Figs ) [33] . These factors are expressed in a pattern complementary to that of neur in the PNC , due to default repression by Su ( H ) in the SOP and synergistic activation by the proneurals and Su ( H ) in non-SOPs ( the “S+P” cis-regulatory code ) [34 , 35] . All three instances of the bHLH-R core binding motif ( CACGYG ) in neur4D and neur1B are conserved in all 12 genomes ( S1 and S2 Figs ) . Based on both their pattern of expression and cis-regulatory logic , the bHLH-Rs would be predicted to confine neur expression to the SOP through the binding motifs in neur4D and neur1B . Indeed , when we mutate the two bHLH-R motifs in neur4D-GFP , we frequently observe many PNC positions in the wing imaginal disc where there is at least one GFP-positive cell in addition to the GFP- and Sens-positive SOP , usually located adjacent to the SOP ( Fig 2A’ and 2C ) . Likewise , in the 12 hr APF notum we observe many regions in between Sens-positive SOPs that display multiple GFP-positive , Sens-negative cells ( Fig 2D ) . We find that this ectopic GFP expression ( outside of the SOP ) is entirely dependent upon proneural cis-regulatory input via PS sites in neur4D ( Fig 2B’ , 2C and 2E ) . This antagonistic functional relationship between bHLH-R and PS motifs was also observed using neur1B-GFP . Mutation of the single bHLH-R motif in neur1B-GFP did not cause ectopic expression as broad as that seen by mutating the neur4D motifs; the position most regularly affected was the posterior dorsocentral . We frequently observed ectopic GFP expression at this position in neur1BRm-GFP wing discs , and it always appeared adjacent to the SOP ( Fig 2F and 2H ) . Moreover , mutating the single PS site in neur1B was sufficient to reduce this ectopic expression significantly ( Fig 2G and 2H ) . These data demonstrate a functional requirement in both neur4D and neur1B for intact bHLH-R cis-regulatory input to confine the proneural-dependent activation of these enhancers to the SOP . Because we had found , first , that both enhancers contribute to neur function in the SOP [13] and , second , that mutation of the bHLH-R input in both the neur4DRm-GFP and neur1BRm-GFP reporters causes ectopic expression in non-SOP cells , we sought to examine if this regulatory relationship can be observed in the context of the neur gene itself . To test this , we created both untagged and C-terminal GFP fusion versions of a wild-type P[acman] construct [36] containing 21 kb of the neur locus , extending into the adjacent genes , along with a variant in which the bHLH-R motifs within neur4D and neur1B are mutated ( Fig 3A ) . Examining third-instar wing imaginal discs from larvae containing the untagged constructs , we saw an expansion of neur mRNA transcript expression , particularly at the wing margin and at the chordotonal organ of the tegula ( Fig 3D and 3E ) . We quantified changes at this latter position using ImageJ software . Discs containing the bHLH-R motif mutant rescue constructs measured a statistically significant increase in the area of staining ( Fig 3F ) , as well as a very significant decrease in average white intensity ( Fig 3G ) , which is due to the increased darkness of the in situ signal . While these results clearly indicate an increase in neur transcript accumulation following disruption of bHLH-R-mediated repression , the spatial resolution of this assay is rather poor . A more conspicuous result was obtained using the GFP-tagged rescue constructs , with which we were regularly able to detect an expansion of Neur-GFP expression from the R motif mutant construct into more cells than just the specified SOPs ( Fig 3K , 3O and 3Q ) . Together , these data demonstrate that mutation of the bHLH-R binding motifs within the two neur SOP enhancers results in the failure to confine neur transcript and protein to the SOP . The logic of confining a fully functional level of Neur protein accumulation to the SOP is clear: It is critical that only one cell in the proneural cluster should have the capacity to inhibit the SOP fate in all of its neighbors . However , the very reliance on proneural input ( whether direct or indirect ) to activate neur expression in SOPs creates the possibility that neur would initially be activated in more PNC cells than just the ultimate committed SOP . Consistent with this expectation , Huang et al . observed neur reporter gene ( neurA101-LacZ ) expression in 2–3 adjacent or nearby cells during macrochaete SOP specification [37] . Likewise , Koto et al . used a neur-GAL4 driver to visualize the appearance of excess neur-positive cells during microchaete SOP determination [38] . We similarly have observed reporter gene ( neur4D-GFP ) expression in two adjacent cells prior to SOP specification , as determined by costaining with anti-Sens ( Fig 1E; see caret ) . We sought more detailed documentation of this phenomenon by detection of either neur transcript or protein during the heterochronic appearance of macrochaete SOPs in the wing imaginal disc . In the notum region of the wing disc , these SOPs are first detected in a consistent temporal order [37] . Furthermore , certain of the individual clusters ( e . g . , dorsocentral and scutellar ) develop exactly two SOPs , with one appearing early in development and the second appearing later in a stereotypical location a few nuclear diameters away ( Fig 4A ) . This developmental pattern allowed us to fix larval imaginal discs at a stage in which a cluster contained both a specified early SOP and a nearby region , the “pre-SOP domain” . Indeed , in several of these heterochronic clusters we were able to find clear examples of neur expression in multiple adjacent cells by detecting either neur nascent transcript or GFP-tagged Neur protein ( Fig 4B and 4F–4H ) . For the former experiment , we utilized the multiplex fluorescent in situ hybridization technique [39] with intron probes to simultaneously visualize nascent transcripts for neur , sca ( to mark PNC membership ) , and CG32150 ( to positively identify a committed SOP ) [24] , while also staining with Hoechst , a DNA dye to mark the nucleus . To be certain of the neur transcript detection , we used versions of the same in situ hybridization probe with two different labels simultaneously; thus , strong colocalization of these two probes unambiguously identifies cells producing neur nascent transcript . Regularly , within the dorsocentral ( DC ) and scutellar ( SC ) PNCs , one nucleus ( the posterior cell ) exhibited colocalization of strong neur probe signal in both channels , as well as a strong signal for sca and CG32150 probes , identifying the first specified SOP in each of these clusters ( Fig 4B , panels 5 and 6 ) . In these same clusters , 1–3 nuclear diameters away , we were often able to find 2–4 cells that each colocalized neur probes ( Fig 4B , panels 1 , 2 , and 4 ) . In these cells , the probe density was not as strong as in the specified SOP , nor did these cells have strongly detectable CG32150 transcript . When CG32150 transcript was detected in this region , it was confined to a single nucleus that also exhibited neur probe colocalization at an increased density . We also examined Neur protein accumulation in these pre-SOP domains , using a wild-type neur GFP-tagged rescue construct . Analogous to the in situ hybridization experiments , we co-stained with anti-Sens antibody to identify committed SOPs , and looked for Neur::GFP signal in a region a few cells away with no detectable Sens . Similar to what was seen in the neur transcript assay , we were able to detect 2–3 adjacent cells with GFP signal above background in these regions , typically in the DC and SC PNCs ( Fig 4F–4H ) . Collectively , these data indicate that prior to demonstrated SOP commitment a subset of cells in the PNC express both neur transcript and protein . As we have seen , neur expression is ultimately tightly restricted to the SOP , yet prior to specification it occurs in more than one cell . We sought to investigate the potential consequences of persistent neur expression outside of the single committed SOP . Enhanced Notch signaling due to ectopic Neur expression in non-SOPs could conceivably interfere with proper lateral inhibition in two main ways . First , it could lead to loss of normal SOPs by preventing or overcoming their commitment to this fate . Alternatively , it could allow multiple cells in the PNC to resist signaling from the SOP and become committed SOPs themselves ( perhaps due to cis-inhibition [40] ) . To explore these possibilities , we utilized two different strategies to misexpress neur and looked for manifestations of either of the predicted phenotypes . We first expressed Neur specifically in non-SOPs within the PNC using the non-SOP-specific , Notch-dependent driver mα-GAL4 . In flies bearing single copies of both the driver and UAS-neur , the dominant phenotype was missing bristles ( Fig 5A ) , which we confirmed to have resulted from loss of the SOP ( Fig 5C ) . Adding an additional copy of the driver primarily enhanced SOP loss , while adding an additional copy of the responder significantly increased the number of extra bristles ( Fig 5A ) . One complication of this strategy for misexpression is the fact that the E ( spl ) mα regulatory region is Notch-regulated [35] . Thus , if Notch signal receipt in non-SOPs is compromised , the expression of GAL4 could accordingly decrease . We therefore sought to examine the consequences of Notch-independent , uniform neur expression in mosaic tissue using the MARCM system [41] . Similar to the mα-GAL4 experiments , we observed both SOP loss ( Fig 5D–5F ) and gain ( Fig 5G–5I ) , depending upon the context . In the latter case , which we observed in the microchaete field of the pupal notum , the effect in the neur-overexpressing tissue was a zone of increased SOP density , with fairly regular spacing . Together , these data demonstrate the danger posed by persistent non-SOP expression of neur , resulting either in failure to establish the normal SOP fate or inappropriate specification of ectopic SOPs . The above data establish both the existence of neur expression in multiple PNC cells prior to SOP specification and the danger posed by persistence of this expression in non-SOPs . Once SOP specification and effective inhibitory Notch signaling are established , the non-SOPs of the PNC prevent the accumulation of new neur transcript by deploying the E ( spl ) -C bHLH-Rs . But what about the Neur protein that is already present in non-SOPs due to the earlier neur expression ? We hypothesized that the activity of this “ectopic” Neur protein is inhibited in non-SOPs by the Notch-dependent expression of the Bearded family proteins ( BFMs ) , which bind directly to Neur and competitively block its interaction with the intracellular domains of Notch ligands , thus preventing any reciprocal signaling back to the SOP [15] . Consistent with this model , co-expression of neur and the BFM E ( spl ) m4 using mα-GAL4 significantly decreases the lateral inhibition disruptions caused by neur expression alone ( Fig 5A ) . Conversely , we also assayed the effect of removing endogenous expression of two BFMs [E ( spl ) mα and E ( spl ) m4] on the neur misexpression phenotype . Adult flies homozygous for a double deletion of both E ( spl ) mα and E ( spl ) m4 display a mild extra-bristle phenotype ( Fig 5A ) . When neur is now misexpressed in this background using just a single copy of driver and responder , the number of extra bristles is greatly increased , far beyond that seen in a wild-type BFM background ( Fig 5A ) . Thus , endogenous BFM expression in non-SOPs does strongly inhibit Neur activity in these cells . Of course , the severity of the extra-bristle phenotype in this experiment is artificially enhanced due to the high levels of Neur produced in response to the GAL4 driver . Therefore , we examined the consequence of loss of the two BFMs in a background homozygous for the neurRC-4D , 1BRm rescue construct , which causes only a modest de-repression of neur in non-SOPs ( Fig 3 ) . Because the phenotypic effects vary substantially among different bristle positions , overall macrochaete counts on the head and thorax ( Table 1 ) can be less informative than more focused assays . If we consider those bristle positions where we routinely observe ectopic reporter transgene activity or Neur::GFP expression , we see a statistically significant increase in bristle numbers in E ( spl ) mα E ( spl ) m4 homozygous deletion flies with the addition of the Rm mutant neur rescue construct ( Table 2 ) . To this point we have established that persistent neur expression in non-SOPs poses a threat to lateral inhibition , and have illuminated the mechanisms these cells use to antagonize the transcriptional activation ( via the bHLH-Rs ) and the function ( via the BFMs ) of neur . Since both of these non-SOP-specific inhibitors are direct targets of Notch signaling from the SOP , in which Neur is a critical cell-autonomous participant , it follows that blocking Neur function specifically in the SOP should lead to ectopic neur transcript accumulation in the other cells of the PNC . We therefore inhibited Neur function in the SOP by ectopically co-expressing two BFMs , Tom and E ( spl ) m4 , in this cell using a neur-GAL4 driver . As predicted , we observed in this genotype multiple positions in the wing imaginal disc displaying both ectopic neur transcript ( Fig 6A and 6C ) and ectopic expression of a neur4D reporter transgene ( Fig 6B and 6B’ and 6D–6D” ) .
The logic of neur activation in SOPs appears remarkably complex . The presence of conserved proneural protein and bHLH-R binding motifs in neur4D and neur1B suggested that a simple “P+R” cis-regulatory code might underlie the operation of these enhancers—direct transcriptional activation by proneural proteins in the PNC , with non-SOP expression directly repressed by bHLH-Rs [13] . Mutating these motifs in the context of reporter transgenes , however , has revealed a more intricate regulatory scheme . We observed that upon bHLH-R binding site mutagenesis in the neur enhancers , only a subset of non-SOP cells displayed ectopic expression . This contrasts with the behavior of previously studied SOP enhancers in the phyllopod ( phyl ) and nervy ( nvy ) genes , which exhibit strong and extensive de-repression in PNCs upon mutation of their bHLH-R motifs [13] . A number of circumstances may contribute to the weak de-repression of the neur enhancers . First , they may be subject to direct repression by additional factors beyond the bHLH-Rs . A strong precedent for this possibility is provided by the downstream SOP enhancer of the senseless ( sens ) gene , which is repressed in non-SOPs by both bHLH-Rs and the Sens protein itself [13] . Only when both of these inputs are eliminated does the enhancer exhibit substantial ectopic activity . Second , unlike the phyl SOP enhancer , the neur enhancers may be relatively unresponsive to the lower levels of proneural protein activity in non-SOP cells . Our finding that mutation of the proneural binding motifs in either neur enhancer fails to completely eliminate its SOP activity indicates that they both receive additional positive inputs , and these may be present at only marginal levels in non-SOPs . Since removal of Ac/Sc proneural activity in trans abolishes the activity of both enhancers [13] , these additional factors most likely lie downstream of the proneurals in a coherent feed-forward regulatory structure [42] . Our results indicate that neur1B and neur4D are differentially dependent on the proneural component of this feed-forward mechanism . Mutating its proneural motifs has a stronger effect on neur1B’s activity , while neur4D likely relies more upon the proneural-dependent activation of several additional regulators . We suggest that SOP-specific enhancers that are targets of the proneurals typically lie at various positions along this spectrum , with their different requirements for direct proneural regulation possibly related to the timing of their activity or to the specific function of the associated gene during SOP specification and differentiation . Other contrasts between neur1B and neur4D are also evident . There are marked differences in overall motif composition and organization; for example , neur4D contains two SMCα motifs , previously associated with activation in SOPs [29] , while neur1B lacks them . In addition , the SOP-specific activity generated from the neur1B region of the locus seems to be distributed over a larger area , since a partially overlapping region , neur1C , also exhibits some weak SOP activity , and a larger fragment ( NRS1 ) containing both neur1B and neur1C drives stronger and slightly expanded expression , including the wing margin [13] . By contrast , we have not detected enhancer activity in the intronic area adjacent to neur4D . Finally , it is noteworthy that neur1B and neur4D display a very different reliance on PS versus PA proneural binding motifs . Overall , the many structural and functional differences between neur1B and neur4D may reflect a role for the two enhancers in ensuring the robustness of neur’s expression in SOPs [43 , 44] . While these modules exhibit a largely overlapping SOP functionality [13] , it may be advantageous for them to rely differentially on various positive and negative inputs in order to better withstand a range of genetic and environmental perturbations . The evolutionary appearance of distinct Atonal and Achaete/Scute subfamilies of proneural proteins likely predates the cnidarian/bilaterian divergence , perhaps 550–600 Mya [33 , 45] . It is perhaps not surprising , therefore , that Ato and Ac/Sc factors have been found to have distinct roles in cell fate specification during development . In Drosophila , for example , the external sensory organs of the peripheral nervous system are dependent on ac/sc gene function , while chordotonal organs and the R8 photoreceptors of the eye rely on ato [46] . Despite this , it is certainly reasonable to imagine—given their shared role in the overall process of neurogenesis—that the target gene repertoires of the Ato and Ac/Sc factors might be substantially overlapping , and indeed many common targets have been identified . In some instances , the two factor types have been found to regulate a common target largely via distinct binding sites , as exemplified by the Brd gene [47 , 48] . By contrast , we have shown here that neur utilizes proneural binding motifs of the CAGATG class to mediate activation by both Ac/Sc and Ato . The logic underlying the use of common versus distinct proneural sites in the same target is not entirely clear , but may reflect constraints imposed by selective interactions with regulatory cofactors [46] . Previous studies of neur expression and function in PNCs during lateral inhibition have relied on reporter genes [13 , 37 , 38] or mutational analysis [13 , 20 , 23] . Our direct analysis of neur transcription and protein accumulation in macrochaete PNCs has demonstrated explicitly that , prior to SOP specification , a distinctive subset of PNC cells activates neur expression . Lack of neur function during Notch-mediated lateral inhibition results in a comparatively modest mutant phenotype by comparison to the effects of losing the activity of other “neurogenic” genes such as Notch itself [20] . Specifically , only a relatively small subset of cells in the PNC commit inappropriately to the SOP fate [20 , 23] . We suggest that these ectopic SOPs correspond to the “pre-SOP” subset identified here by neur expression analysis , and thus that the “pre-SOPs” overlap strongly , or even coincide , with the “neur group” described by Troost et al . [23] . Given the essential role—both direct and indirect—played by proneural gene activity in activating neur expression [13 , 24] , it is likely that this is the principal determinant of which PNC cells are members of the “pre-SOP” group . Thus , the “pre-SOPs” would correspond to those cells with the highest levels of net proneural activity—the cells with the highest levels of proneural protein accumulation and the lowest levels of expression of the inhibitory Extramacrochaetae ( Emc ) protein [49] . The need to specify only one SOP cell within each PNC presents clear regulatory challenges . The very fact that membership in the PNC is defined by expression of proneural factors imposes the strict requirement that the net levels of proneural activity in the non-SOP cells be kept below a threshold that would permit their inappropriate commitment to the SOP fate . Likewise , it is critical that the non-SOPs—either individually or collectively—do not become sufficiently strong Notch signalers as to inhibit the proper specification of the single SOP . Since Neur is a principal determinant of this signaling capacity , it is vital that only the SOP acquires sufficient Neur activity to become a fully effective signal source . Yet neur transcription is both directly and indirectly activated by proneural factors , and while this gives the SOP a clear advantage ( due to its elevated level of proneural protein ) , it also creates the serious risk of one or more non-SOPs developing inappropriately high levels of Neur function . We have shown here that the lateral inhibition network utilizes two distinct mechanisms to counter this threat . The first operates at the level of controlling neur transcription in non-SOPs ( Fig 7 ) . Notch signaling from the SOP activates the expression of multiple Hes-class bHLH repressor proteins specifically in the non-SOPs [35] . These factors are thus ideally suited to the task of inhibiting the expression of SOP genes only in non-SOPs [13] . Direct transcriptional repression of neur by the Hes proteins works , then , to counteract the proneural-dependent activation of the gene in non-SOPs . However , the threat of inappropriate Neur activity in non-SOPs has a second source ( Fig 7 ) . We have demonstrated that , prior to the establishment of effective Notch signaling activity by the presumptive SOP ( and therefore prior to the onset of Hes repressor function in non-SOPs ) , a subset of non-SOP cells ( the “pre-SOPs” ) actively transcribe neur . The resulting neur mRNAs could then encode sufficient Neur protein to confer significant Notch signaling capacity on one or more pre-SOPs , potentially resulting in inhibition of the SOP’s fate commitment . This possibility is countered by a second class of Notch pathway targets , the Brd gene family , transcription of which is likewise activated selectively in non-SOPs [35] . As potent direct inhibitors of Neur’s function in activating Notch ligands [14 , 15 , 50] , the Brd proteins offer an effective post-transcriptional solution to the problem of Neur protein accumulation in non-SOPs . Due to the essential role it plays in establishing the SOP’s Notch signaling capacity , Neur is indirectly responsible for stimulating the expression in non-SOPs of both the Hes repressors and the Brd proteins , both of which act to antagonize Neur activity in these cells ( Fig 7 ) . It follows , then , that the neur gene engages in two distinct modes of cell-non-autonomous negative autoregulation during lateral inhibition , which serve to insure the robustness of the SOP specification process .
The E ( spl ) mα-Gal4 driver was described previously [35] . UAS-neur and UAS-Tom were constructed by Eric Lai , and UAS-FLAGm4 by Joseph Fontana [15] . The E ( spl ) mα E ( spl ) m4 double-deletion line was a generous gift from Joseph Fontana , constructed via two independent homologous recombination events using the methods described [51] . Stocks for generating neur MARCM clones ( y w hs-FLP122 tub-Gal4 UAS-GFP-6xnls; FRT82B tub-Gal80/TM6B and w; FRT82B neur1 cu/TM6B ) were generously provided by Christos Delidakis [20] . UAS-neur was crossed in to create the stock w; UAS-neur; FRT82B neur1 cu/TM6B . Mosaic analyses using the FLP/FRT and MARCM systems have been described [20 , 52–55] . Reporter constructs for neur4DWT ( primers 5'-CCAAGACCCAAATTTAGTTGGTATTCAAGC-3' and 5'-AATAGGCCCCAATCCAGTACACGTATGTGC-3' ) and mutants ( PS and PA , RCANNTG>RAANNGG; Sens , AAATCTGT>AGGTCTGT; bHLH-R , CACGYG>CCCTYT; SMC , AGGGGTTG>AAAAAAAA; for “mystery blocks , ” all nucleotides in S2 Fig converted to A ) were cloned into pH-Stinger [56] or pH-RedStinger [57] . Mutations were generated by overlap extension PCR [58] . At least three independent transformant lines were analyzed before a representative line was selected for all further analysis . Constructs were injected using standard transformation techniques [59] , with w1118 as the recipient strain . Wild-type ( primers NRS1B-u 5'-TCCCAGTTTTGAAACCATTAGCTTACACAG-3' and NRS1B-d 5'-AAAGACAATTGTGAGGCCAGAGGGTAATGC-3' ) and mutant versions of neur1B were generated and cloned into pH-Stinger-attB and injected using the ΦC31 integrase system [60] into the docking site VK00037 [36] . The neur4D and neur1B variants in S5 Fig , as well as the constructs from the promoter-proximal regions shown in S1 Fig , were cloned into pH-Stinger-attB and integrated into the ΦC31 docking site attP2 ( 1B-C: NRS1B-u and NRS1C-d; 1C: NRS1C-u 5'-GCAGACAGCTGCTTCCATTTGCATTTGTCG-3' and NRS1C-d 5'-ATTCCCTTTTGTGTCCGCAGGATTAGTTCG-3'; 1BC: NRS1BC1 . 1-u 5'-TCGATATCCACTGTACCCATCATGATCACC-3' and NRS1BC1 . 1-d 5'-GCAAAGGTAGTAACTCGATCGTAATGGAGG-3'; 1BBC: NRS1B-u and NRS1BC1 . 1-d ) . neurRC-WT-P[acman] constructs were generated by BACR09F04-mediated gap repair of attB-P[acman]-AmpR via recombineering , as described [36] . The region cloned extends to the Eag I sites on either side of the neur locus ( from sequence CGGCCGCCTCCAGGATAAGATGCT to sequence GATATACCCGCTGTGAATCGGCCG , a 21-kb region ) . These constructs were subsequently injected into the docking sites attP40 and attP16 [61] by Genetic Services , Inc . , using the ΦC31 integrase system [60] . Mutant and tagged variants of this starting construct were generated by recombineering using galK-mediated selection [62] , and injected into the attP40 docking site . neurRC-WT-GFP was integrated into the attP40 , attP2 , attP16 , and VK00037 docking sites [36 , 61 , 63]; neurRC-4D , 1BRm-GFP was integrated into VK00037 for comparison with the WT-GFP at the same site . Single-probe in situ hybridizations were performed as previously described [10 , 24 , 64 , 65] . Quantification of in situ signal area and darkness for the neurRC-4D , 1BRm experiment was performed using ImageJ software , taking the average of 9 discs for the WT construct and 21 discs for the Rm construct . Statistical significance was assayed by ANOVA . Multiplex fluorescent in situ hybridizations in third-instar wing imaginal discs were performed basically as described [39]; anti-hapten antibodies ( sheep anti-DIG , mouse anti-biotin , and chicken anti-DNP ) were used at a 1:5000 dilution in 1X PBS + 0 . 1% Triton X-100 ( PBT ) , without using a block solution ( we observed too much background in disc tissue when using the Roche Block mentioned in Kosman et al . ) . Probes were constructed by cloning an intronic DNA fragment into pGEM-T , linearizing , and transcribing RNA using the T7 RNA polymerase following the Kosman protocol . The following probes were used: DNP-sca , DIG-neur , BIO-neur , BIO-CG32150 . Images were captured as described below , adjusting the gain to maximally reveal any coincidence between neur probes . With the exception of GFP antibody staining , immunohistochemistry was performed essentially as described previously [64] . Discs from neurRC-WT-GFP discs also included a blocking step after fixation in 0 . 3% milk in PBT . Blocking was done overnight at 4°C , with primary antibodies added the next morning , also in the milk blocking solution . Secondary antibodies for this stain were added in PBT only . The following antibodies were used: guinea pig anti-Sens ( generously provided by Hugo Bellen ) , 1:2000; mouse anti-Cut ( 2B10 ) [Developmental Studies Hybridoma Bank ( DSHB ) , University of Iowa] , 1:100; rabbit anti-GFP ( Invitrogen ) , 1:500 . All secondary antibodies used were AlexaFluor varieties from Invitrogen and included anti-rabbit-Alexa488 conjugate , anti-guinea pig-Alexa555 conjugate , anti-mouse-Alexa555 conjugate , and anti-mouse-Alexa647 conjugate . Secondaries were always used in staining at a 1:1000 dilution in PBT . For the fluorescent in situ hybridizations , the secondaries were all raised in donkey . Multiple independent transformant lines were collected for each pH-Stinger GFP reporter construct . Imaginal discs from at least 10 larvae were collected for each line and analyzed for variation across the line . To record images , imaginal discs from at least 10 larvae or pupae carrying wild-type and mutant constructs were collected , dissected and fixed , and imaged in parallel under identical confocal settings . Representative images are displayed in the figure panels . For analysis of ectopic GFP reporter expression due to transcription factor binding motif mutations ( Fig 2 ) , 30 third-instar larvae were dissected and all discs with discernible DC and SC positions were analyzed , noting the presence of any cell expressing nuclear GFP but not Sens at these positions . For quantification of bristle phenotypes ( Fig 5; both Tables ) , all macrochaete positions on the dorsal head and thorax were analyzed , and each position scored for either missing or extra bristles , over a total of 25 males and 25 females unless otherwise noted . Statistical significance was determined by pairwise ANOVA . Confocal microscopy procedures have been described previously [64] . Images of fluorescent in situ hybridizations were collected as series of 1-micron sections; antibody stains were collected at low magnification as 2-micron sections , with high-magnification images as 1-micron sections . For the collection of z-sections to generate the cross-sectional view shown in Fig 4E , we shortened the distance to 0 . 75-micron sections . Images were collected using Leica confocal software , cropped with Adobe Photoshop , and combined into figures using Adobe Illustrator . Gene structure and sequence alignment diagrams were constructed using the latest version of the GenePalette software tool ( http://www . genepalette . org ) [66] and were edited in Adobe Illustrator . Additional oligonucleotide primer sequences are available upon request .
|
Much of the process of animal development is concerned with giving cells specific instructions as to what type of cell they are to become—their “fate” . Often , it is even necessary to assign very different fates to cells that are adjacent to each other in the tissue . In such cases , cell-to-cell signaling is frequently utilized as the means of distinguishing the cells’ fates . For example , one cell might send a signal to its neighbors that inhibits them from adopting the same fate as itself . Here , it is obviously vital that there is an asymmetry between the “sending” and “receiving” cells in the ability to transmit such a signal . In the fruit fly Drosophila , the gene neuralized encodes a protein that plays a critical role in establishing the capacity to send such an inhibitory signal . The work we describe here reveals specifically how the receiving cells are prevented from acquiring the ability to send the signal . Remarkably , the Neuralized protein itself is deeply involved in this process . Neuralized function in the sending cell generates two distinct mechanisms that inhibit its own activity in the receiving cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"invertebrates",
"animals",
"notch",
"signaling",
"dna",
"transcription",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"sequence",
"motif",
"analysis",
"morphogenesis",
"drosophila",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"sequence",
"alignment",
"bioinformatics",
"gene",
"expression",
"life",
"cycles",
"imaginal",
"discs",
"insects",
"arthropoda",
"signal",
"transduction",
"eukaryota",
"cell",
"biology",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"larvae",
"organisms"
] |
2018
|
Lateral inhibition: Two modes of non-autonomous negative autoregulation by neuralized
|
Most of the Leishmania genome is reported to be constitutively expressed during the life cycle of the parasite , with a few regulated genes . Inter-species comparative transcriptomics evidenced a low number of species-specific differences related to differentially distributed genes or the differential regulation of conserved genes . It is of uppermost importance to ensure that the observed differences are indeed species-specific and not simply specific of the strains selected for representing the species . The relevance of this concern is illustrated by current study . We selected 5 clinical isolates of L . braziliensis characterized by their diversity of clinical and in vitro phenotypes . Real-time quantitative PCR was performed on promastigote and amastigote life stages to assess gene expression profiles at seven time points covering the whole life cycle . We tested 12 genes encoding proteins with roles in transport , thiol-based redox metabolism , cellular reduction , RNA poly ( A ) -tail metabolism , cytoskeleton function and ribosomal function . The general trend of expression profiles showed that regulation of gene expression essentially occurs around the stationary phase of promastigotes . However , the genes involved in this phenomenon appeared to vary significantly among the isolates considered . Our results clearly illustrate the unique character of each isolate in terms of gene expression dynamics . Results obtained on an individual strain are not necessarily representative of a given species . Therefore , extreme care should be taken when comparing the profiles of different species and extrapolating functional differences between them .
Leishmania are digenic Protozoan parasites endemic worldwide and causing a spectrum of diseases in humans collectively referred to as leishmaniasis . As part of their life cycle , Leishmania alternate between the alimentary tract of the sandfly vector ( where they grow as extracellular flagellated promastigotes and differentiate into infective non-dividing metacyclic forms ) and the phagolysosome of the vertebrate host macrophages ( where parasites differentiate into aflagellated replicative amastigotes ) . Important morphological and biochemical changes underlie the differentiations involved in the life cycle and are most likely the result of regulated changes in gene expression in response to environmental signals ( e . g . temperature change and pH shift ) [1] , [2] , [3] . A recent study assessing gene expression profiles throughout the life cycle of a single strain of L . infantum showed that most of the Leishmania genome is constitutively expressed and that consequently , regulated genes are a minority [4] . Interestingly , most significant events of regulation were shown to occur during promastigote development , with a dramatic down-regulation during the transition from promastigotes to amastigotes , hereby supporting the hypothesis of pre-adaptation of the parasite to intracellular survival in the macrophage [4] . In parallel to these longitudinal studies of gene expression throughout the life cycle of a given strain , another critical dimension needs to be explored , i . e . the gene expression diversity within natural populations of Leishmania . Indeed , these parasites successfully colonized a large range of hosts in various ecological niches , hereby determining different transmission types in which humans play a variable role ( from dead-end host to reservoir ) . Not surprisingly , the parasites are characterized by high phenotypic diversity: infectivity , tissue tropism , clinical pattern or drug susceptibility , among others . This polymorphism is reflected at the taxonomic level , with more than 20 described species . A first comparative transcriptomic study throughout the life cycle of 3 strains belonging to the species L . major , L . infantum and L . braziliensis was recently reported: this provided in the 3 species a similar picture of a mostly conservative gene expression , and a minority of species-specific differences related to differentially distributed genes or the differential regulation of conserved genes , either of which are subject to translational and/or post-translational controls [5] . This type of research being undertaken among other reasons for a better understanding of the differences in virulence and pathogenicity of the respective species , it is of utmost importance to ensure that the observed differences are indeed species-specific or simply specific to the strains selected as representative of the species . In a previous study targeting a limited number of genes , we analyzed the expression profile of 21 L . braziliensis clinical isolates during in vitro promastigote growth [6] . A fraction of genes were up-regulated during differentiation , but the set of regulated genes varied between isolates , providing a picture of intra-species gene expression mosaic . We aimed here to continue our assessment of gene expression diversity within a single species ( L . braziliensis ) by extending it to the whole life cycle of the parasite . This type of study is complicated by the extreme sensitivity of the amastigote stage to disturbance of the intracellular parasite's environment during harvesting , hence extreme care is needed to ‘freeze’ gene expression levels instantly [7] . Our previous work also demonstrated the importance of time-course analyses to better observe the dynamics of gene expression changes during in vitro growth and to allow more reliable comparisons between different strains [6] , [7] . These considerations currently impede high-throughput experiments involving the clinically relevant amastigote stage . We surveyed here 5 clinical isolates of L . braziliensis , characterized by their diversity of clinical and in vitro phenotypes . We applied a standardized and reproducible biological protocol [6] , [7] and used real-time quantitative PCR to assess gene expression profiles at seven time points covering the whole life cycle . We tested 12 genes encoding proteins with roles in transport , thiol-based redox metabolism , cellular reduction , RNA poly ( A ) -tail metabolism and housekeeping functions .
The five parasite isolates used here were obtained from patients with cutaneous leishmaniasis at the Instituto de Medicina Tropical A . von Humboldt , Lima , Peru . Protocol and informed consent were approved by the Research Ethic Committees of Universidad Peruana Cayetano Heredia ( Lima , Peru ) and Institute of Tropical Medicine ( Antwerp , Belgium ) . Written , informed consent was obtained from all participating subjects or their legal guardians . Animal experimentation concerned only the use of peritoneal macrophages obtained from mice . Our animal protocol adhered to the guidelines at the Universidad Peruana Cayetano Heredia , Lima , Peru and was in agreement with the Peruvian and Belgian regulations for the protection and welfare of laboratory animals . Mouse care and experimental procedures were performed under approval of the Ethic Committee of the Universidad Peruana Cayetano Heredia as well as the Animal Ethic Committee of the Institute of Tropical Medicine Antwerp ( PAR-018/2 ) . The five parasite isolates were obtained from Peruvian patients with confirmed cutaneous leishmaniasis and originating from 4 regions of the country ( Table 1 ) . The selected isolates were previously characterized as L . braziliensis by PCR-RFLP assays targeting a range of markers [8] , but we retyped them here to obtain a more precise perception of their genetic diversity . Therefore , we sequenced part of the coding region of the hsp70 genes , a locus recently shown to be phylogenetically very informative [9] and more and more used for species identification of Neotropical Leishmania species [10]–[12] . Sequencing concerned the isolates of this study [GenBank accession numbers: FR715986 ( isolate PER002 ) , FR715987 ( isolate PER006 ) , FR715988 ( isolate PER182 ) , FR715989 ( isolate PER104 ) , FR715990 ( isolate PER163 ) ] and the additional Peruvian L . braziliensis reference strain LEM2222 ( http://www . parasitologie . univ-montp1 . fr/cnrl . htm; GenBank accession number: FR715991 ) . The sequences were aligned with reference strains recently published by our group [9] and available GenBank entries , and compared by Neighbor-Joining analysis of p-distances , using the software package MEGA4 ( http://www . megasoftware . net/ ) . Data on the in vitro SbV and SbIII susceptibility of the studied isolates , as tested by the intracellular amastigote-macrophage model [8] , and of the clinical treatment outcome of respective patients , were available ( see Table 1 for summary of isolates' features ) . Isolates were used here for gene expression analysis within a maximum of 25 in vitro passages post-isolation from patients . The in vitro conditions for promastigote generation have been described previously [6] . Briefly , growth curves were initiated by inoculating 3×106 parasites/mL in 5 mL medium 199 ( M199; Sigma ) containing 20% heat-inactivated fetal bovine serum ( FBS; Lonza Bioscience ) , 25 mM Hepes ( pH 7 . 4 ) , 100 units/mL penicillin and 100 µg/mL streptomycin ( Lonza ) . Two independently grown cultures and corresponding harvests at 24 h ( early-log phase ) , 72 h ( late-log phase ) , 120 h ( early-stationary phase ) and 168 h ( late-stationary phase ) time points were performed in parallel for each isolate ( biological replicates ) . Intracellular amastigotes were obtained after infection with axenic amastigote-like forms , as this was reported to enhance infectivity in species of L . ( Viannia ) subgenus [13] . Briefly , stationary-phase promastigotes were incubated in M199-20% FBS , at pH 5 . 5 and 34°C , during 96 h [14] . Following this incubation , parasites were washed , resuspended in M199-pH7 . 4–10% FBS ( pre-warmed at 32°C ) , and then used to infect murine peritoneal macrophages ( collected from BALB/c mice ) at a parasite to macrophage ratio of 10∶1 . Infected cultures were incubated for 4 h at 32°C in a humidified 5% CO2/95% air environment , then washed with pre-warmed medium to remove free extracellular parasites , and further incubated for 3 days . Intracellular amastigotes were co-harvested with macrophages at 24 h , 48 h and 72 h post-infection using a protocol that has shown to ‘freeze’ L . donovani gene expression levels instantly [7] . Infection rates were monitored using control plates , as described elsewhere [15]; up to 200 cells were counted in order to determine the percentage of infected macrophages and the average number of amastigotes by infected macrophages . The whole experiment was done in 2 biological replicates . RNA sampling protocols , RNA isolation and quality control were performed as described elsewhere [16] . First strand cDNA was synthesized from 150 ng total RNA of promastigote samples and from 1 µg total RNA of mixed amastigote-macrophage samples , using a 18mer oligo ( dT ) and Transcriptor Reverse Transcriptase , according to the manufacturer's instructions ( Roche Applied Science ) . The resulting cDNA was diluted 10-fold with DEPC-treated water ( Ambion ) for further use . From the cDNA diluted samples , 2 µL was used as template in 25 µL SYBR Green-based quantitative PCR ( qPCR ) reactions on the iCycler ( Bio-Rad ) , as previously described [16] , with the only modification that amplification was done for 34 cycles [6] . We analyzed 12 Leishmania-specific genes , with predicted function in transport ( LbAQP1 , MRPA ) , thiol-based redox metabolism ( GSH1 , GSH2 , ODC , TRYR ) , cellular reduction ( ACR2 , TDR1 ) , RNA poly ( A ) -tail metabolism ( PABP , PAP14 ) , cytoskeleton function ( Actin ) and ribosomal function ( S8 ) . Primer sequences , parameters and reproducibility of the respective qPCR assays have been described previously [6] . Analysis was performed on duplicate biological samples that were each assayed in triplicate . The arithmetic average threshold cycle ( Ct ) was used for data analysis . The Ct values of each qPCR run were imported as Excel files into qBasePlus 1 . 3 ( Biogazelle NV , Zulte , Belgium ) , a software for real-time PCR data analysis based on the geNorm method [17] and qBase technology [18] . Three genes ( ACR2 , GSH2 , PAP14 ) showed the most stable expression through parasite life stages in our sample panel ( geNorm stability mean M-value and mean coefficient of variation lower than 0 . 45 and 20% , respectively ) and data were normalized to their geometric mean . The use of multiple control genes allows more accurate and reliable normalization of gene expression data [17] . The linear component of the variability of the expression level of each gene during in vitro growth was modelled independently for each isolate using linear regression . The main predictor was the development time of the parasite during in vitro promastigote or intracellular amastigote growth/differentiation . The time point 24 h was considered as the baseline time in the regression models , in order to prevent negative values for gene expression levels , a biological impossibility . The constant terms of the regression models were used as measure of the baseline expression of a given gene , with the criterion of significance that the constant terms differed by at least 2-fold from the lowest constant term in the series . In addition , the fold change ( FC ) of gene expression of ( i ) stationary versus logarithmic phase promastigotes ( time points 120 h and 168 h versus 24 h and 72 h of the growth curves , respectively ) ; ( ii ) late-differentiating phase amastigotes versus early-differentiating phase amastigotes ( time points 72 h versus 24 h and 48 h post-infection macrophages , respectively ) ; and ( iii ) stationary phase promastigotes ( time points 120 h and 168 h of the growth curves ) versus early-differentiating phase amastigotes ( time points 24 h and 48 h post-infection macrophages ) was determined for each parasite isolate . FC were considered significant if they satisfied a 2-fold cutoff and P<0 . 01 , as recommended elsewhere [19] . All analyses were performed using the statistical software Stata 10 ( StataCorp ) .
Out of the 5 isolates , 4 ( PER002 , -006 , -104 and -182 ) clustered in a well-supported group ( bootstrap value of 92% ) constituted by several reference strains of the L . braziliensis complex ( Fig . 1 ) . The fifth isolate , PER163 , has 100% sequence identity with the MLEE typed L . braziliensis reference strain LEM2222 , and they cluster with a bootstrap support of 78% . Even though PER163 has more sequence similarity ( 99 . 64% ) with the main L . braziliensis group than with L . naiffi ( 99 . 49% ) , it clusters with the latter without bootstrap support . A series of parameters were measured during the growth of promastigotes and the macrophage infection by amastigotes for phenotypic comparison of the 5 isolates . At the promastigote level , our observations led us to the empirical description of 3 main phenotypes ( Table 1 ) : ( i ) slow growth , but high density at stationary phase in PER163 and PER006 , ( ii ) rapid growth and high density at stationary phase in PER104 and PER182 , and ( iii ) slow growth and lower density at stationary phase in PER002 . At the amastigote level , 2 main phenotypes were observed ( Table 1 ) : ( i ) high percentage of infected macrophages and high number of amastigotes per macrophage in PER163 and PER104 , and ( ii ) low values for both parameters in PER002 , PER006 and PER182 . We obtained highly reproducible gene expression measurements between biological replicates , in both the promastigote and intracellular amastigote developmental time series ( Fig . 2 , 3 , 4 ) , thereby confirming that parasites were manipulated in a standardized way . Visual inspection of the graphs revealed notable differences between life stages in general and between genes according individuals . For instance , LbAQP1 showed in PER163 and PER006 a clear up-regulation during promastigote development , followed by a 5-fold down-regulation during the transition to amastigotes ( Fig . 2 ) . In sharp contrast , TRYR showed a similar expression during the 4 time points of promastigote growth curve in PER163 , while in PER006 a clear up-regulation was observed ( Fig . 3 ) . A last example concerns the expression of Actin shown to be down-regulated from promastigote to amastigote in PER163 and PER006 , whereas it appeared to be clearly constitutive and at relative low levels in PER182 through the 7 time points analyzed here ( Fig . 4 ) . In order to summarize and better visualize the gene expression profiles during the life cycle , we examined 5 parameters in the 5 isolates: the baseline expression levels ( given by the constant term of the linear regression models ) in ( 1 ) promastigotes and ( 2 ) amastigotes , and the fold change in mRNA abundance ( 3 ) from logarithmic to stationary phase promastigotes ( further called FC-PRO ) , ( 4 ) during the transition from stationary phase promastigotes to early-differentiating phase amastigotes ( further called FC-PRO-AMA ) , and ( 5 ) from early- to late-differentiating phase intracellular amastigotes ( further called FC-AMA ) . Results are schematized in Figures 5 and 6 . First , when analyzing all parameters together , we identified 3 genes that were constitutively expressed during the whole life cycle and in the 5 isolates: ACR2 , GSH2 and PAP14 . Secondly , analysis of the baseline expression in promastigotes ( Fig . 5A ) discriminated 2 groups of parasites: ( i ) PER002 and PER182 , which showed the lowest baseline levels for the 12 genes studied here and ( ii ) PER163 , PER006 and PER104 which showed significantly higher baseline levels for 2 to 5 genes ( out of the following set: Actin , LbAQP1 , MRPA , PABP , S8 and TRYR ) . Thirdly , analysis of the baseline expression in amastigotes ( Fig . 5B ) also discriminated 2 groups of parasites: ( i ) PER163 and PER006 showed the lowest baseline for the 12 genes at the amastigote stage , which is in sharp contrast to promastigote data , and ( ii ) PER002 , PER104 and PER182 which showed significantly higher baseline expression of 4 to 6 genes ( out of the following set: Actin , LbAQP1 , GSH1 , ODC , PABP , TDR1 and TRYR ) . Fourthly , up-regulation of gene expression during promastigote differentiation ( FC-PRO ≥2 , Fig . 6A ) was observed for up to 3 genes out of the following set: LbAQP1 , GSH1 and TRYR . As did other parameters , the FC-PRO also discriminated PER163 and PER006 from other isolates , as the former showed the lowest number of up-regulated genes . Fifthly , down-regulation during the promastigote-amastigote transition ( FC-PRO-AMA ≥2 , Fig . 6B ) was observed in all isolates , but to a very different extent: ( i ) involving 7 and 8 genes in PER163 and PER006 respectively , ( ii ) 3 genes in PER002 and ( iii ) one gene only in PER104 and PER182 . Sixthly , and in stark contrast to previous parameters , we did not observe up-regulation of any gene in any isolate during amastigote development ( FC-AMA <2 , Fig . 6C ) .
The accession numbers for the genes analyzed in this study are as annotated at L . braziliensis GeneDB database ( http://www . genedb . org/Homepage/Lbraziliensis ) . LbAQP1 ( GeneDB ID LbrM . 31 . 0020 ) encodes an aquaglyceroporin . MRPA ( LbrM . 23 . 0280 ) codes for an ABC-thiol transporter . GSH1 ( LbrM . 18 . 1700 ) encodes a putative gamma-glutamylcysteine synthetase ( γ-GCS ) . GSH2 ( LbrM . 14 . 0880 ) codes for a putative glutathione synthetase ( GS ) . ODC ( LbrM . 12 . 0300 ) codes for a putative ornithine decarboxylase . TRYR ( LbrM . 05 . 0350 ) encodes trypanothione reductase . ACR2 ( LbrM . 32 . 2980 ) encodes a putative SbV/AsV reductase according to the orthologous ACR2 sequence in L . major ( GeneDB ID LmjF . 32 . 2740; GenBank accession number AY567836 . 1 ) . TDR1 ( LbrM . 31 . 0550 ) encodes a thiol-dependent reductase 1 . PABP ( LbrM . 30 . 2560 ) encodes a putative RNA-binding protein . PAP14 ( LbrM . 14 . 1350 ) codes for a putative poly ( A ) polymerase . Actin ( LbrM . 04 . 1250 ) codes for the actin protein . S8 ( LbrM . 24 . 2160 ) encodes a putative 40S ribosomal protein S8 .
|
Leishmania is a group of parasites ( Protozoa , Trypanosomatidae ) responsible for a wide spectrum of clinical forms . Among the factors explaining this phenotypic polymorphism , parasite features are important contributors . One approach to identify them consists in characterizing the gene expression profiles throughout the life cycle . In a recent study , the transcriptome of 3 Leishmania species was compared and this revealed species-specific differences , albeit in a low number . A key issue , however , is to ensure that the observed differences are indeed species-specific and not specific of the strains selected for representing the species . In order to illustrate the relevance of this concern , we analyzed here the gene expression profiles of 5 clinical isolates of L . braziliensis at seven time points of the life cycle . Our results clearly illustrate the unique character of each isolate in terms of gene expression dynamics: one Leishmania strain is not necessarily representative of a given species .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"biology",
"microbiology",
"parasitology"
] |
2011
|
Comparative Gene Expression Analysis throughout the Life Cycle of
Leishmania braziliensis: Diversity of Expression Profiles
among Clinical Isolates
|
Sickle cell disease ( SCD ) is a hematological disorder leading to blood vessel occlusion accompanied by painful episodes and even death . Red blood cells ( RBCs ) of SCD patients have diverse shapes that reveal important biomechanical and bio-rheological characteristics , e . g . their density , fragility , adhesive properties , etc . Hence , having an objective and effective way of RBC shape quantification and classification will lead to better insights and eventual better prognosis of the disease . To this end , we have developed an automated , high-throughput , ex-vivo RBC shape classification framework that consists of three stages . First , we present an automatic hierarchical RBC extraction method to detect the RBC region ( ROI ) from the background , and then separate touching RBCs in the ROI images by applying an improved random walk method based on automatic seed generation . Second , we apply a mask-based RBC patch-size normalization method to normalize the variant size of segmented single RBC patches into uniform size . Third , we employ deep convolutional neural networks ( CNNs ) to realize RBC classification; the alternating convolution and pooling operations can deal with non-linear and complex patterns . Furthermore , we investigate the specific shape factor quantification for the classified RBC image data in order to develop a general multiscale shape analysis . We perform several experiments on raw microscopy image datasets from 8 SCD patients ( over 7 , 000 single RBC images ) through a 5-fold cross validation method both for oxygenated and deoxygenated RBCs . We demonstrate that the proposed framework can successfully classify sickle shape RBCs in an automated manner with high accuracy , and we also provide the corresponding shape factor analysis , which can be used synergistically with the CNN analysis for more robust predictions . Moreover , the trained deep CNN exhibits good performance even for a deoxygenated dataset and distinguishes the subtle differences in texture alteration inside the oxygenated and deoxygenated RBCs .
Sickle cell disease ( SCD ) , also known as sickle cell anemia , is a type of inherited RBC disorder associated with abnormal hemoglobin S ( HbS ) [1] . When HbS molecules polymerize inside RBCs , due to lack of oxygen , they affect greatly the shape , elasticity , and adhesion properties of RBCs . Moreover , the RBCs become stiff and more fragile , with vastly heterogeneous shapes in the cell population [2] , which makes this problem an ideal candidate for the examination of morphological heterogeneity . Unlike the normal RBCs , which are flexible and move easily even through very small blood vessels , sickle RBCs promote vaso-occlusion phenomena . Hence , SCD patients are afflicted with the risk of life-threatening complications , stroke and organ damage over time , resulting in a reduced life expectancy . According to a recent study [3] , as of 2013 about 3 . 2 million people have SCD while an additional 43 million have sickle-cell trait , resulting in 176 , 000 deaths in 2013 , up from 113 , 000 deaths in 1990 , mostly of African origin . The prime hallmark of SCD is that is surprisingly variable in its clinical severity . Available methods for treating SCD are mainly supportive and mostly aim at symptom control , but lack the active monitoring of the health status as well as the prediction of disease development in different clinical stages [4] . Recent developments in advanced medical imaging technology and computerized image processing methods could provide an effective tool in monitoring the status of SCD patients . Indeed , Darrow et al . [5] recently demonstrated a positive correlation between cell volume and protrusion number using soft X-ray tomography . Van beers et al . [6] have also shown highly specific and sensitive sickle and normal erythrocyte classification based on sickle imaging flow cytometry assay , a methodology that could be useful in assessing drug efficacy in SCD . Therefore , implementing an automated , high-throughput cell classification method could become an enabling technology to improve the future clinical diagnosis , prediction of treatment outcome , and especially therapy planning . However , there are several major technical challenges for automatic cell classification: 1 ) RBCs may touch or overlap each other or appear as clusters in the image , which makes it difficult to detect the hidden edge of cells . 2 ) The RBC region and the background may have low contrast in the intensity . 3 ) The boundaries of RBCs may be blurry due to the influence of imaging procedure . 4 ) Very complex and heterogeneous shapes of RBCs are present in SCD . 5 ) Artifacts may be present , for instance , dirt on the imaging light path , various halos and shading . 6 ) Finally , because RBCs lack a nucleus , methods utilizing the nuclei location as an apparent marker for cell counting and detection are not applicable . The objective of the current work is to develop an automated algorithm for sickle RBC classification test , which may prove a powerful complementary clinical test for a ) assessing patient’s disease severity via longitudinal tracking and patient-specific RBC mapping , and b ) intervention strategies via personalized medicine treatment monitoring . Next , we present a brief overview of the state-of-the-art techniques involved in cell segmentation and classification . Cell detection methods are prevalent , see e . g . [7–10] , and some open source software ( e . g . , CellProfiler [11] , CellTrack [12] , Fiji [13] and CellSegm [14] , etc . ) for 2D and 3D cell detection and counting has emerged recently . However , in SCD we need cell classification , which is quite difficult due to the heterogeneous shapes of RBCs and the existence of touching and overlapped RBCs in the raw microscopy image , and existing software cannot be directly used to obtain RBC boundaries and cannot distinguish among the many different types of RBCs . Presently , there are two kinds of cell classification approaches , i . e . , manual and automatic . In the manual approach one inspects the blood samples using the microscope to count the number of cells and examines the outliers in each frame . This , apparently , is subjective , labor intensive and time consuming for batch data processing . Coulter Counters and Laser Flow Cytometers enable cell sorting automatically by detecting the current and light refraction changes during cell pass through the channel . However , there are some shortcomings , such as the high cost and low processing speed ( 106 cells/hour ) , and in particular , these instruments are not suitable for the classification of heterogeneous cells . Thus , some cellular data analysis tools have been recently developed targeting this problem . For example , ACCENSE [15] adopted two clustering methods ( k-means and DBSCAN ) to facilitate the cellular classification automatically , however , the clustering performance relies on the properly initiation of parameters by hand; moreover , the performance of cell classification degrades for the clusters with different size and different density . More recently , RSIP Vision ( http://www . rsipvision . com ) has developed a commercial software package , allowing the recognition and count of RBCs by using a classifier to classify the hand-crafted morphological features; however , the main drawback of this method is that it requires domain-specific expertise on the feature extraction , f and is also a time-consuming procedure . In addition , the accuracy of the method has not yet been demonstrated for cell classification . Both of the aforementioned methods use machine learning tools but not deep learning algorithms . Likewise , some other similar studies on the HEp-2 cell classification based on the traditional machine learning methods have emerged recently , such as in [16] where multi-variant linear descriptors were adopted to extract the features and applied the SVM method to realize HEp-2 cell classification with an accuracy of 66 . 6% . Other methods include superpixels-based sparse coding method approach [17] , k-nearest clustering method for red blood cell and white blood cell classification [18] , etc . Due to ineffectiveness of the aforementioned methods and given the recent advances of deep learning technique , Gao et al . [19] performed HEp-2 cell classification based on deep CNNs . Also , in order to improve the diversity of single HEp-2 cell data samples , Li et al . [20] carried out classification experiments based on deep CNNs by using four different patients’ datasets under different lighting conditions . However , for the currently available automated machine learning methods , which could be used for cell classification , the following are still drawbacks: 1 ) the classification studies are mostly directly based on already prepared single HEp-2 cellular data , hence , ignoring the initial key procedure of single cell extraction from the raw image data; 2 ) the adopted conventional machine learning methods are time consuming for the hand-crafted feature extraction and need specific human expertise; moreover , they need an accurate cell segmentation; 3 ) the classification accuracy is limited by the selected features and the performance of selected classifier . For our application , since RBCs of SCD exhibit special characteristics in terms of heterogeneous shapes and variant sizes , there is still no efficient tool that can be used to facilitate the automated inspection and recognition of various kinds of RBC patterns which are present in SCD blood . The main focus of our paper is to develop an automated , high-throughput sickle cell classification method based on the deep Convolutional Neural Networks ( dCNNs ) , taking advantage of the hierarchical feature learning goodness of dCNNs . The rest of this paper is organized as follows: In Section 3 we present our methodology , and in Section 4 we present the experimental results , a comparative analysis , and a discussion . Finally , in Section 5 we present the conclusion . S1–S4 Appendix contain some more details of the collection of raw data , the shape factor analysis , the CNN architecture , and deoxygenation method of sickle RBCs .
In the traditional learning-based cell image segmentation or classification method , the two most common techniques to obtain the training patches are the exhaustive pixel-wise sliding window with the same size method [22] and the ground truth bounding box method , e . g . Li et al . [20] . However , the major drawback of the pixel-wise block splitting method for the application of RBC classification is that it generates a large number of unwanted and redundant patches for the background and artifacts ( e . g . , dirt or debris in the light path ) to feed for training and testing of the neural network . This redundancy and artifacts significantly hinder the efficiency of the method to take into account the high resolution of the microscopy data and large background area . The ground truth bounding box method was based on manual labeling of cells present in raw images , a process which is labor intensive and needs specific domain knowledge . In addition , due to the fact that sickle cells are always heterogeneous in shape and at times touch or overlap , it can be difficult to obtain all single RBC patches by using the sliding or bounding box window with a fixed pixel size . Therefore , in our study , a hierarchical RBC patch extraction method was developed to overcome the above problems . The complete flowchart of the proposed method is shown in Fig 2 . Firstly , the raw microscopy images were divided into overlapped patches by using the sliding window technique , with the block size N * N . Then , the entropy containing in each image block was estimated by Eq ( 1 ) below: E = - ∑ i = 1 L P i log P i , ( 1 ) where L is the maximum grayscale level , Pi refers to the probability of occurrence for each intensity level that is encountered in the image block , and it can be derived from the ith histogram count f ( i , j ) divided by the amount of pixels in each subblock image ( the size of the block is N ) , as shown in Eq ( 2 ) below: P i = f ( i , j ) N 2 . ( 2 ) We have employed the information entropy to measure the uncertainty in RBC regions and the background region; the high entropy regions were extracted as the ROI ( region of interest ) , i . e . the RBC regions in the raw microscopy images . The detailed ROI extraction procedure is shown in Fig 3 . First , raw microscopy images ( Fig 3A in high resolution were split into overlapped blocks . Next , the information entropy was calculated for all sub-blocks ( including the edges and noises blocks ) . The blocks with high entropy are shown in white color in Fig 3B , where the entropy threshold ( 5 . 0 ) was obtained from our validation experiments on different datasets . The corresponding ROI mask image was generated by filling the holes and removing the artifacts with area smaller than a common RBC prior area ( 6*10*10 ) for the result of Fig 3B . The result is shown in Fig 3C with each color representing a single ROI region . Fig 3D shows the “cleaned” RBC ROI region result corresponding to the ROI mask image . The entropy estimation method can effectively extract the complete RBC regions from the raw images , especially for those RBC boundaries in a low intensity contrast . Moreover , it can also detect the RBC region correctly from various datasets regradless of their brightness differences . Thus , it can effectively overcome the shortcomings of the previous commonly used methods ( e . g . , Ostu , watershed and Sobel , etc . ) . To obtain the RBC patch images for the deep CNNs , the high-level ROI boundary is detected and by searching the minimum coordination of pixel ( x0 , y0 ) and maximum pixel coordination ( x′ , y′ ) from the boundary pixels , the ROI patches are illustrated as shown in Fig 3E . It should be noted , however , that for the particular situation of overlapped and touching RBCs that may be present in the raw microscopy image , we may obtain some extracted ROI regions containing multiple cells; see the yellow smaller sized box in Fig 3F , where 8 ROI patches contain two or more RBCs , and the pink smaller sized box that includes all segmented single RBC patches . The subimages in the two boxes were obtained by calculating the corresponding bounding boxes of the ROI . Overlapping RBCs were removed from the input of deep CNNs in our work . Therefore , we only focused on the “touching” RBC separation problem by applying the random walk method [23] in conjunction with the distance transform [24] to generate the RBC boundary . This method can obtain the RBC seed points identification automatically . The specific separation procedure is shown in Fig 4 . Because of the RBC heterogenity in size , shape and orientation , the generated single RBC patches from section B were of different sizes ( see Fig 3E ) . In addition , due to varying brightness and intensity contrast conditions during the procedure of raw RBC microscopy data collection , the background of RBC patch images appeared to differ among datasets . Currently , commonly used image scaling methods for the image size normalization are prone to reducing the RBC patch image fidelity ( e . g . , intensity contrast , noise and distortion ) , which will accordingly affect the RBC classification accuracy of the CNN . Therefore , to overcome the above issues , a size-invariant RBC patch normalization method based on statistic intensity linear mapping was employed . The algorithmic workflow is shown in Fig 5 . In our work , we adopted a deep CNN architecture with 10 layers , including 3 convolutional layers ( C1 , C3 and C5 ) , 3 pooling/subsampling layers ( P2 , P4 and P6 ) , dropout layers ( D7 and D9 , where p = 0 . 5 ) and a fully connected layer ( F8 ) . As a result of the computational efficiency , the grayscale RBC image patches were initially resized to 78 * 78 . Next , these were then fed into the neural network . A ReLU non-linear activation function was then applied . Following the F7 layer , a logistic regression method combining the softmax function ( see Eq ( 4 ) ) with a cross-entropy loss function ( see Eq ( 5 ) ) was implemented to obtain the final learning probability and predicted labels . The softmax function can “squash” the obtained score vector Q = {qi|i = 1 , 2 , … , N} to a N-dimension probability vector δ ( qi ) , so as to aid RBC classification efficiency . Q ′ = δ ( q i ) = e q i / ∑ j = 1 N e q j , ( 4 ) D ( Q ′ , Q ) = - ∑ i = 1 N Q i log ( Q i ′ ) , ( 5 ) According to different shape division level for the original RBC patches , two kinds of RBC labeling principles were employed in the experiment: coarse labeling ( output = 5 ) and refined labeling ( output = 8 ) . Thus , the output layer had two different dimensions ( 5 or 8 categories ) . More details about the deep CNN architecture applied in this paper are shown in Fig 7 ( see also S3 Appendix for the specific illustration of the layers of deep CNN ) . As mentioned previously in the text , RBCs from SCD patients vary significantly in morphology/shape [25] . In the previous section , deep CNNs was applied to train and learn the diverse RBC patterns from RBC microscopy imaging data ( see Table 1 ) . Hence , by utilizing this deep CNNs we can classify sickle RBC in different types according to training . In addition to RBC type classification we perform shape factor analysis for each RBC type to further quantify specific RBC shape parameters derived from the contour analysis of the individual RBCs . Three kinds of shape factors were calculated in this work . The shape factors’ formulas and pseudo-code for the specific implementation of the automatic RBC shape factors quantification method are given in S2 Appendix . On the basis of the above automated image-based shape factor analysis scheme , we can perform a comprehensive shape analysis for the classified RBCs or unclassified RBCs according to specific practical applications and requirements .
In order to test the performance of the deep convolutional neural network model , we conducted systematic convergence studies with respect to the number of iterations and the learning rate; here we show some representative results . For the case of 4 patients ( Exp_I ) , we evaluated the training error and loss in the configuration of different learning rates ( 0 . 01 and 0 . 03 ) , batch size = 20 , image size is 78*78 and weight decay is 0 . 01 . In Fig 9 , we observe that both the train and the loss errors decay with the increasing number of epochs , and the higher learning rate can accelerate the decay speed , see the corresponding plots of the loss and train error results for two comparative experiments ( T1 and T2 ) with different learning rate settings . Moreover , another significant observation in Fig 9 is that both the train error and loss results start fluctuating after 15 iterations for T2 and 25 iterations for T1 . In particular , the fluctuations in the loss increase as the number of iterations increases , but the train error has a relatively smaller fluctuation . In order to better understand the fluctuation problem ( so-called “over-training” or “overfitting” ) , we optimized the batch size and use the “dropout” scheme proposed in [28] to overcome this problem . As described before , the dropout layer is implemented after the convolution layer ( p = 0 . 5 ) . Finally , when the number of iterations reaches 60 , our RBC-dCNN model achieved optimal prediction performance . We plotted the two normalized confusion matrices with respect to different number of maximum iteration times ( 30 and 60 ) in Fig 10 . In Fig 10 , we observed that the Discocytes and Granular classes have relative low prediction accuracy among the 8 classes of RBC before the convergence of loss and training error . However , when the maximum number of iterations was 60 , there was a significant improvement in the accuracy of different class prediction due to further decay of the loss and training errors . Table 2 gives detailed performance analysis of the running time , train error , test error and loss with respect to different maximum iteration times based on Exp_I dataset . Despite a learning model being trained to fit the statistics , the model cannot be assumed to have a successful predictive capability . This is due to the regularization which increases the performance , while the performance on test is optimal within a range of values of the regularization parameter . Thus , accurate evaluation of predictive performance is a key step for validating the precision and recall of a deep neural network classification model . K-fold cross-validation is an effective way to measure the predictive performance for the deep CNNs model [29] . The K-fold cross-validation procedure is shown in Fig 11 . First , the total RBC population was divided into k non-overlapped subsets with equal number of RBCs ( here k was chosen to be 5 ) . Then , for every fold or experiment , one of the 5 subsets was chosen as the validation set ( green color data block ) and the other k − 1 subsets were combined to form the training set ( orange color data blocks ) . Finally , the average validation scores obtained from the five folds were calculated as the final prediction score . Every class of RBC images is divided into 5 equal subsets , the quantity of training data and validation data can in each class can be expressed as Eq ( 6 ) . { Sum ( V ij ) = C ( i ) / 5 , Sum ( T ij ) = 1 - C ( i ) / 5 , i ∈ [ 1 , n ] , j ∈ [ 1 , 5 ] ( 6 ) Where , C ( i ) is the number of RBC in the ith class , Vij describes the i-th class and j-th validation sub-dataset , and Tij is the corresponding training dataset . n can take the values of 5 or 8 for our studies . Finally , 5 folds can be generated by collecting the same subset from different classes alternately . For example , the j-th fold can be represented by Eq ( 7 ) . fold ( j ) = { V 1 j , V 2 j , … , V n j } ∪ { T 1 j , T 2 j , … , T n j } ( 7 ) The main advantage in using k-fold cross validation is that each image is limited to one use during the validation process . This can effectively avoid the inaccurate and unstable phenomenon while artificially forcing multiple common samples into both training and testing . Hence , in order to evaluate the general performance of our RBC-dCNN model , we performed 5-fold cross validation for the new datasets Exp_II ( 7 patients ) , in which we created a supplement for the number of echinocyte , granular , sickle and reticulocyte categories . To evaluate the performance of our deep CNN model for the SCD RBC classification and determine the importance of different types of RBCs present in SCD blood , we perform the experiments according to the following principles: Precision = T P / ( T P + F P ) ( 8 ) Sensitivity = T P / ( T P + F N ) ( 9 ) Specificity = T N / ( F P + T N ) ( 10 ) F - score = 2 * T P / ( 2 * T P + F P + F N ) ( 11 ) Accuracy = ( T P + T N ) / ( T P + F P + F N + T N ) ( 12 ) Here , TP , TN , FP and FN are , respectively , the true positive , true negative , false positive and false negative number of RBCs being classified for each class . The above five metrics can help us measure the dCNN’s performance from different perspective; e . g . , the precision , or positive predictive value ( PPV ) can be viewed as a measure of a classifiers exactness . A low precision can also indicate a large number of False Positives . The sensitivity—also called recall or true positive rate ( TPR ) – measures the proportion of positives that are correctly identified; it can be viewed as a measure of a classifiers completeness . A low recall indicates many False Negatives; the specificity ( SPC ) —also known as true negative rate ( TNR ) – measures the proportion of negatives that are correctly identified . F1-score considers both precision and recall; it gets the best accuracy when it reaches 1 , worst corresponds to 0 . The ROC-AUC curve is a plot for TPR and NPR ( Negative Positive Rate ) , which is explained in the experiments below . In the following , the experimental results based on 5-fold cross validation for the two kinds of labeling datasets are presented respectively . Evaluation of coarse-labeled RBC dataset ( 5 categories ) : In this experiment , all RBC patch images were coarsely labeled into 5 categories: 1 ) Dic+Ovl , 2 ) Ech , 3 ) El+Sk , 4 ) Grl , 5 ) Ret . In accordance with the cross validation scheme in Fig 11 , the divided 5-fold cross validation datasets for 5 types of RBC and their corresponding evaluation results are given in Table 3 . As seen from Table 3 , the mean accuracy for training of 5 types of RBC classification under different folds is 91 . 01% , and the mean evaluation accuracy is 89 . 28% . Here , in order to better visualize the discriminative capability of the training deep CNNs model for RBC classification and investigate the sensitivity of the deep CNN model to various RBC categories , the Receiver Operating Characteristic ( ROC ) curve was used to plot the true positive rate ( TPR ) against false positive rate ( FPR ) for different classes of the 5-fold test . The top-left corner of a ROC plot is the “ideal point” while the diagonal dashed line indicates random chance or luck probability . Therefore , the closer the curve followed the left-top border of the ROC space , the more accurate the test can be considered . We also computed the AUC ( Area Under the Curve ) for each ROC curve to evaluate the prediction performance of our RBC-dCNN model . Fig 12 shows the corresponding ROC-AUC results for RBC classification with 5 target categories . In the ROC-AUC plot , the average ROC curve was calculated and shown in blue color , and the corresponding averaged AUC for each fold is at least 0 . 97 . Regarding the prediction performance of the five RBC classes , Granular and Echinocytes received a relative low AUC value , and the other two classes ( “Discocytes+Oval”and “Elongated+Sickle” ) obtained a high AUC value . In addition , Fig 13A shows the corresponding confusion matrix , which can guide humans to observe the confusing classes in red circles; for instance , Ret and Ech are a pair of confusing classes , and the diagonal represents the correctly predicted number of each observation . The calculated sensitivity ( right column ) and precision ( bottom ) for each class in yellow color are consistent with the bars in the statistic Fig 13C . Except for these measures , three other measures are also computed for the performance analysis of the experiment , however , the difference in accuracy among the 5 type of RBCs is small because it refers to the true predictions ( TP and TN ) among the total validation dataset . However , high accuracy is not enough to demonstrate the goodness of the classifier , nor it can tell the sensitivity , precision , specificity and F-score . Therefore , it is necessary to explore these measures for a more in-depth analysis in the experiment . In Fig 13C , the Ret RBCs have a low recall ( sensitivity ) , and the Ech get the lowest precision among the five classes . F-score can be applied to harmonize the above two evaluation metrics; the comparison results of F-score , precision and recall of 5 classes are shown in Fig 13B . Throughout all the evaluation measurements , we can obviously observe that the deep CNN model get a high accuracy and precision in predicting the different types of RBCs , in particular for “Dic+Ovl” , “El+Sk” and “Ret” types . Evaluation of refined-labeled RBC dataset ( 8 categories ) : To evaluate the robustness of the deep CNN model in the application of more rich types of RBC classification , a refined labeling dataset “Exp_II” was generated , which included 8 types of RBC: Dic , Ech , El , Grl , Ovl , Ret , Sk and Sto . Similarly , 5-fold cross validation was carried out and the classification result is shown in Table 4 . The mean evaluation accuracy for the 8 types of RBC classification was 87 . 50% . The corresponding mean ROC-AUC result for the refined labeling test is shown in Fig 14 . The average AUC value for 8 types of RBC is 0 . 94 as opposed to an average AUC value of 0 . 97 for the coarse labeling RBC classification . The RBC Categories ( El and Ovl ) got a relative low classification performance with an AUC value of 0 . 92 . In addition , in Fig 15A , we see a more detailed confusion matrix for classification of 8 RBC categories . This shows the most confused classes ( red circles ) for each type of RBC; Fig 15b and 15c give a performance comparison among the 8 categories . Dic reached the best values for each metric and exhibited a sensitivity of 94 . 4% with high class-specific precision on testing sets of 1434 RBC images . Ovl type achieved the lowest precision and recall due to the misclassification with Dic and El types . From the prediction result example in Fig 16 we can observe that some Ovl type RBCs ( e . g . the RBC in red frame of Fig 16 ) are misclassified as Dic and the El type RBCs are prone to be classified to Ovl type and Sk type , e . g . , the RBCs in blue and green frames in Fig 16 . Based on the proposed deep RBC-CNN model , we perform an independent RBC classification test on 8 raw microscopy images in the highest density RBC fraction i . e . fraction 4 ( typically associated with severe SCD ) . Statistical quantification results for the number of different types of RBC are shown in Fig 17 . Notice the significant heterogeneity of cell types even at the highest density fraction . In addition to the above two experiments ( EXP_I , EXP_II ) on coarse-labeled and refine-labeled sickle RBC classification , in order to test the RBC-dCNN model for oxygenated and deoxygenated RBCs in SCD , we also performed a patient-specific experiment on the classification of a new experimental dataset that includes the previous coarse-labeled five catergories under normoxic conditions ( Oxy ) and a new catergory: “El+Sk under deoxygenation ( DeOxy ) ” , see appendix for details on the experimental methodology . The specific experimental dataset ( EXP_III ) is shown in Table 5 ( row 6 ) and it includes 81 El+Sk ( DeOxy ) RBCs , which after data augmentation correspond to an equivalent sample of 486 DeOxy RBCs . In order to appreciate the differences in RBCs under Oxy and DeOxy conditions , we present in Fig 18 images of RBCs before and after deoxygenation . Even under Oxy , these particular RBCs have crenated shape because they are irreversibly sickled . However , upon deoxygenation we see that there is further polymerization of sickle hemoglobin ( HbS ) inside the RBCs , manifested by the roughening of the contours of RBCs as well as the alteration of the texture inside the RBCs . While the change in the overall shape of these deoxygenated RBCs is relatively small compared to their Oxy state , the differences are subtle and hence they present a new challenge for our dCNN . Having this new mixed Oxy-DeOxy dataset ( EXP_III ) and the particular RBC inner pattern alteration characteristics , we carried out the dCNN model training and testing using the previous similar 5-fold cross validation schema , which involves four folds for training and one fold for testing . We have a total of 988 RBCs for training , which we arrange in 50 batches of 20 images each except the last one that has only 8 RBCs; see Fig 19A . So each batch contains 20 different RBCs , which may be in any of the six categories that the dCNN model should learn . The RBCs are randomly shuffled before input to dCNN . The hierarchical features can be extracted by dCNN layer-by-layer . For instance , the learned feature maps in the hidden 5th-layer for batch 1 is shown in Fig 19B . We observe that the convolutional operation can extract and highlight image features based on the raw image data field directly and hierarchically , such as detecting the image key points , edges , curves , etc . This is further illustrated in Fig 20 , which presents a sequence of feature maps for different layers ( layers 5 , 6 , 8 ad 10 ) corresponding to four different classes of RBCs in Oxy and DeOxy states . As we move to high layer numbers , we pick up more features from low level to high level , hence bridging the gap between high level representation and low level features . Within each layer different filters can be learned from the data to help extract different features . The images shown in Fig 20 correspond to arbitrary selection of filters for each layer . The original raw images are shown on the first column of Fig 20 . Hence , the learned hierarchical convolutional features corresponding to variant learning filters play an important role in classifying RBCs in SCD , in particular for the classification of Oxy and irreversibly DeOxy sickle RBCs . The final prediction result for the classification of deoxygenated RBCs is shown in Fig 21 for the elongated and sickle ( DeOxy ) category . If there is an obvious intracellular pattern change , then the accuracy of our trained dCNN model can obtain a high recall ( 93 . 8% ) but a relatively low precision ( 60 . 0% ) . The main reason for this phenomenon can be justified as follows: Taken together , the above observations imply that both intracellular patterns and RBC contours play a significant role in classification ( see Fig 18 , second row ) . Our proposed RBC classification methodology also relies on the extraction of individual RBCs shape factors that is complementary to RBC classification . Two of the most prevalent shape factors are the Circularity Shape Factor ( CSF ) and Ellipticity Shape Factor ( ESF ) ( Also see S2 Appendix ) [30–32] . We computed the CSF and ESF shape factors for the classified RBCs obtained with the RBC-dCNN methodology ( see Fig 22 ) . The graph is a statistical visual representation of the classified RBCs ( i . e . , Elongated , Oval and Discocytes ) within the ellipticity and circularity shape factor mapping . In addition to these two factors , we can implement in the workflow and compute any of the additional 12 shape factors mentioned in Table S-I to quantify SCD patient-specific RBC shape parameters . The results here are consistent with results described by Horiuchi et al . [30] . In summary , we have used patient-specific microscopy images to develop an automated , high-throughput , ex-vivo RBC classification method for the sickle cell disease based on pre-extraction of RBC region and deep CNNs . We employed a hierarchical RBC patch extraction method followed by a shape-invariant RBC patch normalization technique for the input of our deep nets , which can exclude unnecessary background patches and save time during both the training and the learning procedures . Moreover , our experiments for two kinds of labeling datasets ( 5 and 8 classes ) based on different partition levels demonstrate the great capability and robustness of our RBC-dCNNs model on the classification of various RBC categories with characteristics of complex patterns and heterogeneous shapes without the need for hand-crafted feature pre-extraction . While most of the dCNN training was done based on oxygenated SCD RBCs , we also conducted the classification of deoxygenated RBCs , and demonstrate that our model can detect the deoxygenated RBCs with high accuracy capturing the subtle intracellular texture alterations . Furthermore , the explicit shape analysis at the end of the procedure can offer a robust morphological quantitative tool expanding the proposed framework to high-throughput , ex-vivo RBC classification . Our program is written in Python language and C language , and it currently runs on CPUs , but it can also be updated to run on GPUs . It is mainly based on Python open-source libraries Theano , Numpy , SciPy and matplotlib , etc . The program takes only a few seconds on a standard desktop to test over a thousand RBCs using the trained deep neural network model . In SCD , the shape of sickle RBCs is directly related to the polymerization process inside the RBC , which , in turn , depends on the de-oxygenation rate and hence the specific human organ where a sickle cell crisis may occur , consistent with clinical observations . The ability to perform high-throughput morphological classification utilizing deep CNNs of individual RBCs or other cell types , ( e . g . , white blood cells ) opens up complementary avenues in medical diagnostics for highly heterogeneous cell populations such as in hematological diseases and stored blood used for transfusion . The framework presented here is powerful but many aspects can be further improved in future work . For example , new work should aim to: ( 1 ) develop an accurate segmentation method for the overlapped RBCs in the microscopy image; ( 2 ) increase the dataset scale on the number of rare categories , e . g . sickle , granular , stomatocytes , etc . ; and ( 3 ) build a golden standard library containing diverse SCD RBC categories . Given the success of dCNN in classifying deoxygenated RBCs , having been trained mostly with oxygenated RBCs , we believe that with the proper training of dCNN , the overall methodology for classification we propose could be effective in other hematological disorders , e . g . , diabetes mellitus , elliptocytosis , spherocytosis , as well as in classifying other cells , e . g . , cancer cells , and even detecting the activation state of platelets .
|
There are many hematological disorders in the human circulation involving significant alteration of the shape and size of red blood cells ( RBCs ) , e . g . sickle cell disease ( SCD ) , spherocytosis , diabetes , HIV , etc . These morphological alterations reflect subtle multiscale processes taking place at the protein level and affecting the cell shape , its size , and rigidity . In SCD , in particular , there are multiple shape types in addition to the sickle shape , directly related to the sickle hemoglobin polymerization inside the RBC , which is induced by hypoxic conditions , e . g . , in the post-capillary regions , in the spleen , etc . Moreover , the induced stiffness of RBCs depends on the de-oxygenation level encountered in hypoxic environments . Here , we develop a new computational framework based on deep convolutional networks in order to classify efficiently the heterogeneous shapes encountered in the sickle blood , and we complement our method with an independent shape factor analysis . This dual approach provides robust predictions and can be potentially used to assess the severity of SCD . The method is general and can be adapted to other hematological disorders as well as to screen diseased cells from healthy ones for different diseases .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results/discussion"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"neural",
"networks",
"genetic",
"diseases",
"neuroscience",
"factor",
"analysis",
"mathematics",
"red",
"blood",
"cells",
"statistics",
"(mathematics)",
"artificial",
"intelligence",
"hemoglobinopathies",
"autosomal",
"recessive",
"diseases",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"deoxygenation",
"imaging",
"techniques",
"animal",
"cells",
"mathematical",
"and",
"statistical",
"techniques",
"chemistry",
"hematology",
"clinical",
"genetics",
"sickle",
"cell",
"disease",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"cellular",
"types",
"statistical",
"methods",
"machine",
"learning"
] |
2017
|
A deep convolutional neural network for classification of red blood cells in sickle cell anemia
|
The flavoenzyme pyranose dehydrogenase ( PDH ) from the litter decomposing fungus Agaricus meleagris oxidizes many different carbohydrates occurring during lignin degradation . This promiscuous substrate specificity makes PDH a promising catalyst for bioelectrochemical applications . A generalized approach to simulate all 32 possible aldohexopyranoses in the course of one or a few molecular dynamics ( MD ) simulations is reported . Free energy calculations according to the one-step perturbation ( OSP ) method revealed the solvation free energies ( ΔGsolv ) of all 32 aldohexopyranoses in water , which have not yet been reported in the literature . The free energy difference between β- and α-anomers ( ΔGβ-α ) of all d-stereoisomers in water were compared to experimental values with a good agreement . Moreover , the free-energy differences ( ΔG ) of the 32 stereoisomers bound to PDH in two different poses were calculated from MD simulations . The relative binding free energies ( ΔΔGbind ) were calculated and , where available , compared to experimental values , approximated from Km values . The agreement was very good for one of the poses , in which the sugars are positioned in the active site for oxidation at C1 or C2 . Distance analysis between hydrogens of the monosaccharide and the reactive N5-atom of the flavin adenine dinucleotide ( FAD ) revealed that oxidation is possible at HC1 or HC2 for pose A , and at HC3 or HC4 for pose B . Experimentally detected oxidation products could be rationalized for the majority of monosaccharides by combining ΔΔGbind and a reweighted distance analysis . Furthermore , several oxidation products were predicted for sugars that have not yet been tested experimentally , directing further analyses . This study rationalizes the relationship between binding free energies and substrate promiscuity in PDH , providing novel insights for its applicability in bioelectrochemistry . The results suggest that a similar approach could be applied to study promiscuity of other enzymes .
Generally , enzymes are perceived as being specific for both their substrates and the reactions they catalyze [1] . Deviations from such behavior are often seen as unwanted side effects or even errors in the biological function of the enzyme that come at an additional energetic cost for the organism . Although this feature has long been recognized to be useful in other contexts , for example in the recognition of multiple antigens by the same germline antibody [2]–[4] , such enzymes are often characterized by poor overall catalytic efficiencies and termed promiscuous . Starting in 1976 , this paradigm started to shift when Jensen drew a link between promiscuity and protein evolution [5] . He hypothesized that the first enzymes had very broad substrate specificities that evolved to more specialized forms via duplication , mutation , and selection of the corresponding genes . This was corroborated by later studies that investigated the evolutionary implications of promiscuity such as the adaption of enzymes towards novel xenobiotics , e . g . halogenated compounds or antibiotics , in the course of a few decades [6] , [7] . Although systematic screens for promiscuous enzyme functions are not feasible because of the vast number of possible different substrates and reactions , there are many indications and examples that promiscuity is rather the rule than the exception [6] . Especially in the past two decades , enzyme promiscuity received considerable attention , and enzymes that can take over the function of related enzymes in an organism via their promiscuous activities have been extensively investigated [8]–[10] . These studies suggest that metabolic pathways are intertwined in many unexpected ways , which ultimately gives the organism a higher survival potential under changing environmental conditions . Regulation of such metabolic pathways as well as promiscuity itself at the gene- , transcript- , protein- , and localization-level and the associated reaction conditions are other thriving research areas [1] , [11] . Moreover , promiscuity is often observed for close homologs in protein families and distant homologs within superfamilies [11] . Individual family members have frequently evolved from a common ancestor through gene duplication and subsequent specialization . These members share the same fold and catalytic strategy , and consequently the main activity of one family member is often found as the promiscuous activity of another family member . Nobeli and coworkers refer to this phenomenon as ‘family’ promiscuity as opposed to ‘individual’ or ‘pure’ promiscuity , which is associated to multiple functions of a single enzyme [11] . The molecular mechanisms underlying promiscuity are manifold , including post-translational modifications , multiple domains , oligomeric states , protein flexibility , partial recognition , multiple interaction sites or a single site with diverse interacting residues , allosteric interactions , flexibility as well as size and complexity of the interaction partner , chemical scaffolds , and protonation states of active site residues [6] , [11] , [12] . Hydrophobic interactions , diverse hydrogen bonding , flexibility , and nonpolar van der Waals interactions combined with negligible electrostatics were found to be the main driving forces for promiscuity [11] , [13]–[15] . Consequently , understanding the molecular mechanisms and energetics leading to enzyme promiscuity is a valuable asset in the field of protein design and engineering as well as drug development , and therefore they have been investigated extensively [1] , [16] . In view of various causes and effects involving promiscuity , it is not surprising that the definition of the term is not exact and combinations of different definitions occur [1] , [6] , [11] , [16] , [17] . In this article , the term ‘promiscuity’ is used in the context of relaxed substrate specificity [1] , [18] in order to perform similar chemical reactions on related substrates [17] . A prototypical example of ‘family’ promiscuity [11] can be found in the structural family of GMC oxidoreductases , named after three representatives utilizing either glucose , methanol , or choline as their substrate [19] . Although the four initially characterized GMC family members share only between 23 and 32% sequence similarity and possess diverse catalytic activities with a wide range of substrate specificity , they share the same overall architecture and catalyze a similar chemical reaction , the oxidation of an alcohol moiety . Cavener speculates about an ancestral protein of this family that could bind to many different substrates [19] , which it converted with low catalytic efficiencies while mutations paved the way for high specificities towards the individual substrates . A more recent addition to the GMC oxidoreductase family is the glycosylated enzyme pyranose dehydrogenase ( PDH , EC 1 . 1 . 99 . 29 ) , reacting with many different carbohydrates . It contains a monocovalently bound flavin adenine dinucleotide ( FAD ) and has a mass of approximately 65 kDa . Although PDHs from other sources with similar biochemical properties have been studied [20]–[22] , the enzyme from Agaricus meleagris has been characterized most extensively so far: a wealth of biochemical data [13] , [23]–[25] and a high-resolution X-ray structure with PDB code 4H7U [26] are readily available . To date , its exact physiological role is unknown . However , because of the natural habitat of Agaricus meleagris on lignocellulose-rich forest litter and PDH's reactivity with a multitude of different carbohydrates found during lignin degradation , PDH is most likely involved in lignocellulose breakdown [25] . Compared to other GMC oxidoreductases , it oxidizes many different aldoses and ketoses in pyranose form as well as heteroglycosides , glucooligosaccharides , sucrose , and lactose , which can be ( di ) oxidized at C1–C4 . A comprehensive list of its impressively broad substrate spectrum can be found in the paper of Sedmera and coworkers [27] and an updated version in the review of Peterbauer and Volc [25] . The reactivity towards many different carbohydrate-substrates makes PDH a very interesting enzyme to study in the context of substrate-promiscuity . In this manuscript , we investigate computationally the promiscuous nature of PDH towards the pyranose form of monosaccharides that are turned over by the enzyme . All 32 possible combinations of α- and β-anomers as well as d- and l-stereoisomers of glucose , mannose , galactose , talose , allose , altrose , gulose , and idose will be considered . Molecular dynamics ( MD ) simulations of PDH were applied to study the interactions with the monosaccharides described above . The aim of this study was to gain a deeper understanding of the promiscuous nature of PDH towards monosaccharides . This involved a generalized approach of extensive MD simulations and free energy calculations using the one-step perturbation ( OSP ) method [28] , [29] to calculate monosaccharide binding and solvation . The OSP method is an efficient means to obtain free-energy differences of similar molecules from a simulation of a carefully designed reference molecule for which the sampling is such that configurations are sampled that are representative of the molecules or states in which one is interested . In the past , OSP has successfully been applied to reproduce and predict binding free energies of a series of compounds to a common receptor [29] , [30] , to study the stereoselective binding of substrates to a promiscuous enzyme [31] , [32] , to study conformational preferences of molecules that show slow transitions in regular simulations [33] , [34] or the effect of changes in force-field parameters on conformational equilibria [35] , [36] . In the current work , we investigate ( i ) the solvation free energies of the 32 above-mentioned monosaccharides in water; ( ii ) the free energy differences of the α/β-anomers of the d-stereoisomers in water; ( iii ) the relative binding free energies for all monosaccharides and ( iv ) the occurrence of reactive poses for all monosaccharides . Where experimental data were available , comparisons were made and good to excellent agreements were observed . Furthermore , our work offers predictions of properties that have not yet been described experimentally .
The structure preparations were essentially performed as reported previously [13] . In short , a preliminary version of the 1 . 6 Å resolution X-ray structure of PDH ( PDB code 4H7U ) served as starting point [26] . The covalent monoatomic oxygen species at C ( 4a ) , which is most likely an X-ray artifact , was removed . As a glycoprotein , the structure of PDH comprised covalently attached sugar moieties at surface residues Asn-75 and Asn-294 . The influence of these glycosylated residues on the active site is expected to be negligible and consequently the glycan structures were removed . A PO43- ion at the surface , which is most likely a crystallization buffer artifact , was removed as well . The amino and carboxy termini were charged; all arginines , cysteines and lysines were protonated , and all aspartates and glutamates were deprotonated . In our previous study , we propose that PDH oxidizes its sugar substrate via a general base proton abstraction [13] , which requires one of the two active site histidines ( His-512 and His-556 ) being neutrally charged . The most stable protonation state fulfilling this requirement was obtained when His-512 was fully protonated and His-556 was in its neutral state ( proton at Nε ) . The selection of the tautomeric state for the neutral His-556 was such that in the X-ray structure its deprotonated nitrogen atom pointed towards the active site . The remaining histidines were doubly protonated , except for His-103 , which is covalently attached to the FAD and was protonated at Nδ . No structure of PDH comprising a monosaccharide-substrate in the active site was available at commencement of this work . Therefore , PDH and the closely related GMC oxidoreductase pyranose 2-oxidase ( P2O , EC 1 . 1 . 3 . 10 ) were superimposed with an atom-positional root-mean-square deviation ( RMSD ) of 0 . 13 nm for all heavy atoms of their sugar-binding sites . Two different P2O structures were used , in which the bound sugar roughly differs in a 180° rotation around the axis going through C2 and C5 of the tetrahydropyrane ring to allow for ( di ) oxidations at all possible sites ( C1–C4 ) . Superposition of PDH and P2O in complex with 3-fluoro-3-deoxy-β-d-glucose ( PDB: 3PL8 ) [37] yielded pose A ( Fig . 1A ) , whereas pose B ( Fig . 1B ) was obtained by aligning PDH with P2O in complex with 2-fluoro-2-deoxy-β-d-glucose ( PDB: 2IGO ) [38] . After grafting the monosaccharide coordinates into PDH's active site , the fluorine of the sugar was replaced by a hydroxyl group . This procedure ultimately yielded system PDH-SUG , with the monosaccharide bound to PDH according to pose A or pose B . For simulations of sugar without PDH , the coordinates of β-d-glucose from P2O-PDB 2IGO [38] were used . For the description of the interactions with the sugar , a united atom force field was used . Chirality around CH-groups in such a force field is enforced through an improper dihedral potential energy term . In order to allow transitions between equatorial and axial positions of the attached hydroxyl groups and to sample all 32 possible monosaccharides in a single MD simulation , following changes were made to the topology of β-d-glucose , following suggestions in references [28] and [31] as indicated in Fig . 2: The building block and the changes made to the β-d-glucose topology to define the three reference molecules are further detailed in the supplementary material ( Table S1 ) . MD simulations were conducted using the GROMOS 11 software package [39] employing the 53A6 force field [40] . In this parameter set , carbohydrates are described according to the parameter set of [41] , and the topology of β-d-glucose is given in the supplementary material . Note that the GROMOS force field is a united atom force field , which is crucial for the modifications of the sugar interactions described in the previous section . His-103 and FAD were covalently attached to each other and their topologies and force field parameters were adapted accordingly . All systems were energy-minimized in vacuo employing the steepest-descent algorithm: for the PDH-SUG complexes , the sugar atoms were energy-minimized with constrained PDH coordinates after which both SUG and PDH atoms were energy-minimized . A 1-µs stochastic dynamics ( SD ) simulation of SUG in vacuum was performed , referred to as SUGvac , for which the energy-minimized structure of SUG was used . For MD simulations of SUG and the complex PDH-SUG in water ( SUGwater and PDH-SUG , respectively ) the structures were placed into a rectangular , periodic , and pre-equilibrated box of SPC water [42] . All water molecules within 0 . 23 nm of any solute atom were removed from the box and a minimum solute to box-wall distance of 0 . 8 nm was enforced . In order to relax unfavorable atom-atom contacts between the solute and the solvent , energy-minimization of the solvent was performed while keeping the solute positionally restrained using the steepest-descent algorithm . Finally , five water molecules with the most favorable electrostatic potential for replacement by a positive ion were substituted with sodium ions to achieve electroneutrality in systems PDH-SUG . The following protocol was used to thermalize and equilibrate the system: initial velocities were randomly assigned according to a Maxwell-Boltzmann distribution at 50 K . All solute atoms were positionally restrained through a harmonic potential energy term with a force constant of 2 . 5×104 kJ mol−1 nm−2 in order not to disrupt the initial conformation , and the systems were propagated for 20 ps . In five subsequent 20 ps MD simulations , the positional restraints were reduced by one order of magnitude and the temperature was increased by 50 K . Subsequently , the positional restraints were removed , roto-translational constraints introduced on all solute atoms [43] , and the systems were further equilibrated , each for 20 ps at 300 K . Finally , an equilibration at a constant pressure of 1 atm was conducted for 300 ps . After equilibration , production runs at constant pressure ( 1 atm ) and temperature ( 300 K ) were performed . For the SUGwater systems , one production run of 100 ns was performed . For the PDH-SUG systems , each with SUGa , SUGab , or SUGabc bound according to pose A or pose B , two independent 50 ns production runs ( termed md1 and md2 ) were conducted , leading to a total of 12 independent PDH-SUG simulations . Pressure and temperature were kept constant using the weak-coupling scheme [44] with coupling times of 0 . 5 and 0 . 1 ps , respectively . The isothermal compressibility was set to 4 . 575×10−4 kJ−1 mol nm3 , and two separate temperature baths were used for solute and solvent . The SHAKE algorithm was applied to constrain all solute bond lengths [45] as well as the solvent geometry in simulation SUGwater . Because of simulation efficiency , the SETTLE algorithm was applied to constrain solvent geometry [46] in system PDH-SUG . In all cases , constraining the bond lengths allowed for 2-fs time-steps . Nonbonded interactions were calculated according to a triple range scheme . Interactions within a short-range cutoff of 0 . 8 nm were calculated at every time-step from a pair list that was updated every fifth step . At these points , interactions between 0 . 8 and 1 . 4 nm were also calculated explicitly and kept constant between updates . A reaction field [47] contribution was added to the electrostatic interactions and forces to account for a homogenous medium outside the long-range cutoff using a relative dielectric constant of 61 as appropriate for the SPC water model [48] . Coordinate and energy trajectories were stored every 0 . 5 ps for subsequent analysis . The one-step perturbation ( OSP ) method relies on the application of Zwanzig's perturbation formula which is exact in the limit of infinite sampling [49] . In practice , the free-energy difference between a possibly unphysical reference molecule represented by Hamiltonian and a physically relevant compound represented by Hamiltonian is accurately estimated if a simulation of the reference molecule samples a sufficiently high number of configurations relevant for the real compound . In those cases the free energy can efficiently be calculated using ( 1 ) where the angular brackets indicate an ensemble average computed from the simulation of the reference state and represents Boltzmann's constant multiplied by the absolute temperature . Since only the energy difference appears in the exponential , only the few energy terms that are different between the compounds need to be re-evaluated over the real compounds , here involving only the covalent interactions indicated in Fig . 2 . The free energy differences can subsequently be used to estimate various physically relevant free energy differences , such as the solvation free energies , relative to the reference state ( 2 ) The free energy difference between α- and β-anomers of specific sugars can be computed as ( 3 ) where the subscripts α and β refer to the α- and β-anomers of a single monosaccharide . The binding free energy relative to the reference compound is calculated as ( 4 ) and relative to another compound as ( 5 ) One limitation of the OSP approach is the fact that most simulation effort is spent on unphysical reference molecules , reducing the direct insight into the structure and dynamics of the real compounds . However , the ensemble average of any property Q for the real compounds may be obtained using [50] , [51] ( 6 ) which was used here to analyze the average occurrence of reactive poses for the real compounds . The distances between H-atoms HC1 , HC2 , HC3 , and HC4 and the N5 atom of the FAD cofactor were calculated for the reference state simulations as . Consistent with our previous study [13] , a particular conformation was considered as reactive for a specific carbon if the corresponding value of was below 0 . 3 nm , such that the average occurrence can be calculated as ( 7 ) where H ( x ) is the Heaviside step function , i . e . for and for . By replacing Q in equation ( 6 ) by , we obtain the average amount of catalytically active poses of the real compounds . In the current work , multiple reference compounds R were applied ( SUGa , SUGab , and SUGabc ) whereas individual estimates were combined by transferring the free energy estimates to a common reference state . One can easily show that expressing the ensemble average for reference compound R1 of equation ( 1 ) as an umbrella-weighted ensemble , calculated from a simulation of reference state R2 using equation ( 6 ) , can be expressed as ( 8 ) where both terms on the right-hand side are readily calculated from the simulation of R2 . This way , simulations of the three reference states lead to three estimates of , which can be exponentially averaged to obtain ( 9 ) where the overbar indicates an average over three values of i [52] . Statistical error estimates for the individual ensemble averages used in equation ( 1 ) were obtained from covariances and the statistical inefficiency as described in [53] . The uncertainty in a series of N correlated sequential observations xn , with expectation value , becomes where g is the statistical inefficiency , defined as g = 1+2τ , with τ the auto-correlation time of the normalized autocorrelation function , [53] . The individual error estimates of were subsequently weighted by to obtain the statistical uncertainty on .
To find a suitable reference state , which is crucial for reliable free energy calculations according to the OSP method , MD simulations of system SUGwater with changes to the topology according to SUGa , SUGab , and SUGabc were conducted . As a typical example , Fig . 3 shows the distributions of the improper dihedral angle 5 ( ID5 ) centered on atom C5 for the three 100 ns MD simulations . For ( black ) , ID5 is not evenly distributed and samples mostly the region around +30 degrees . ( red ) and ( blue ) both show more equal distributions , indicating that both stereo-configurations are equally sampled . To use a topology with minimal changes with respect to the real compounds , was selected as the most suitable reference state in water . Similarly , was used for the 1 µs SD simulation in vacuo ( ) . In contrast , SUGa , SUGab , and SUGabc ( Fig . 2 ) were all selected for simulations and analysis of system PDH-SUG , in order to sample as many stereoisomers as possible . Consequently , 12×50 ns MD simulations of system PDH-SUG were conducted: three different SUG topologies , two different SUG binding poses ( pose A and pose B ) and two independent simulations for each ( md1 and md2 ) . MD simulations of system PDH-SUG will be referred to as e . g . ( pose A ) . Table 1 shows the 32 simulated stereoisomers , their 5-digit ID code , and the corresponding sugar names . For systems and , the relative free energies of individual stereoisomers with respect to the reference state in kJ/mol are listed . The range in relative free energies amounts to 13 . 6 kJ/mol ( 20 . 2–33 . 8 kJ/mol ) for system and to 18 . 7 kJ/mol ( 26 . 9–45 . 6 kJ/mol ) for system . In achiral environments such as vacuum and water , no differences in the relative free energies are expected between enantiomers . In Table 1 , sugar-pairs with codes 1 and 32 , 2 and 31 , 3 and 30 , etc . represent enantiomers . Except for enantiomers β-d-talose ( 40 . 8 kJ/mol ) and α-l-allose ( 45 . 6 kJ/mol ) in system , the relative free energies for the enantiomers match very well within both systems: absolute differences between 0 . 1–1 . 8 kJ/mol in system and 0 . 1–2 . 6 kJ/mol in system are roughly within the thermal noise of kBT . The relative free energies for β-d-talose and α-l-allose match qualitatively as they are the two largest within system . The value for α-l-allose seems exceptionally high and omission of this value reduces the range for system to 13 . 9 kJ/mol , similar to the vacuum value . Overall , the small free-energy differences between enantiomers give confidence in the applicability of the reference compound and in subsequent calculations . In the last column of Table 1 , the calculated relative solvation free energies ( ΔΔGsolv ) are listed . To the best of our knowledge , values for these quantities have not been reported in the literature previously , neither from experimental nor from computational sources . While the values of ΔΔGsolv are relative to the reference state , the differences between these values provide insights into the solvation of monosaccharides . Hydrolysis at the pyranose C1 atom allows for interconversion between the α- and β-anomers characterized by a corresponding equilibrium . Table 2 lists the free energy differences of β/α-anomers ( ΔGβ−α ) of all simulated pyranose d-stereoisomers in water , for which comparison with the available experimental data is possible . The calculated values were obtained from equation ( 3 ) , while the experimental values were calculated from previously published experimental estimates of the β/α-pyranose ratios . The experimental values ( at 30°C ) were found to be largely temperature-insensitive [54] , and can be readily compared to the simulation data obtained at 300 K or 26 . 85°C . Estimates of ΔGβ−α from the experimental β/α-pyranose ratios were calculated according to ( 10 ) The ΔGβ−α in Table 2 obtained from MD simulations or experiment have very small absolute deviations in a range between 0 . 0–2 . 3 kJ/mol , which is smaller than the thermal noise , with a mean absolute deviation of 0 . 5±1 . 3 kJ/mol . For each of the 32 simulated monosaccharides of system , we investigated the occurrence of each of the 14 possible ring conformations [55] , [56] of the six-membered pyranose ring by correlating the observed ring conformations with the values of the improper dihedral angles in simulation . We found that sugars with code 1–16 ( d-stereoisomers ) occurred predominantly in the 4C1 chair conformation and sugars with code 17–32 ( l-stereoisomers ) in the 1C4 chair conformation . This again agrees with experimentally observed ring-conformational preferences of the d- or l-series of the studied aldohexopyranoses [55]–[57] , which gives additional confidence in the conducted MD simulations . Fig . 4 shows ( i ) the occurrence of each of the 32 stereoisomers as a function of time and ( ii ) the number of occurrences with a lifetime ≥1 ps . The left two panels are derived from the 100 ns MD simulation of system , the right two panels represent the 50 ns MD simulation of system ( pose A ) , which was selected as a representative example . System nicely samples all stereoisomers and indicates many transitions between the monosaccharides , leading to good statistics for subsequent analysis . System ( pose A ) shows significantly less sampling and transitions of the stereoisomers . Therefore , six MD simulations ( systems , , ; two independent runs each ) were conducted for each pose as mentioned previously . In some of the simulations of the PDH-SUG complexes , the unphysical reference state compound was observed to leave the active site . This may very well represent the proper behavior of these molecules , but unbound mixtures of PDH and SUG are ( i ) not expected to be relevant for real molecules binding to PDH and ( ii ) not part of the thermodynamic cycle to calculate the binding free energies according to equations ( 4 ) and ( 5 ) . For this reason , simulations and for pose A and and for pose B were excluded from the following analyses and four independent simulations of each pose remained . For the time series of relevant distances between PDH and SUG for all simulations see Figures S1 and S2 in the supplementary material . The remaining four MD simulations for each pose were exponentially averaged according to equation ( 9 ) to calculate the free-energy differences ( ΔG ) of individual stereoisomers and their relative binding free energies ( ΔΔGbind ) . According to Fig . 4 , system ( pose A ) clearly samples l-stereoisomers ( sugar code 17–32; 5th digit of improper dihedral code is 4 ) better than d-stereoisomers ( sugar code 1–16; 5th digit of improper dihedral code is 2 ) . This is not surprising , as the transitions of the large CH2-OH group attached at this position are sterically the most hindered ( see Fig . 2 ) . Fig . 5 shows the distributions of all five improper dihedrals ( ID ) for the MD simulations in water and in protein . For system ( top five panels in Fig . 5 ) , the distributions of the ID are derived from the single 100 ns MD simulation , which nicely sampled all 32 possible stereoisomers ( see Fig . 4 , left two panels ) . For system PDH-SUG ( pose A ) ( middle five panels in Fig . 5 ) , and for system PDH-SUG ( pose B ) ( lowest five panels in Fig . 5 ) , the occurrences of the IDs of the four selected MD simulations were arithmetically averaged . Except for ID5 in pose B and to a lesser extent ID3 in pose A , all improper dihedrals show fairly equal distributions and consequently very good sampling . In spite of lower occurrences for one configuration , ID5 ( pose B ) and ID3 ( pose A ) sample both stereoconfigurations . As mentioned previously , a large CH2-OH group is attached at ID5 ( compare Fig . 2 ) and consequently transitions of this group are most sterically hindered in the MD simulations within the protein . Table 3 lists the free-energy differences ( ΔG ) of the 32 stereoisomers simulated in system PDH-SUG ( pose A or pose B ) . The reported ΔG values were obtained by exponentially averaging the four selected MD simulations for pose A or pose B . Because of the chiral environment within the protein , the span of ΔG values significantly increased ( 18 . 9–101 . 8 kJ/mol for pose A; 21 . 7–52 . 8 kJ/mol for pose B ) compared to the MD simulation of system ( 26 . 9–45 . 6 kJ/mol; see Table 1 ) . Moreover , the chiral protein-environment causes significant differences in ΔG between enantiomers . Enantiomers correspond to sugar-pairs with codes 1 and 32 , 2 and 31 , 3 and 30 , etc . The ΔG between enantiomers range from 0 . 7–60 . 1 kJ/mol for pose A and from 1 . 8–24 . 1 kJ/mol for pose B ( in absolute values ) . In addition , the relative binding free energies ( ΔΔGbind ) are listed in Table 3 . They were calculated by subtracting the ΔG values for a certain monosaccharide in the MD simulation of system ( see Table 1 ) from the ΔG of the identical monosaccharide in system PDH-SUG in either pose A or pose B ( Table 3 ) . Note that these values are relative to the reference states and only differences between them have physical relevance . The range for ΔΔGbind for pose A is −11 . 7 to 56 . 2 kJ/mol and −10 . 1 to 23 . 1 kJ/mol for pose B . Interestingly , the ΔΔGbind values for the α- and β-anomers of the d-stereoisomers of glucose are among the lowest of all simulated monosaccharides in both poses ( range between −10 to −2 . 9 kJ/mol ) . Table 4 gives an overview of the relative binding free energies ( ΔΔGbind ) calculated from the experimentally derived Km values [23] and from the combined MD simulations of system PDH-SUG for either pose A or pose B . The experimental values were approximated from the corresponding Km values according to the following formula: ( 11 ) Experimental data were available only for the four listed d-stereoisomers . Because the α- and β-anomers spontaneously interconvert in solution via mutarotation , they cannot be distinguished in experimental binding . The ΔΔGbind values for pose A or pose B were calculated from the MD simulations by first exponentially averaging the free-energy differences of the α- and β-anomers of the respective d-stereoisomers simulated in system ( Table 1 ) or in system PDH-SUG in pose A or pose B ( Table 3 ) . Then , the averaged ΔG values in system were subtracted from system PDH-SUG ( pose A or pose B ) to obtain the ΔΔGbind . For easier comparison , the ΔΔGbind for d-glucose was set to zero . The ΔΔGbind for pose A and pose B were not averaged , as the preference of the reference states for a certain pose is unknown . The ΔΔGbind values derived from simulations of pose A agree well with experiment . Only the difference for d-talose between the experimental and calculated ΔΔGbind ( 5 . 9 kJ/mol ) is larger than the thermal noise . The agreement for pose B matches qualitatively , with differences between the experimental and calculated ΔΔGbind values of 6 . 9 kJ/mol for d-mannose and 7 . 9 kJ/mol for d-talose , which are both above the thermal noise . To conclude , the ΔΔGbind values derived from simulations and experiments match quite well . For successful oxidation , a hydride transfer takes place from the SUG-oxidation site to the N5 atom of FAD [13] . Fig . 6 shows the distances between H-atoms HC1–HC4 of SUG and the N5-atom of FAD . The position of the hydrogen atom in our united-atom representation of the reference state was determined according to ideal geometries and a C-H bond length of 0 . 1 nm . The occurrence of the distances of all four simulations of pose A and pose B were arithmetically averaged . As reported previously [13] , a 0 . 3 nm cutoff was considered in order for a hydride transfer to occur between HC1–HC4 and N5 . Color codes are the same as in Fig . 2A . In pose A ( left panel ) , only HC1 ( green ) and HC2 ( yellow ) are below the mentioned cutoff . In pose B , only HC3 ( red ) and HC4 ( blue ) are below the 0 . 3 nm cutoff . This corresponds very well to previously published data for d-glucose oxidation by PDH , where pose A represents the C2 oxidation mode and pose B the C3 oxidation mode of this particular sugar [13] . In Fig . 7 , the average number of observations of distances between hydrogens attached to C1–C4 and the N5-atom in FAD below 0 . 3 nm for pose A and pose B are shown as calculated for all monosaccharides using equations 7 and 6 . The bars in this logarithmic diagram are non-additive . These reactive distances are compared to the experimentally detected oxidation products . When the distances are below 0 . 3 nm , we will use the ΔΔGbind to evaluate the likelihood of the corresponding monosaccharide to bind to PDH and consequently for a reaction to take place . Note that also low values of can already explain reactions , as substrate binding can easily be a much slower process than the actual reaction . The ΔΔGbind value of β-d-glucose is set to zero in each pose and the ΔΔGbind values of the other sugars are reported here relative to β-d-glucose . Experimental d-glucose conversions yield ( di- ) oxidations at C2 and C3 [25] . This observation is reproduced by MD simulations: in pose A , 0 . 1% of the HC2-N5 distance is below 0 . 3 nm in β-d-glucose; in pose B , the HC3-N5 distance is below the chosen cutoff for 3 . 4% ( α-d-glucose ) and 6 . 5% ( β-d-glucose ) . Again , this observation corresponds very well to our previously published work [13] , where d-glucose is oxidized at C2 in pose A and at C3 in pose B . Experimentally , l-glucose is observed to have C2- and C3 ( di- ) oxidations as well . However , in our MD simulations we do not see any of the relevant HC2-N5 or HC3-N5 distances below the 0 . 3 nm cutoff . Moreover , its relative activity was experimentally measured to be 42% of d-glucose [25] , which does not correspond to the predicted highly unfavorable ΔΔGbind values between 18 . 5–39 . 9 kJ/mol for l-glucose bound in either pose . For d-galactose , MD simulations gave a HC2-N5 distance below 0 . 3 nm 0 . 8% of the time for α-d-galactose in pose A , which corresponds to its experimentally observed C2 oxidation [25] . Moreover , we predict a relatively favorable ΔΔGbind value of 6 . 2 kJ/mol for α-d-galactose in pose A . The sugar d-mannose is a substrate for PDH , however , its oxidation sites have not yet been determined experimentally . The most prominent reactive distance for this sugar is HC3-N5 ( β-d-mannose bound in pose B ) , which is below 0 . 3 nm for 5 . 9% of the time and has a predicted ΔΔGbind value of 12 . 5 kJ/mol . For d-allose , C1 oxidation has been experimentally reported [25] . In the MD simulations , the corresponding HC1-N5 distance is below the 0 . 3 nm cutoff 2 . 9% of the time for α-d-allose ( pose A ) . The predicted ΔΔGbind value for α-d-allose ( pose A ) is 13 . 3 kJ/mol , corresponding to the experimentally determined relative activity of 15% of d-glucose [25] . Experiments revealed solely C1 oxidation for d-talose , which was reproduced by MD simulations with the HC1-N5 distance below 0 . 3 nm 1 . 0% of the time for α-d-talose bound to PDH in pose A . The predicted ΔΔGbind value for α-d-talose ( pose A ) is 5 . 8 kJ/mol , which agrees qualitatively and to a lesser extent quantitatively with the experimentally determined binding affinity according to its Km value ( see also Table 4 ) . The HC4-N5 distance for α-d-talose ( pose B ) is below the 0 . 3 nm cutoff 8 . 8% of the time and the corresponding ΔΔGbind value is reasonably low for binding ( 3 . 1 kJ/mol ) . Nevertheless , C4 oxidation is not reported experimentally , which might be caused by steric clashes of the adjacent hydroxymethyl-group attached to the C5 carbon resulting in poor binding . The last experimentally determined oxidation site for a monosaccharide investigated in this study is available for d-gulose , which is oxidized at C1 [25] only . In the MD simulations , the HC1-N5 distance for α-d-gulose ( pose A ) is indeed below 0 . 3 nm 4 . 4% of the time . The activity of d-gulose was reported to be 7% of d-glucose [25] , which corresponds to an unfavorable ΔΔGbind value for α-d-gulose ( pose A ) of 13 . 9 kJ/mol . In addition to the experimentally determined oxidation sites , we made some striking observations during our distance analyses , which can direct future experiments . High percentages of reactive poses suggesting C1 oxidation are observed for the following sugars bound in pose A: α-d-idose ( 17 . 2% ) , α-l-mannose ( 13 . 3% ) , α-l-galactose ( 85 . 5% ) , α-l-altrose ( 22 . 9% ) , α-l-gulose ( 62 . 3% ) , and α-l-idose ( 45 . 0% ) . Some of these possible oxidation products can be neglected , as the predicted ΔΔGbind value for the corresponding monosaccharides is very unfavorable in pose A: α-l-gulose ( 53 . 6 kJ/mol ) , α-l-galactose ( 42 . 3 kJ/mol ) , and α-l-mannose ( 21 . 1 kJ/mol ) . Others could have low , but measurable activities: α-d-idose ( 12 . 8 kJ/mol ) and α-l-altrose ( 6 . 7 kJ/mol ) . Lastly , oxidation for α-l-idose ( −2 . 6 kJ/mol ) with HC1-N5 below 0 . 3 nm 45% of the time is predicted in pose A . For pose B , we predict HC4 oxidation of α-l-gulose , for which the HC4-N5 distance is below 0 . 3 nm 14 . 9% of the time and the predicted ΔΔGbind value is 13 . 9 kJ/mol , allowing for low but measurable activity . Interestingly , monosaccharides bound to PDH in pose A , for which additional oxidations were observed , are all α-compounds and oxidized at HC1 . This can be rationalized , as the hydroxyl-group attached to C1 defines whether a sugar is an α- or β-anomer . Consequently , if the HC1-N5 is within the reactive distance , the hydroxyl-group attached to that C1 has to be on the opposite side of the HC1 , which ( in pose A ) corresponds to the α-anomer of the respective sugars .
In this study , we presented a generalized approach to simulate monosaccharide solvation in water , as well as binding and product formation in the enzyme PDH . Introducing changes to the monosaccharide topology according to Fig . 2 created systems SUGa , SUGab , and SUGabc , out of which system SUGab was selected as the most suitable reference state for subsequent analysis in water . This allowed for sampling of all 32 possible aldohexopyranoses in only one MD simulation of the reference compound in water or using a handful of simulations of the reference state compounds within PDH . Free energy calculations according to the one-step perturbation method revealed that systems and show a similar range of relative free energies for the simulated monosaccharides . Moreover , the relative free energies for the enantiomer-pairs ( sugar codes 1 and 32 , 2 and 31 , etc . ) match very well within systems and . Because both vacuum and water represent an achiral environment , this outcome is expected and gives confidence in the conducted simulations . We reported calculated values for the relative solvation free energies ( ΔΔGsolv ) of all 32 aldohexopyranoses ( Table 1 ) . To the best of our knowledge , these ΔΔGsolv values have not been reported previously , giving new fundamental insights into the solvation of aldohexopyranoses . For all simulated pyranose d-stereoisomers , we report the free energy differences of the corresponding β/α-anomers in water ( ΔGβ−α ) . The deviations from experimental values [23] , [25] are very small ( within a range of 0 . 0–2 . 3 kJ/mol ) , further increasing the confidence in the conducted simulations . In addition , the pyranose ring conformations for each of the 32 stereoisomers were investigated . Sugars with codes 1–16 ( d-stereoisomers ) occurred predominantly in the 4C1 chair conformation and sugars with codes 17–32 ( l-stereoisomers ) in the 1C4 chair conformation . These results are in line with the experimentally obtained preferences of ring conformations for d- or l-stereoisomers [55]–[57] , furthermore strengthening the validity of the performed simulations . For system PDH-SUG , six MD simulations of 50 ns each were performed for either pose A or pose B . Of these six , two were discarded for each pose from subsequent analysis , because the monosaccharide left the active site . For each pose , the results of the four selected MD simulations were exponentially averaged . All 32 possible stereoisomers were sampled extensively in pose A and pose B . Compared to system ( 26 . 9–45 . 6 kJ/mol ) , the chiral protein environment caused a significant increase in the span of the free-energy differences ( ΔG ) : 18 . 9–101 . 8 kJ/mol for pose A and 21 . 7–52 . 8 kJ/mol for pose B . The relative binding free energies ( ΔΔGbind ) range between −11 . 7 to 56 . 2 kJ/mol for pose A and between -10 . 1 to 23 . 1 kJ/mol for pose B . Where available , the calculated ΔΔGbind values were compared to the experimental binding free energies estimated from Km values . For pose A , the agreement is quite reasonable . Pose B shows qualitative agreement between calculated and experimental ΔΔGbind values . Taking an arbitrary cutoff of +15 kJ/mol relative to β-d-glucose for binding , the predicted binding affinities indicate that measurable binding free energies can be expected for 21 monosaccharides in pose A as well as in pose B . This is in line with the expected promiscuity of the PDH enzyme and suggests that these monosaccharides can be anticipated to inhibit the enzyme and possibly are also substrates themselves . A distance analysis between hydrogens attached to the sugar carbons 1–4 and the N5 atom of FAD revealed that monosaccharide oxidation is possible at HC1 or HC2 in pose A and at HC3 or HC4 in pose B , which is in line with previously published findings [13] . We could reproduce the experimentally detected oxidation products by monitoring the HCX–N5 distance for d-glucose , d-galactose , d-allose , d-talose , and d-gulose . Only for l-glucose , the experimentally observed C2- or C3- oxidation could not be reproduced by the HCX-N5 distance analysis . With a combination of HCX-N5 distance analysis and binding free energy calculations , we predict oxidation products for some sugars , which have not yet been reported experimentally: low but measurable oxidation at HC1 for l-altrose and d-idose as well as at HC3 for d-mannose and at HC4 for l-gulose; strong oxidation at HC1 for l-idose – a challenge for future experiments . To conclude , this study presents a generalized approach to simulate all 32 possible aldohexopyranoses in the course of just a few simulations . It contributes to the rationalization of PDH's substrate promiscuity with a combination of binding free energies and distance analyses for each sugar . This provides insights into PDH's applicability in bioelectrochemistry . We believe that this approach is readily transferable to other promiscuous enzymes , whose substrates differ mainly in the stereochemistry of their reactive groups .
|
Generally , enzymes are perceived as being specific for both their substrates and the reaction they catalyze . This standard paradigm started to shift and currently enzyme promiscuity towards various substrates is perceived rather as the rule than the exception . Enzyme promiscuity seems to be vital for proteins to acquire new functions , and therefore for evolution itself . The driving forces for promiscuity are manifold and consequently challenging to study . Binding free energies , which can be calculated from computer simulations , represent a convenient measure for them . Here , we investigate the binding free energies between the enzyme pyranose dehydrogenase ( PDH ) and a sugar-substrate by computational means . PDH has an extraordinarily promiscuous substrate-specificity , making it interesting for e . g . bioelectrochemical applications . By introducing modifications to the sugar-structure used for the molecular dynamics simulations , we could simultaneously study all 32 possible aldohexopyranoses from a single simulation . This saves costly computational resources and time for setting up and analyzing the simulations . We could nicely reproduce experimental results and predict so far undetected sugar-oxidation products , directing further experiments . This study gives novel insights into PDH's substrate promiscuity and the enzyme's applicability . A similar approach could be applied to study the promiscuity of other enzymes .
|
[
"Abstract",
"Introduction",
"Methods",
"Results/Discussion",
"Conclusions"
] |
[
"computational",
"chemistry",
"molecular",
"dynamics",
"biology",
"and",
"life",
"sciences",
"chemistry",
"physical",
"sciences",
"computational",
"biology",
"biophysics",
"biophysical",
"simulations"
] |
2014
|
Pyranose Dehydrogenase Ligand Promiscuity: A Generalized Approach to Simulate Monosaccharide Solvation, Binding, and Product Formation
|
Cryptococcosis is an important fungal infection in immunocompromised individuals , especially those infected with HIV . In Brazil , despite the free availability of antiretroviral therapy ( ART ) in the public health system , the mortality rate due to Cryptococcus neoformans meningitis is still high . To obtain a more detailed picture of the population genetic structure of this species in southeast Brazil , we studied 108 clinical isolates from 101 patients and 35 environmental isolates . Among the patients , 59% had a fatal outcome mainly in HIV-positive male patients . All the isolates were found to be C . neoformans var . grubii major molecular type VNI and mating type locus alpha . Twelve were identified as diploid by flow cytometry , being homozygous ( AαAα ) for the mating type and by PCR screening of the STE20 , GPA1 , and PAK1 genes . Using the ISHAM consensus multilocus sequence typing ( MLST ) scheme , 13 sequence types ( ST ) were identified , with one being newly described . ST93 was identified from 81 ( 75% ) of the clinical isolates , while ST77 and ST93 were identified from 19 ( 54% ) and 10 ( 29% ) environmental isolates , respectively . The southeastern Brazilian isolates had an overwhelming clonal population structure . When compared with populations from different continents based on data extracted from the ISHAM-MLST database ( mlst . mycologylab . org ) they showed less genetic variability . Two main clusters within C . neoformans var . grubii VNI were identified that diverged from VNB around 0 . 58 to 4 . 8 million years ago .
Infection by Cryptococcus species is considered one of the most important disease in patients living with HIV . It is estimated that around 624 , 000 deaths occur annually due to cryptococcosis , with most of them occurring in sub-Saharan Africa and Southeast Asia [1 , 2] . The infection is mainly acquired through inhalation of dehydrated yeast cells from environmental sources , including pigeon excreta and plant debris [3–5] . Within the lungs , this fungus may cause pneumonia , and is able to disseminate to the central nervous system ( CNS ) where it infects the meninges and brain parenchyma [6–8] . The mortality associated with cryptococcal meningitis varies among different countries and is dependent on several factors , such as the availability and the patient access to antiretroviral therapy ( ART ) , antifungals , as well as the time of diagnosis and elevated CNS opening pressure [9 , 10] . In Africa , despite the increasing availability of ART and amphotericin B in some regions , the mortality rate varies from 17 to 62% [11–14] . Despite ART being provided free of charge by the public health service and thus readily available in Brazil , the mortality in the first week of admission is still 42–51% , which is attributed to the advanced immunosuppression at start of HIV treatment and late diagnosis [9 , 15] . These rates of mortality are higher than those observed in countries with high GNP ( gross national product ) , such as in Europe ( 6 . 5–32% ) [10 , 16] and North America ( 15–26% ) [17 , 18] . The disease is caused by two sibling species subdivided in major molecular types according to different techniques [19–24]: Cryptococcus neoformans , with the major molecular types VNI/VNII/VNB ( C . neoformans var . grubii , serotype A ) , VNIV ( C . neoformans var . neoformans , serotype D ) , and VNIII ( AD hybrids ) ; and Cryptococcus gattii with VGI , VGII , VGIII , and VGIV ( serotypes B and C ) . In addition , some interspecies hybrids such as AB and BD can occur [25–27] . Recently , a new proposal to elevate the major molecular type status to species level was published [28] . However , throughout the current study the classical nomenclature of the C . neoformans / C . gattii species complex is used . Several studies have been performed around the world to identify the major molecular types of the C . neoformans / C . gattii species complex from clinical and environmental sources , as well as to develop a better understanding of their molecular epidemiology [29–33] . The lack of a standardized typing technique to compare the results obtained from different countries , lead to standardization and adaptation of a consensus MLST scheme by the working group for genotyping of C . neoformans and C . gattii of the International Society for Human and Animal Mycology ( ISHAM ) . This MLST scheme has become an important tool for the characterization of the population genetic structure of the Cryptococcus species [20] . Using this MLST scheme , previous studies on C . neoformans var . grubii clearly differentiated VNI/VNII/VNB in addition to the presence of three subpopulations in Asia and up to five subpopulations around the world [34] . Recently , a high genetic diversity of southern African isolates was reported using this MLST scheme [35 , 36] . Interestingly , the African continent not only has the distinct VNB genotype , but it was also found to have both VNI and VNII populations more widely distributed [29 , 36] . This observation led to the African-origin hypothesis for the evolutionary history of C . neoformans [29 , 37] . A number of epidemiological studies have been performed in Brazil using a range of PCR-based techniques [31 , 38 , 39] . However , no data on the population structure of C . neoformans using sequencing-based methods are available . Therefore , it is unknown how the Brazilian isolates fit into a context of the global population structure . To initiate a better overview of the population genetic structure of C . neoformans , the major aetiological agent of cryptococcosis in HIV-positive patients in Brazil , the ISHAM consensus MLST scheme was applied to genotype 143 clinical and environmental C . neoformans isolates from the southeastern Brazilian state Minas Gerais , and the obtained data were then placed in a global context via comparison against the ISHAM-MLST database .
From 1999 until 2014 , one hundred and forty-three C . neoformans isolates recovered from clinical and environmental samples were collected in the regional centre for infectious diseases at the teaching hospital of the Triangulo Mineiro Federal University , Uberaba , Minas Gerais state , Brazil ( S1 and S2 Tables ) . Of these , 108 ( 75% ) were clinical isolates , which were recovered from the following body sites: 82 ( 76% ) from cerebrospinal fluid ( CSF ) , 13 ( 12% ) from blood , 11 ( 10% ) from urine , one ( 1% ) from skin , and one ( 1% ) from bronchoalveolar lavage ( BAL ) fluid . One isolate per patient was selected from most of the cases throughout the study , except in seven situations , where two isolates from different body sites and/or from serial samples were included . For the HIV-positive patients , the CD4+ T-cell count and HIV viral load were determined at maximum two weeks after the diagnosis of cryptococcosis for all patients . The remaining 35 ( 25% ) isolates were recovered from bird droppings obtained in pet shops from different neighbourhoods and debris of trees from surrounding hospital areas ( S1 Table ) . Restriction fragment length polymorphism ( RFLP ) analysis of the orotidine monophosphate pyrophosphorylase ( URA5 ) gene was used to confirm the major molecular type of the isolates . Strains WM 148 ( serotype A , VNI , AFLP1 ) , WM 626 ( serotype A , VNII , AFLP1A and AFLP1B ) , WM 628 ( serotype AD , VNIII , AFLP3 ) , WM 629 ( serotype D , VNIV , AFLP2 ) , WM 179 ( serotype B , VGI , AFLP4 ) , WM 178 ( serotype B , VGII , AFLP6 ) , WM 161 ( serotype B , VGIII , AFLP5 ) , and WM 779 ( serotype C , VGIV , AFLP7 ) , were used as controls for the identification of the major molecular types of the C . neoformans / C . gattii species complexes [40] . The mating type allelic profiles of the STE20 gene was identified by PCR using the following primers: Aα ( JOHE7264/JOHE7266 ) , Aa ( JOHE7270/JOHE7271 ) , Dα ( JOHE7267/JOHE7269 ) , and Da ( JOHE7273/JOHE7274 ) [41 , 42] . To exclude misidentification with AD hybrids homozygous at mating type locus , the GPA1 gene specific for serotype A ( JOHE2596/JOHE3241 ) and serotype D ( JOHE2596/JOHE3240 ) in addition to the PAK1 gene specific for serotype A ( JOHE3066/JOHE3236 ) and serotype D ( JOHE3066/JOHE3065 ) were also amplified as previously described [41 , 42] . The C . neoformans strains H99 ( Aα , VNI ) , KN99 ( Aa , VNI ) , JEC20 ( Da , VNIV ) , and JEC21 ( Dα , VNIV ) were included as controls in all analyses . Cells were grown in yeast peptone dextrose ( YPD ) broth overnight with agitation and then washed twice with phosphate buffered saline ( PBS ) . Following washing the cells they were fixed overnight in 70% ethanol at 4°C with mild agitation . Approximately 107 cells were washed with 1 mL NS Buffer ( 10 mM Tris-HCl pH 7 . 5 , 0 . 25 M sucrose , 1 mM EDTA , 1 mM MgCl2 , 0 . 1 mM CaCl2 , 0 . 1 mM ZnCl2 , 0 . 55 mM phenylmethylsulfonyl fluoride , 0 . 049% 2-mercaptoethanol ) , resuspended in 0 . 2 mL of NS Buffer containing 14 μL RNase A ( Sigma-Aldrich , 1mg/mL ) and 14 μL propidium iodide ( Sigma-Aldrich , 1mg/mL ) , and incubated in the dark at room temperature for 4–6 hours . Then , 50 μL of the stained cell mixture was added to 0 . 5 mL of 1 M Tris pH 7 . 5 and 1 . 5 mL PBS [43] . Flow cytometry was performed on 10 , 000 cells at a slow flow rate with a Becton-Dickinson FACSCanto II . The results were analysed using the software FlowJo ( FlowJo , LLC , OR , USA ) . The haploid reference strains H99 C . neoformans var . grubii VNI and strain CDCR265 C . gattii VGII and the diploid hybrid strain WM 09 . 184 C . neoformans VNII/VNIV were included as controls . The ISHAM consensus MLST scheme for C . neoformans and C . gattii was applied in the current study using the amplification conditions previously described [20] . PCR products of the six housekeeping genes CAP59 , GPD1 , LAC1 , PLB1 , SOD1 , URA5 , and the IGS1 region were commercially purified and sequenced by Macrogen Inc . , Seoul , South Korea . Sequences were manually edited using the software Sequencher 5 . 3 ( Gene Codes Corporation , Ann Arbor , MI , USA ) and aligned using Muscle algorithm available in MEGA 6 . 06 [44] . The allele types and the sequence types ( ST ) were identified via sequence comparisons with the C . neoformans MLST database at http://mlst . mycologylab . org/ . The extent of DNA polymorphisms , such as the number of polymorphic sites ( S ) , number of haplotypes ( h ) , haplotype diversity ( Hd ) , nucleotide diversity ( π ) that gives the proportion of nucleotide differences in all haplotypes , average number of nucleotide differences ( k ) , and Watterson’s estimate per sequence ( θS ) were calculated using DNAsp v5 . 10 . 1 available at http://www . ub . edu/dnasp/ [45] . The neutrality tests Tajima’s D , Fu & Li’s D* , Fu & Li’s F* , and Fu’s Fs were also calculated using this program . Negative results in these tests suggest evidence of purifying selection or population size expansion while positive results suggest balancing selection or a decreasing in population size . The nucleotide diversity was calculated in the southeastern Brazilian dataset and the ISHAM-MLST expanded global dataset by including one of each ST per country and excluding the high number of clonal STs from the ISHAM-MLST database , http://mlst . mycologylab . org/ . The populations were assigned according to continent of origin . The phylogenetic analyses and the geographic distribution of C . neoformans var . grubii VNI isolates were performed as follows: First , the best model for the concatenated dataset was chosen from the Bayesian information criterion ( BIC ) using the software jModelTest 2 . 1 . 7 [46 , 47] . The Tamura Nei model with invariable sites and gamma distribution ( TrNef + I + G ) with p-inv: 0 . 955 and alpha shape: 0 . 832 was selected and used in the phylogenetic analysis . The unrooted maximum likelihood ( ML ) phylogenetic tree was calculated applying a bootstrap of 1 , 000 replicates in MEGA v6 . 06 . In addition , the dataset was submitted to the neighbour-joining ( NJ ) algorithm based on the TrNef + G model [48] to infer the tree congruence . For the ML method , all sites were included in the analyses , while for NJ , all positions containing alignment gaps were eliminated . To evaluate patterns of evolutionary descent among genotypes according to their source and geographic region , the allelic profile of the southeastern Brazilian and the global dataset were applied to the goeBURST algorithm in PHILOVIZ software available at http://www . phyloviz . net/wiki/ [49] . In this analysis , differences between STs are presented as single locus variant ( SLV ) , double locus variant ( DLV ) , and triple locus variant ( TLV ) respectively . The concept of a clonal complex ( CC ) was adopted when a SLV linkage with the founder ST was observed [49 , 50] . The genetic differentiation of pre-defined C . neoformans var . grubii VNI populations ( e . g . clinical and environmental , and according to the continent of origin ) was calculated using the hierarchical Analysis of Molecular Variance ( AMOVA ) implemented in the software GenoDive v2 . 0 [51] and Principle Component Analysis ( PCA ) using the Adegenet v2 . 0 . 1 package for statistical software R v3 . 2 . 4 ( https://www . R-project . org/ ) . In addition , the population differentiation test ( FST ) from an AMOVA [52 , 53] , assuming that the isolates were all haploids or homozygous diploids , was used to test the null hypotheses ( H0 ) of no population differentiation . Values of FST can range from 0 , which implies complete panmixis and means that the two populations are interbreeding freely ( in these scenario we accept H0 and the p value is greater than >0 . 05 ) , to 1 where all genetic variation is explained by the population structure , and that the two populations do not share any genetic diversity . The FST test was calculated with one ST per population ( clone-corrected dataset ) using 1 , 000 permutations in the software GenoDive v2 . 0 and using the Hierfstat package in R applying the Mont Carlo test to infer difference among populations in the PCA analysis . The presence of recombination within the C . neoformans var . grubii VNI population was inferred using the classical index of association IA and rBarD . Both indices were calculated using clone corrected allelic profiles in the software Multilocus v1 . 3 [54] using 1 , 000 randomizations , which simulate infinite panmixis and compare the values of the observed dataset with those artificially generated by the randomization process . Absence of difference between both datasets ( p>0 . 05 ) supports the null hypotheses of linkage equilibrium and sexual recombination while significant differences supports linkage disequilibrium ( LD ) and clonality . The minimal number of recombination events per gene and per population was then calculated in the software DNAsp v . 5 . 10 . 1 [45] . The presence of recombination per gene ( intragenic ) , in the concatenated dataset ( intergenic ) of each population and in the expanded global dataset was also checked by phylogenetic compatibilities of nearby polymorphic sites along single and concatenated sequences in the software SplitsTree v . 4 . 13 . 1 ( http://www . splitstree . org/ ) [55] . Recombination events were visualized by the formation of parallelograms between neighbours using the reticulated algorithm NeighborNet [55] . This analysis was calculated applying the second best model [Kimura 2-parameter with invariable sites and gamma distribution ( K80 + I + G ) with Ti/Tv: 2 . 895 , p-inv: 0 . 957 , and alpha shape: 0 . 920] available in the software jModelTest 2 . 1 . 7 due to the absence of the TrNef model in SplitsTree v . 4 . 13 . 1 [56] . The Pairwise Homoplasy Index ( PHY ) test was used to infer if there was statistical significance for recombination . The actual number of populations ( K ) and their distribution according to source and geographic region was calculated using the Bayesian statistical algorithm [57] implemented in Structure v2 . 3 . 4 ( http://pritchardlab . stanford . edu/structure . html ) . The admixture model was selected due to possible presence of migrant individuals among each population . Twenty runs were performed for each value of K , ranging from 1 to 10 . Each run consisted of Markov Chain Monte Carlo ( MCMC ) simulations of 1 , 000 , 000 interactions and a burn-in period of 100 , 000 generations [58] . The K number was calculated using the software Structure Harvester available at http://taylor0 . biology . ucla . edu/structureHarvester/ [59] . The results were graphed using Clumpp and Distruct software’s available at https://web . stanford . edu/group/rosenberglab/software . html [60 , 61] . The coalescence analysis was performed using the Bayesian molecular clock method as implemented in the BEAST v1 . 8 . 3 software [62] . The Tamura Nei model with invariable sites and the gamma distribution ( TrNef + I + G ) model was selected in the software jModelTest 2 . 1 . 7 and used in BEAST . In order to find the best-fitting clock model we used the stepping stone sampling marginal likelihood estimator available in the MrBayes v3 . 2 software [63] . The relaxed lognormal clock was selected to infer the time scale through the incorporation of one internal node calibration in the node separating the VNI and VNB from VNII . The calibration date of 4 . 5 million years based on genetic distances of different gene sequences was selected from previous studies [64 , 65] . A normal prior age distribution with a standard deviation ( SD ) of 0 . 25 million years covering the ages of the two previous publications was used in the analysis . Next , the input XML file ( S1 Dataset ) was generated in the software BEAUTI v1 . 8 . 3 with a run of 108 generations , 1 tree sampled per 1 , 000 generations , and a burn-in of 10% [62] . Two independent runs were performed and the respective files were combined using the LogCombiner v1 . 8 . 3 distributed with BEAST ( http://beast . bio . ed . ac . uk/ ) using a burn-in of 10% . The effective sample size ( ESS ) was higher than 200 in all analyses and was visualized using the software Trace v1 . 6 . 0 distributed with BEAST . The tree with the highest log clade credibility was selected in the software TreeAnotator v1 . 8 . 3 , also distributed with BEAST . The tree presenting the posterior mean and 95% confidence intervals of the time to the most recent common ancestor was visualized using the software FigTree v1 . 4 . 3 ( http://tree . bio . ed . ac . uk/software/figtree/ ) . The pattern of ancestry for single genes using the global dataset was also estimated using statistical parsimony , which infers the gene genealogies and the most ancestral haplotype using the software TCS v1 . 2 . 1 available at http://darwin . uvigo . es/software/tcs . html [66] . The normality test Shapiro-Wilk was used to infer if the continuous variables presented normal distribution . The Student’s t test , calculated using the car package for statistical software R ( https://www . R-project . org/ ) , was used to analyse differences between two continuous variables . The univariate analysis and odds ratio with 95% confidence interval ( CI ) was performed with the epiR package in R . Those variables that presented a p value < 0 . 25 in the univariate analysis were included in the multivariate logistic regression analysis calculated using the stats package in R . All samples of the study were retrieved from the culture collection of Mycology Laboratory of the Triangulo Mineiro Federal University . All data were de-identified . Institutional Human Research Ethics approval for the study was obtained from the Research Ethics Board of the Triangulo Mineiro Federal University ( protocol #1350 ) . All sequences are deposited in GenBank and their accession numbers are described in S1 Table .
Of the 108 clinical samples studied , only one isolate per patient was included in the 101 cases . In the remaining seven cases , two isolates were included because they were collected from serial CSF samples and/or from different body sites ( S2 Table ) . The underlying clinical information was available for 103 patients , the majority of them ( 95 ) had AIDS , three were kidney transplant recipients with or without concurrent diabetes mellitus , three had only diabetes mellitus and/or diabetes mellitus and nephritis , one had systemic lupus erythematosus , and one had Crohn's disease ( S2 Table ) . For statistical analysis , we excluded the duplicate isolates and the patients without the respective information: outcome , CD4+ T-cell count , and HIV viral load . Overall , complete clinical information was available for 95 patients , 87 of them HIV-positive . The majority were male ( 72 , 78 . 9% ) , with a mean age of 40 . 17 years [Standard Error ( SE ) ±11 . 57] . There was no difference in age between the genders ( male = 39 . 69 ± SE 11 . 29 vs . female = 41 . 65 ± SE 12 . 06 , p = 0 . 48 ) . Among the patients whose isolates were included , 56 ( 58 . 9% ) had fatal outcome , regardless of gender ( p = 0 . 238 ) , and 64 . 9% ( 55/87 ) when only the HIV-infected population was considered . HIV infection was strongly correlated with death in both univariate and multivariate analyses and presented an odds ratio of 14 . 84 times when compared with non HIV-infected patients ( Table 1 ) . Within the HIV-infected group with CD4+ T-cell count and HIV viral load information available ( 77 patients ) , the CD4+ T-cell counts revealed strong association with immunosuppression as half of the patients had counts below 50 mm3/ml . On the other hand , neither CD4+ T-cell count nor HIV viral load showed a statistical association with fatality ( Table 1 ) . URA5-RFLP analysis identified all isolates as C . neoformans var . grubii major molecular type VNI . They were all mating type alpha and serotype A based on PCR with primer sets specific for the mating type locus STE20 , as well as the GPA1 , and PAK1 genes . The majority of the isolates were haploid , but 10 ( 9 . 2% ) of the clinical and two ( 5 . 7% ) of the environmental isolates were diploid ( double DNA content ) based on flow cytometry ( Fig 1 ) . Taken together , these results show a high prevalence ( n = 131; 91 . 6% ) of haploid , serotype A/mating type alpha ( Aα ) isolates and a minority ( n = 12; 8 . 4% ) of diploid ( AαAα ) isolates ( S1 Table ) . MLST analysis of the 143 southeastern Brazilian isolates demonstrated the presence of 13 sequences types ( ST ) . The majority of them represented “high frequency” STs , such as ST93 ( 91; 63 . 6% ) , ST77 ( 19; 13 . 3% ) , ST23 ( 7; 4 . 9% ) , and ST63 ( 6; 4 . 2% ) . The remaining STs were represented by one to three isolates , and ST540 was a newly described sequence type . Ten of the 13 STs were identified in clinical isolates , while seven were present in environmental isolates . ST23 , ST32 , ST39 , ST63 , ST289 , and ST540 were only found among clinical strains , while ST2 , ST15 , and ST77 were found only in environmental isolates . ST93 represented 10 clinical homozygous diploid isolates while ST77 represented the two environmental strains . Of the seven patients from whom two samples each were studied , one patient had isolates ( UFTM 14 . 61/UFTM 14 . 75 ) of the same ST ( e . g . ST93 ) , but different ploidy ( e . g . haploid and diploid ) . Two other patients were co-infected by two isolates presenting different STs with different ploidy ( UFTM 14 . 112 , ST93 diploid/UFTM 14 . 117 , ST23 haploid , and UFTM 14 . 96 , ST32 haploid/UFTM 14 . 114 , ST93 diploid ) ( Fig 1 , S1 and S2 Tables ) . The remaining patients were co-infected by the same STs , all of them haploids . To place the southeastern Brazilian isolates in a global context , we expanded the analysis by including one of each of the STs per country , excluding the high number of isolates from clonal STs from the ISHAM-MLST database ( http://mlst . mycologylab . org/ ) . It is important to mention that human migration will have a blurring effect on the selection of the ST per country , especially because cryptococcal cells can be dormant for decades in the host’s body . To overcome this fact , an effort was made to select most STs from references that didn’t mention migration , or that most of the isolates were from autochthonous cases . A total of 179 isolates and 91 different STs were included , with 148 clinical , 23 environmental , three veterinary , and five isolates from unknown sources ( S1 Table ) . The majority ( 150; 83 . 8% ) belonged to mating type alpha , four ( 2 . 2% ) to mating type a , and the mating type information for the remaining isolates was not available . Serotype A information was available for 44 isolates ( S1 Table ) . The nucleotide sequences of the southeastern Brazilian C . neoformans var . grubii VNI isolates showed between 0 to 13 polymorphic sites ( Table 2 ) . The IGS1 region presented the highest nucleotide diversity ( π = 0 . 0045 ) and mutation rate ( θS = 2 . 348 ) , followed by GPD1 ( π = 0 . 0012; θS = 1 . 084 ) . In contrast , SOD1 was the least variable genetic locus , with only one allele type . The average estimates of these statistics for the concatenated sequences were also low ( Hd = 0 . 507 , π = 0 . 0012 , and θS = 4 . 877 ) , reflecting the low genetic diversity of the isolates . The neutrality tests Tajima’s D , Fu and Li’s D* , Fu and Li’s F* , and Fu’s FS showed evidence of balancing selection or expansion of rare polymorphisms for all loci in the southeastern Brazilian population ( Table 2 ) . Using the expanded global dataset and distributing the isolates into subpopulations , African populations presented the highest genetic diversity ( h = 43 , π = 0 . 0027 , and θS = 13 . 88 for the concatenated sequences ) . The lowest genetic diversity was observed among South American populations ( h = 13 , π = 0 . 0012 ) , followed by North American ( h = 22 , π = 0 . 0020 ) , Asian ( h = 37 , π = 0 . 0020 ) , and European populations ( h = 29 , π = 0 . 0022 ) . Overall , the IGS1 locus was the most variable region ( π = 0 . 0074 ) , followed by SOD1 ( π = 0 . 0053 ) , and LAC1 ( π = 0 . 0021 ) . The neutrality tests for the overall C . neoformans var . grubii VNI population showed evidence of purifying selection or population expansion for all loci ( Table 2 ) . The maximum likelihood analysis of the global dataset revealed the presence of two major clusters within the C . neoformans var . grubii VNI population ( Fig 2; S1 Fig ) . The first ( minor ) cluster was composed of 30 different STs and had higher bootstrap support than the second ( major ) cluster with 59 STs . Three STs ( e . g . ST238 , ST239 , and ST241 ) were grouped between the two major groups in the phylogenetic analysis . Both clusters were composed of isolates recovered from five continents , although isolates recovered from Europe and Asia were mainly found within the major cluster . All but the newly identified ST540 from Brazil had been previously reported from different regions of the world ( Fig 2 ) . The goeBurst analysis was applied to infer patterns of evolutionary descent among clusters of related genotypes and to identify groups within populations in the southeastern Brazilian as well as in the global datasets ( Fig 3 ) . The goeBurst analysis differentiated the southeastern Brazilian isolates into two main clusters and two more distant STs , ST71 and ST5 ( Fig 3A ) . The clonal complex ( CC ) CC32 and its descendants ST39 , ST93 , and ST31 , in addition to an additional three linked STs ( e . g . ST540 , ST15 , and ST77 ) were clustered together . These seven STs were separated by three alleles [e . g . triple locus variant ( TLV ) ] from the second group , which was composed of ST289 , ST63 , ST23 , and ST2 . The results did not show any pattern of differentiation based on their source ( e . g . clinical or environmental ) , although the group that contained ST63 and its descendants was mainly represented among clinical isolates ( Fig 3A ) . Expanding this analysis to the global dataset that contained isolates differentiated by continents , the two main clusters of the southeastern Brazilian isolates were also separated ( Fig 3B ) . The first major cluster composed of CC174 and its single locus variant ( SLV ) descendants ST3 , ST6 , ST175 , ST53 , and ST81 and another 10 CCs ( e . g . CC6 , CC4 , CC5 , CC212 , CC69 , CC63 , CC1 , CC2 , CC67 , CC61 , and CC23 ) is presented in a star-like shape , indicating the clonal distribution of C . neoformans var . grubii VNI . This major cluster is separated from the second minor cluster due to a difference in one allele ( e . g . SLV ) . The minor cluster is composed of CC31 , CC32 , CC93 and its descendants ( Fig 3B ) . Five STs ( ST241 , ST320 , ST322 , ST240 , and ST21 ) presented a difference in three alleles ( TLV ) in the overall minimum-spanning tree generated by goeBurst . There was no visual clustering among geographic distribution and population groups found in the goeBurst analysis , although isolates from Europe were more frequently clustered in the major group . To better understand the distribution of the genetic diversity , Principle Component Analysis ( PCA ) and hierarchical Analysis of Molecular Variance ( AMOVA ) were performed . The PCA showed that the southeastern Brazilian and global C . neoformans var . grubii datasets differentiate two major clusters previously identified in the goeBurst analysis ( Fig 4A and 4B ) . Using the global dataset , no support for differences between the clinical and environmental populations was found using PCA ( Fig 4C ) . AMOVA showed that the majority of variance components were found distributed within , rather than between populations ( Table 3 ) . In contrast , when dividing the populations according to continents , statistical support ( p = 0 . 002 ) was found that accounted for differences between populations ( Table 3 and Fig 4D ) . The pairwise FST tests were calculated among populations , and showed that Asia and Europe had distinct subpopulations , while Africa , North America , and South America did not have statistical support for differentiation ( see map of Fig 3B ) . The index of association ( IA ) and rBarD values of the different C . neoformans var . grubii VNI subpopulations were calculated using a clone corrected dataset in order to avoid bias of “high frequency” sequence types in the analysis . Both results strongly reject ( p<0 . 005 ) the null hypothesis of linkage equilibrium and free recombination in the African , European , and South American subpopulations , while this hypothesis was not rejected for the Asian and North American populations ( Table 4 ) . Next , we checked the presence of recombination using the PHI test and the minimal number of recombination events per gene was calculated within each population group and per gene in the global dataset . The PHI test did not show evidence of recombination for any of the populations studied using the single gene datasets . However , all but the North America population showed evidence for recombination using the concatenated dataset ( p value of PHI test: Africa: 0 . 000004 , Asia: 0 . 004 , Europe: 0 . 001 , North America: 0 . 107 , South America: 0 . 013 ) . In addition , one recombination event could be identified in the SOD1 gene in the African population and one in the GPD1 gene of the Asia population ( Table 4 ) . Taken together , the results from the global dataset showed that the European and South American isolates reproduced clonally . The Asian population showed linkage equilibrium and one event of recombination for the GPD1 gene while the African isolates present mainly a clonal reproduction with the highest number of recombination events that were uncovered in our dataset . The low numbers of STs available in the database from North America limited a detailed interpretation of these results . When the global C . neoformans var . grubii VNI population was analysed together , no significant evidence of linkage equilibrium was observed , and the intragenic PHI test showed no statistical significance for recombination events per gene , even though a few recombination events in the SOD1 and GPD1 genes in addition to the IGS1 region were detected ( Table 4 ) . In the concatenated ( intergenic ) dataset the PHI test showed evidence ( PHI , p<0 . 0001 ) of recombination ( S2 Fig ) . The two major clusters identified in the previous analyses were again differentiated by this analysis ( S2 Fig ) . Structure software was used to identify the number of populations ( K ) in our dataset and to check for the presence of migrant individuals . We first analysed the global dataset and the results showed three subpopulations with the highest explanatory power . Structure could differentiate both clusters ( e . g . major and minor ) in addition to one more subpopulation within the major cluster ( Fig 5A , S3A Fig ) . Using the southeastern Brazilian and global datasets , pre-defining the clinical and environmental populations , we also observed three subpopulations , without any differentiation ( FST test , p = 0 . 086 ) between them ( Fig 5B , S3B Fig ) . Furthermore , despite three subpopulations being identified when the isolates were divided by continents ( Fig 5C , S3C Fig ) , some continents were composed of more individuals belonging to one ( e . g . Asia , mainly represented by one subpopulation , coloured in red ) or another ( e . g . Europe , represented mainly by isolates coloured in green ) population . The remaining three continents contained isolates from all three subpopulations . Taken together , these data support that there are two main clusters with three subpopulations in C . neoformans var . grubii VNI , which can be found in both clinical and environmental isolates . Recombination events in some isolates could be observed as mosaics of multiple small chromosomal chunks depicted in the Structure analyses ( Fig 5 ) . Finally , coalescence analysis also clearly differentiated both clusters and showed that VNI isolates diverged from VNB ( ST7 ) around 0 . 58 to 4 . 0 million years ago [effective sample size ( ESS ) = 5 , 733] according to the best representative sample of the model used ( S1 Dataset ) . The time to the most recent common ancestor ( TMRCA ) of the two main VNI clusters was 0 . 29 to 2 . 8 million years ago ( ESS = 1 , 604 ) ( Fig 6 ) . The ancestral pattern based on single genes was calculated and showed that , for the majority of the MLST genes , the African alleles are situated on the top , as well as in a terminal position , of the network inferring its descendant profile ( S4 Fig ) .
Cryptococcus neoformans is by far the most common pathogen causing meningitis in HIV-infected patients around the world . In Brazil , there are around 734 , 000 ( 610 , 000–1 , 000 , 000 ) estimated people living with HIV and 16 , 000 ( 9 , 900 to 23 , 000 ) deaths just in 2014 according to UNAIDS ( http://unaids . org . br/ ) . Amongst developing countries , Brazil was one of the first to offer free of charge ART in 1996 through the public health system , which has contributed to the dramatic decline in mortality caused by AIDS ( http://unaids . org . br/ ) . The mortality attributed to cryptococcal meningitis has also decreased from as high as 90% before the ART era to 30–55% mortality at 10-week after its introduction [9 , 67] . Despite the broad free availability of ART , these numbers still remain high ( up to 60% in our study ) compared to countries with high GNP ( 6 , 5–32% ) [10 , 16–18] and require more investment from the public health system . Our data also reinforce the current clinical picture of cryptococcal meningitis in Brazil , which is mainly represented by adult males , in the median age of life , with late HIV diagnosis , severe immunosuppression , and disseminated fungal disease at admission for HIV treatment . This late HIV diagnosis is in contrast to freely available HIV diagnostic tests throughout the public health system in Brazil . This observation reflects the stigma that HIV infection has in the Brazilian society and represents one of the main challenges for public health policy-makers in Brazil . On the other hand , despite C . gattii infection has been described in children [68] , our results also confirm the rarity of cryptococcosis due to C . neoformans in pediatric cases [69 , 70] with the youngest patient presenting at age 18 . Few studies on C . neoformans in Brazil and other South American countries have linked clinical data with molecular epidemiological data . Furthermore , most of the studies to date conducted to better understand the genetic population structure of C . neoformans in these countries have only been performed using a range of PCR-based techniques using anonymous genetic markers [39 , 40 , 71–74] . As a consequence , little information about molecular epidemiology of this pathogen has been generated using sequencing-based methods . These methods aid in the integration of results into global databases , providing insights into the evolutionary ecology and molecular epidemiology of the pathogen . Since 2009 , the ISHAM consensus MLST scheme has been used in several parts of the world , especially in regions with high prevalence of HIV-infected patients , such as Southeast Asia [34 , 75 , 76] and Africa [35 , 36 , 77] , but also in different countries across Europe [10 , 78] , and North America [79] . Our results show a high prevalence of ST93 ( 91/143 , 63 . 6% ) for both clinical and environmental isolates and ST77 in half ( 19/35 , 54 . 2% ) of the environmental isolates ( S1 Table ) . ST93 has been mainly recovered from HIV-positive patients in several countries , but is more prevalent in Indonesia , India , and Uganda [34 , 77 , 80] . ST77 was mostly recovered from clinical samples in India , but also in few patients from France , Thailand , and Uganda [34 , 77 , 80] . The presence of both of these STs in clinical and environmental samples , in addition to the absence of the genetic differences between these two populations suggests that STs have not adapted specifically to environmental or clinical sources ( see , e . g . , Figs 3A , 4C and 5B ) . Interestingly , some of these “high frequency” STs were diploid , some of them co-infecting patients with haploid strains ( S2 Table ) . To rule out any misinterpretations of these results due to contamination with different colonies , two independent flow runs were performed using single colonies for all diploid isolates . In the C . neoformans / C . gattii species complex , diploid AD hybrids , which are the result of matting between C . neoformans strains of serotypes A and D [81] , occur more frequent than hybrids , which are the result of matting between the two sibling species C . neoformans and C . gattii [19 , 25–27] . Homozygous diploids ( AαAα or AaAa ) have also been found in clinical specimens from South Africa [82] where the percentage ( 9% ) is similar to this study ( 9 . 8% ) . Although , herein all isolates presented the diploid profile AαAα , as analysed by flow cytometry , and PCR screening for the mating type specific STE20 gene and serotype specific GPD1 and PAK1 genes . In the southeastern Brazilian region , dual infections were previously described in two AIDS patients , one by two RAPD subtypes of C . neoformans var . grubii VNI genotype , and the other by two isolates each from C . neoformans var . grubii VNI and C . gattii VGII in a reference center of Rio de Janeiro [83] . A previous study with 500 isolates from different geographic regions also showed that 8% of the isolates were diploid [84] . Two mechanisms have been proposed to explain the presence of diploid isolates in a population where the majority of the isolates are haploid . First , intra-varietal diploidization produced by fusion of two genetically distinct alpha cells . Second , autodiploidization generated when identical copies of the genome arise via either endoreplication or clonal mating [84 , 85] . Although we did not detect any mating type a among our isolates , we cannot rule out the possibility of its presence in the environment , which would keep the normal mating cycle of C . neoformans ( e . g . α-a sexual reproduction ) going . However , autodiploidization seems to be an alternative possibility , as has been previously suggested [84 , 86] . The presence of both haploids and diploids in a population and/or a shift in the ploidy has also been shown to increase fitness during stress conditions such as growth in the presence of different concentrations of azoles , nitrosative and oxidative stress [87 , 88] . This mechanism may be used by tetraploid titan cells , which play an important role in establishment of the pulmonary infection and subsequent dissemination to the CNS [87 , 89] . Surprisingly , all but one of the MLST STs described here were also present in different regions of the world . This result is highlighted by the goeBurst analysis , where all isolates from southeastern Brazil occur either in the same clonal complex or form a close network with STs from other countries and continents ( Fig 3B ) . However , differences in prevalence are also important to highlight . For example , ST5 , which was one of the most prevalent STs amongst the clinical isolates from Europe and Asia [34 , 90–92] , was only identified in two clinical and one environmental isolate from Brazil . Other STs seem to be very uncommon , for instance , ST289 was only found in one clinical sample in this study and in few occasions in Germany [90] . Since some STs are more prevalent in different regions , and to develop an overarching view of the C . neoformans var . grubii VNI global population , we separated the entire MLST dataset according to continents . This analysis clearly showed that the highest genetic diversity was observed in the African population and the lowest in the South American population ( Table 2 ) . The African isolates exhibited the greatest number of polymorphic sites , haplotypes , nucleotide diversity , as well as mutation rate . The high genetic diversity within the African population has already been previously described [36 , 37 , 75 , 93] , however in these studies , all three molecular types of C . neoformans var . grubii serotype A ( e . g . VNI , VNB , and VNII ) and only few or no isolates from South America were included . These results highlight that the high variability is not just only due to the African VNB genotype , but also when only the VNI subpopulation were considered . The low variability observed amongst South American isolates studied here could be due to the limited geographical scale of our study . However , as the first sequence based typing study of C . neoformans var . grubii VNI in South America , the current study was conducted on a wide variety of clinical and environmental isolates obtained in an infectious diseases centre of a university teaching hospital which receives patients from 27 cities in Southeast Brazil , increasing the variability sampled . To verify the suggested finding of a mainly clonal C . neoformans var . grubii VNI population responsible for the high mortality in HIV-positive patients in Brazil , a larger study including isolates from different regions of the country and from different countries in South America would be desirable . Brazil lacks systematic epidemiological studies on cryptococcosis and the diagnosis is markedly delayed . To overcome these obstacles , 32 multidisciplinary research groups from a variety of universities and medical centers in Brazil set up the Brazilian Cryptococcosis Network ( RCB ) in 2013 , which aims to study the epidemiological markers and clinical outcome of primary and opportunistic cryptococcosis cases , identifying the risk factors and differences in virulence of cryptococcal isolates ( Marcia Lazéra , personal communication ) . Differences in the global distribution of the genetic diversity suggest that ancient evolutionary events can be detected in the invasion-history of major molecular type VNI . Both the AMOVA and PCA analyses showed that the diversity was statistically significant when the populations were assessed according to continent . In this case , the pairwise FST tests showed Africa , North America , and South America had the same groups of STs , while Europe and Asia presented more differentiated subpopulations ( see , e . g . , map of Figs 3B and 5C ) . The African-origin hypothesis is supported by different studies [37 , 75] , in addition to the presence of the highest diversity and the presence of more mating type a isolates amongst African isolates . It is very probable that the global VNI population originated in Africa , from where the ancient human migration to Asia [75 , 93 , 94] and Europe [94] took place as observed in this study . The data presented here and the absence of a statistical difference between the FST values from Africa , South America , and North America suggests more recent multiples dispersal events from Africa to the Americas . The clonal characteristics of the VNI population in America in addition to the few evolutionary events observed in the coalescence analysis indicate a more recent evolution . Analyses using goeBurst and coalescence methods to determine the number of populations in the global dataset , further corroborated the existence of two main clusters . Structure analysis recovered these two clusters , and further divided the major cluster into two subpopulations . Three subpopulations within the C . neoformans var . grubii VNI major molecular type ( VNIa , VNIb , and VNIc ) , including variation in virulence , were recently described using genetic analysis in a South African population [35] . The current study did not find any correlation between the fatal outcome of the patients and the two major clusters analysed ( Table 1 ) . This lack of demonstrable variation in virulence is likely explained by the close genetic relationships of the C . neoformans subpopulations studied . The coalescence genealogy showed that the two main clusters diverged from the ancestral VNB around 0 . 58 to 4 . 0 million years ago . Arising from a subpopulation mainly restricted to Africa , the C . neoformans var . grubii VNI major molecular type likely split into two main groups around 0 . 29 to 2 . 8 million years ago . Results of the structure analysis showed that isolates from Africa , South America , and North America were present in both clusters ( Figs 4D and 5C ) , corroborating the lack of difference , which was found by FST . In contrast , isolates from Europe were comprised of one subpopulation of the major cluster while those from Asia were comprised of a different subpopulation . As evolution is a continuous process , it is possible that these two clusters within the C . neoformans var . grubii VNI would gain specific characteristics as time passed . Thus , further characterization of these subpopulations within the VNI genotype using more discriminatory techniques such as whole genome sequencing will likely add insights into different virulence traits and clinical outcomes , as has already been demonstrated for C . neoformans var . grubii VNB and for the Vancouver Island genotypes of the sibling species C . gattii VGII [23 , 24 , 35] . The persistence of widespread clones ( e . g . ST93 in Southeast Brazil; ST5 in Asia ) that are stable in space and time may follow the features of clonal evolution . Clonal evolution in microbes has been defined as a result of the absence or restriction of genetic recombination due to two main manifestations: ( i ) strong linkage disequilibrium and ( ii ) widespread genetic clustering [95] . Our results fit the two manifestations of this concept as follows: First , the majority of populations and the clone corrected dataset showed statistically significant results for tests of non-random association of alleles at different loci ( IA and rBarD ) , demonstrating an overwhelmingly clonal population structure ( Table 4 ) . Second , a star-like shape distribution or clustering of the isolates was clearly observed using the allelic profile in the goeBurst analysis ( Fig 3B ) , in addition to the convergent results using the phylogenetic , coalescence , PCA , and structure analyses . Furthermore , only a few recombination events were found for SOD1 in Africa and GPD1 in Asia , and three of the seven MLST loci in the overall C . neoformans var . grubii VNI population . The few recombination events that we found are likely to be too rare to break the clonal population structure of C . neoformans var . grubii VNI . One main theoretical consequence of our observation of a widespread clonality is the accumulation of deleterious mutations in the genome , known as Muller's ratchet . This phenomenon would be expected in C . neoformans var . grubii VNI as the majority of the population is of one mating type . However , unisexual reproduction can restore mutant strains of C . neoformans to wild-type genotype and phenotype , including prototrophy and growth rate , thus reverting Muller's ratchet [96] . Overall the herein presented study highlights a clonal population structure of the C . neoformans var . grubii VNI major molecular type in clinical and environmental isolates from southeastern Brazil using the ISHAM consensus MLST scheme . The southeastern Brazilian isolates revealed a highly clonal population structure and were less variable than populations from other continents , despite representing a limited geographic sampling . The finding that ST93 was recovered from the majority of the patients in this study , which was also associated with high mortality in Uganda [77] , would suggest that this genotype may be associated with a higher mortality . From a clinical point of view , despite free ART in Brazil , late HIV diagnosis and disseminated cryptococcosis at admission for HIV treatment are still the major problems , which need to be overcome in order to reduce the high mortality of cryptococcosis .
|
The members of the Cryptococcus neoformans / Cryptococcus gattii species complex are the cause of cryptococcosis , a life-threatening human disease responsible for 624 , 000 deaths annually . Infection is acquired through inhalation of dehydrated yeast cells from environmental sources . After reaching the lungs , the fungus disseminates to the central nervous system causing meningoencephalitis . The majority of meningitis cases in HIV-infected patients are caused by C . neoformans , a species well studied in regions with a high prevalence of HIV infection , such as Asia and Africa . A similar high prevalence has been reported from Brazil however the epidemiology of these infections is less well understood . We studied clinical and environmental isolates from the southeast region of Brazil using MLST . The results that we obtained showed a clonal population structure of C . neoformans var . grubii VNI , with low variability when compared against populations from different continents . This lower variability is probably the result of multiple recent dispersal events from Africa to the Americas . The majority of clinical isolates were of one sequence type ( ST93 ) , which was also found in environmental samples . By expanding the analysis to isolates from around the globe , it was possible to identify two major groups among C . neoformans var . grubii VNI .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"cryptococcus",
"neoformans",
"medicine",
"and",
"health",
"sciences",
"cryptococcus",
"gattii",
"cryptococcus",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"population",
"genetics",
"geographical",
"locations",
"microbiology",
"fungi",
"phylogenetic",
"analysis",
"molecular",
"biology",
"techniques",
"population",
"biology",
"fungal",
"pathogens",
"africa",
"research",
"and",
"analysis",
"methods",
"europe",
"mycology",
"south",
"america",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"brazil",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"people",
"and",
"places",
"asia",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"organisms"
] |
2017
|
MLST-Based Population Genetic Analysis in a Global Context Reveals Clonality amongst Cryptococcus neoformans var. grubii VNI Isolates from HIV Patients in Southeastern Brazil
|
Invertebrates rely on innate immunity to respond to the entry of foreign microorganisms . One of the important innate immune responses in arthropods is the activation of prophenoloxidase ( proPO ) by a proteolytic cascade finalized by the proPO-activating enzyme ( ppA ) , which leads to melanization and the elimination of pathogens . Proteolytic cascades play a crucial role in innate immune reactions because they can be triggered more quickly than immune responses that require altered gene expression . Caspases are intracellular proteases involved in tightly regulated limited proteolysis of downstream processes and are also involved in inflammatory responses to infections for example by activation of interleukin 1ß . Here we show for the first time a link between caspase cleavage of proPO and release of this protein and the biological function of these fragments in response to bacterial infection in crayfish . Different fragments from the cleavage of proPO were studied to determine their roles in bacterial clearance and antimicrobial activity . These fragments include proPO-ppA , the N-terminal part of proPO cleaved by ppA , and proPO-casp1 and proPO-casp2 , the fragments from the N-terminus after cleavage by caspase-1 . The recombinant proteins corresponding to all three of these peptide fragments exhibited bacterial clearance activity in vivo , and proPO-ppA had antimicrobial activity , as evidenced by a drastic decrease in the number of Escherichia coli in vitro . The bacteria incubated with the proPO-ppA fragment were agglutinated and their cell morphology was altered . Our findings show an evolutionary conserved role for caspase cleavage in inflammation , and for the first time show a link between caspase induced inflammation and melanization . Further we give a more detailed understanding of how the proPO system is regulated in time and place and a role for the peptide generated by activation of proPO as well as for the peptides resulting from Caspase 1 proteolysis .
Melanization is the result of the oxidation of mono- and/or diphenols by a redox enzyme , often phenoloxidase , and it is an important reaction in most multicellular organisms , both animals and plants . Intruding microorganisms are frequently melanized in invertebrates , and during this process , low-molecular-weight phenolic substances are converted into polymeric melanin in a multi-step chain of reactions . The initiation steps of this reaction are catalyzed by the prophenoloxidase activating system ( proPO system ) , and other steps occur spontaneously . The proPO system is a proteolytic enzyme cascade and its primary function is to recognize minuscule amounts ( picograms per liter ) of cell wall products from microorganisms ( lipopolysaccharide ( LPS ) , peptidoglycan ( PGN ) and glucans ) and respond to the microorganism by activation of the system and the subsequent generation of immune factors . This cascade requires careful regulation to achieve spatial and temporal control to avoid dangerous side effects [1] . A number of regulatory factors are involved in controlling the activation of the proPO system , including proteinase inhibitors [2] and the melanization-inhibiting protein ( in insects , crayfish and shrimp ) [3] , [4] , [5] . The importance of melanization ( proPO system ) in controlling a number of specific host–pathogen encounters has been demonstrated over the past few years . One example of this is the bracovirus protein Egf1 . 0 , which inhibits the prophenoloxidase ( proPO ) -activating proteinase in the insect Manduca sexta [6] . Two other recent examples are found in the parasitoid wasp Leptopilina boulardi , which targets the Drosophila phenoloxidase cascade by producing a specific serpin inhibitor [7] , and in the bacterium Photorhabdus luminescens , which secretes a small organic molecule that acts as a negative regulator of PO activity [8] . In addition , a pathogenic Aeromonas hydrophila strain becomes highly virulent to crayfish when the PO transcript levels are experimentally reduced [9] . The proPO activation system , or melanization cascade , bears functional resemblance to the complement system , although the final reaction , melanization , is different . Intriguingly , recently we succeeded in showing that all of the steps in the proPO-cascade in Tenebrio molitor are shared with the proteinase cascade that leads to the activation of the Toll pathway for the production of antimicrobial peptides [10] . This shared cascade has been confirmed in several other insects [11] , [12] . In the present study , we found that caspases are very important for the rapid degradation of proPO , which prevents oxidation in places where it is not appropriate . Caspases , or cysteine-aspartic proteases , are a family of the cysteine proteases that are known for their function in apoptosis [13] . Some caspases are involved in the inflammatory system via the regulation of pro-inflammatory cytokines , and these caspases are necessary regulators of the unconventional secretion of leaderless proteins [14] , [15] . In humans , caspase-1 is not only required for the activation of pro-interleukin ( IL ) -1β and pro-IL-18 , but also functions as an activator of nuclear factor of the kappa-enhancer in B-cells ( NF-κB ) and p38 mitogen-activated protein kinase ( MAPK ) [16] . Interleukin-1β is produced as a cytosolic precursor and is dependent on caspase-1 cleavage for its activation and secretion [15] , [17] . The proPO is also produced as a leaderless protein , most likely in the cytosol , and is secreted by an unknown mechanism . We , therefore , searched the sequence for caspase-1 cleavage sites and found two in the middle of the Cu-binding region . Therefore , we asked whether caspase-1-like cleavage of proPO is involved in the regulation of PO activity . We also asked whether the caspase-cleaved fragments have biological functions and whether these fragments are involved in immune functions even in the absence of PO activity . We also studied whether the peptide fragments generated by the cleavage of proPO into active PO by the prophenoloxidase activating enzyme ( ppA ) , which gives a peptide of approximately 20 kD , might have some immune function during or immediately prior to melanization . Our studies provide new information about the function of caspase-1-like activity in freshwater crayfish , in which it acts as a negative regulator of the proPO system . For the first time , we provide results showing that the fragments resulting from caspase or ppA cleavage have important biological functions .
ProPO , the inactive form of PO , is present in crayfish hemocytes , especially in the granular cells ( GC ) . GCs are densely filled with granules , as indicated by their name . Upon activation by different environmental challenges such as microbes , exocytosis is induced , which causes the release of several proteins from the granules of the GCs and the release of proPO into the external milieu [18] . Immunostaining of proPO in GCs revealed that proPO is present in the cytoplasm but not in the granules , and not all GC cells express proPO ( Figures 1A–C ) . ProPO is cleaved extracellularly to produce active PO upon activation by ppA . However , the mechanism by which proPO is released is still unknown . In beet armyworm and Drosophila , prostaglandins and JNK can stimulate cell lysis and subsequent proPO release [19] , [20] . In mammal , there are many reports showing that inflammasomes and caspase-1 activation are involved in the secretion of proteins without signal peptides [21] , [22] . Thus , we asked whether caspase-1 plays a role in proPO release and/or regulation . To answer this question , the presence of caspase-1 in the crayfish was examined . The transcriptome analysis of the freshwater crayfish P . leniusculus ( unpublished data ) revealed the presence of a translated amino acid sequence that has 36% identity to Drosophila caspase interleukin-1 beta converting enzyme ( GenBank: NP524551 ) . Additionally , by using an antibody against human caspase-1 , two bands with sizes about 40 and 50 kDa were detected from crayfish hemocyte lysate ( Figure 1D ) . These bands are probably two isoforms of the crayfish procaspase-1 like proteins . In comparison , in human six different isoforms of caspase-1 have been found . The 50 kDa procaspase-1 like protein was also detected in crayfish plasma and the level of this protein in plasma was decreased 1 h after an injection of E . coli or A . hydrophila compared to the control ( 0 . 15 M NaCl ) ( Figure 1E ) . Notably , when the 50 kDa band decreased , a 20 kDa band appeared in the plasma ( Figure 1E ) . The 20 kDa band is similar in size to the p20 subunit of mammalian caspase-1 , which is the active subunit of this protein . The active caspase-1 could only be detected in the supernatant and not in human keratinocyte lysates [15] . This is probably because it is rapidly degraded or released to the outside of cells and is therefore not present in cell lysates . Another explanation may be that active caspase-1 has a very short half-life , and therefore , its concentration under physiological conditions is very low [22] , [23] . Caspase-1 activity was also found to be slightly increased in plasma at 1 h after injection with E . coli but no statistically significant difference could be observed ( Figure 1F ) and this activity could be decreased by incubation of the caspase-1 inhibitor , Z-YVAD-FMK . Our results suggest that caspase-1-like activity is present in crayfish and that this activity can be activated during infection . As mentioned above , the activation of caspase-1 in vertebrates is involved in the secretion of several proteins , such as IL-1 β , but to date , no such mechanism has been identified in invertebrates , although the secretion of leaderless proteins such as proPO [18] and Pl-β-thymosins [24] has been observed . Therefore , the amino acid sequence of proPO was analyzed to determine whether there is a potential caspase-1 cleavage site in the proPO sequence . Two caspase-1 cleavage sites were found , after amino acids 363 and 389 , which would give rise to two N-terminal proPO fragments with predicted sizes of approximately 42 kDa ( proPO-casp1 ) and 45 kDa ( proPO-casp2 ) , respectively ( Figure 2A ) . These two cleavage sites are located downstream of the cleavage site for ppA and would give rise to a small N-terminal fragment ( 20 kDa of proPO-ppA ) and a C-terminal active PO [25] . Therefore , cleavage as the result of caspase-1-like activity has the potential to reduce the PO activity and act as a negative regulator of the proPO system . To investigate whether any proPO-caspase fragment ( proPO-casp ) is released from hemocytes , plasma proteins from bacteria-infected crayfish were subjected to western blotting . The results in Figure 2B show that two caspase-cleaved proPO fragments were present in the plasma after bacterial infection and that the levels of both proPO-casps increased with time , whereas the plasma proPO level decreased . Moreover , the level of proPO-casp1 was higher than that of proPO-casp2 at all time points . This result suggests that proPO-casp2 may be degraded faster than proPO-casp1 in vivo . When non-virulent E . coli was injected , the released proPO was rapidly cleaved by a caspase , and high levels of the fragments were detected at 1 and 3 hours post injection . In contrast , the injection of a virulent A . hydrophila strain resulted in slower caspase cleavage , and a high level was reached after 3 hours ( Figure 2B ) . Interestingly , when the levels of the C-terminal fragments of proPO in the samples were analyzed , neither active PO , nor proPO-casps could be detected , and only inactive proPO was found in the plasma . This result suggests that the C-terminal fragments of proPO produced by caspase-1-like cleavage , as well as ppA cleavage , are degraded rapidly . Because cleavage by caspase-1 may reduce PO activity , the enzyme activity in the plasma was measured after bacterial infection ( Figure 2C ) . The PO activity decreased after E . coli infection , but there was no significant difference , and interestingly , the plasma PO activity was significantly higher at 1 h after A . hydrophila infection , at time at which the plasma proPO-casp levels were low , and the enzyme activity markedly decreased by 3 h , when the levels of these fragments were higher . Notably , all animals died 4–6 hours after A . hydrophila infection . Ca2+ has been reported to induce exocytosis in crayfish hemocytes [18] and regulates inflammasome activation and thus caspase-1 activation [17] , [26] . Therefore , we investigated the effect of Ca2+ on proPO-casp release in in vitro experiments with isolated GCs . As shown in Figure 3A , the release of proPO and both proPO-casp fragments was Ca2+ and time dependent . When the antibody against the C-terminus of proPO was used , proPO but not active PO or proPO-casps could be detected . Again , this result suggests that the C-terminal fragments of proPO are rapidly degraded . To confirm that the proPO-casp fragments are the result of a caspase-1-like cleavage , the effect of the caspase-1 inhibitors Z-YVAD-FMK or Ac-WEHD-FMK on the release of proPO-casp fragments was examined . The results presented in Figure 3B clearly show that the production of proPO-casps was markedly decreased in the presence of Z-YVAD ( Figure 3B ) or Z-WEHD-FMK ( Figure S1 ) . The amount of released proPO into the medium was also decreased in the presence of Z-YVAD-FMK , and a higher level of proPO in granular cell lysate was observed when the cells were incubated with caspase-1 inhibitors ( Figure S1 ) . Furthermore , dsRNA caspase-1 treatment of granular hemocytes caused a complete reduction of the caspase-1 like transcript ( Figure 3C ) , but no obvious reduction in protein level could be observed ( data not shown ) . However , when the cells were treated with Ca2+ for 3 h to induce release of caspase-1 at 65 h of dsRNA treatment , the caspase-1 knockdown cells fail to produce new procaspase-1 like protein after another 24 h culture in L-15 ( Figure 3D ) . The lower level of caspase-1 like protein resulted in a reduction of the levels of proPO-casp fragments both in cell lysate and medium . No change in total proPO protein could be observed after the RNAi treatment of granular cells ( Figure 3D ) . Because putative caspase-1 cleavage products were clearly detected outside GCs in vitro and in vivo , we decided to determine if these fragments possess biological functions . The N-terminal parts of proPO produced by cleavage at the putative caspase-1 cleavage site between Asp363 and Ala364 ( proPO-casp1 ) , and the fragment produced by cleavage between Asp389 and Asn390 ( proPO-casp2 ) were produced as recombinant proteins with estimated sizes of 43 and 47 kDa , respectively . In addition , the N-terminal peptide fragment of proPO generated by cleavage by ppA between Arg176 and Thr177 [25] was produced ( proPO-ppA ) . To determine whether any of the proPO fragments are involved in the immune system , the bacterial clearance activities of these proteins were assessed . The fragments were mixed with bacteria and then injected into crayfish . Bacterial number in hemolymph was examined at 40 min and 3 h post injection . After 40 min , the E . coli titer was already significantly decreased in the proPO-ppA , proPO-casp1 , and GFP treatment groups , whereas the injection of proPO-casp2 had no significant effect on the number of E . coli ( Figure 4A ) . Because GFP also caused reduction of bacterial number at this time point , this might be a general protein effect . However , at 3 h post injection , the numbers of E . coli in the proPO-ppA , proPO-casp1 , and proPO-casp2 injection groups were significantly lower than that in the non-protein injection groups and the GFP group ( Figure 4B ) . The antimicrobial activities of all three fragments were then tested in vitro to determine if the bacterial clearance was caused by the proteins themselves or if other components were involved in the clearance process . The titer of E . coli decreased significantly after proPO-ppA treatment compared with the non-protein treatment , whereas the other fragments had no significant effect and did not exhibit any antibacterial activity ( Figure 5A ) . When the treated bacteria were observed under the microscope , very strong agglutination was detected after treatment with the proPO-ppA peptide , whereas no signs of agglutination occurred with the proPO-casp fragments or GFP ( Figure 5B ) . The minimal agglutinating concentration for proPO-ppA was the lowest for E . coli and Staphylococcus aureus , and the proPO-ppA fragment appeared to have the ability to agglutinate all of the tested bacterial species ( Table 1 ) . When the bacteria were observed by SEM after 15 and 40 min of incubation , we could see that proPO-ppA disrupted the E . coli cell morphology , causing the cell walls to shrink . After 15 min of incubation , the E . coli treated with proPO-ppA started to show signs of cell wall disruption , and a longitudinal line was observed ( Figure 5C ) , in contrast to the GFP-treated bacteria ( Figure 5D ) . After 40 min of incubation , the E . coli treated with proPO-ppA clearly formed clumps ( Figure 5F ) , and the cells were flat ( Figures 5H and 5I ) . Then , a bacteria viability assay was performed to determine if the strong agglutination killed the bacteria . Fluorescence microscopy clearly revealed that the proPO-ppA fragment greatly decreased the cell viability compared with the control treatments as measured by the red staining of dead bacteria ( Figure 6 ) . A few agglutinated bacteria were stained with only SYTO9 and appeared green in the merged picture ( live cells ) .
The proPO system is an important innate immune response and is composed of a cascade of proteinases that terminates with the activation of the proenzyme proPO . After proteolytic cleavage , proPO becomes an active redox enzyme , PO , which forms melanin and other antimicrobial products in the non-catalytic pathway from quinone to melanin [1] , [27] , [28] . Because the product of PO is highly toxic , it is necessary to keep the proPO system under strict control to avoid deleterious effects of an activated proPO system , principally the redox enzyme PO . Several factors that can control this system have been described , including a multitude of proteinase inhibitors [1] , [28] , [29] . Moreover , if PO is generated , the melanization inhibition protein ( MIPs ) can inhibit melanin formation [4] , [5] . Another way to protect against the inappropriate activation of proPO is to keep this proenzyme and its activation cascade in separate subcellular compartments . Thus , all arthropod proPOs are produced as leaderless proteins and are presumably located in the cytoplasm , whereas the activation system ( proPO-AS ) is located in secretory granules in crustaceans . This arrangement is similar to that of IL-1β , which is formed as a precursor in the cytoplasm and is then released to the outside of the cell during or after activation . Because caspase-1 cleavage is necessary for this activation and release steps and because such cleavage has been shown to be of importance for the secretion of several leaderless proteins [5] , we looked for caspase-1 cleavage sites in proPO . We identified a new important regulator of the proPO system , caspase-1-like activity , which can efficiently cleave proPO at two cleavage sites and make the enzyme catalytically inactive . Moreover , we found that the N-terminal products of this cleavage have effects on bacterial clearance . There are no previous reports of inflammasomes in invertebrates , and NOD-like receptors ( NLRs ) , which are part of the vertebrate caspase-1-activating inflammasome complex , have not been found in invertebrate genomes except that of the sea urchin [17] . However , both vertebrates and invertebrates express several pattern recognition receptors , and there might be other still undiscovered inflammasome sensor molecules responsible for the activation of invertebrate caspase-1-like activity . Recently , the structure of a Drosophila apoptosome composed of the Apaf-1-like protein Dark was reported . After binding to Dark , the initiator caspase Dronc cleaves the caspase DrICE and initiates an intrinsic cell death pathway [30] . The NLR-inflammasome and different apoptosomes are all examples of the oligomerization of CARD domain proteins involved in caspase activation , and their roles in cell death and immune responses are only beginning to be understood . Our discovery may add the proPO system to the list of immune responses balanced by such caspase regulation . The activation of proPO by ppA occurs via proteolytic cleavage near the N-terminal , and a peptide of approximately 20 kDa is released during this process . Once activated , the PO activity must be strictly localized to where melanization is needed . In analogy with the complement system in vertebrates , we asked whether the cleaved activation peptide also has biological activities , as C3-cleaved peptides do . To study whether this peptide ( proPO-ppA ) and the fragments resulting from caspase cleavage are involved in bacterial clearance , we injected E . coli together with these different fragments separately into crayfish and measured the number of bacteria in the hemolymph after the injection . Interestingly , all three peptides had the ability to decrease the bacterial number in the hemolymph compared to the injection of a control protein . To further investigate the mechanisms of action of these three peptides , we incubated each peptide directly with E . coli to identify any putative antibacterial activity , and we observed clear antibacterial activity for the proPO-ppA peptide . The observation of antibacterial activity in vitro suggests that the observed antimicrobial activity was a result of this peptide itself , whereas the proPO-casp1 and proPO-casp2 seem to require other components in crayfish to promote bacterial clearance . After incubation with the proPO-ppA fragment , both Gram-negative and Gram-positive bacteria were found to be heavily agglutinated . Moreover , live/dead staining of proPO-ppA-incubated E . coli revealed that the agglutinated bacteria contained a high percentage of dead cells . The generation of a highly antibacterial peptide after cleavage is similar to the case of hemocyanin , for which proteolytic cleavage in the crayfish plasma produces astacidin 1 [31] . These findings correspond to the antimicrobial function of human eosinophil cationic protein ( ECP ) . Incubation with ECP can cause bacterial agglutination and decreased viability . One further example is the C-terminal region of human extracellular superoxide dismutase ( SOD ) , which also exhibits antimicrobial activity against Gram-negative and Gram-positive bacteria [32] , [33] . The SEM study showed that the antibacterial effect of proPO-ppA seemed to be on the bacterial cell wall . The cell wall appeared to shrink in the presence of proPO-ppA and then the E . coli cells were flattened . This antimicrobial activity is similar to the activity of funnel web spider venom on Shigella sp . [34] . We present new findings that may explain how the proPO system is regulated and how its activity is localized ( Figure 7 ) . After ppA cleavage , the small N-terminal proPO-ppA peptide causes the agglutination of bacteria at the site of infection , and PO activity then may localize melanization to these bacterial aggregates . We also provide evidence that the release of proPO , a leaderless protein , from cells may involve caspase-1-like activity , similar to that regulating IL-1βrelease . Furthermore , caspase-1-like cleavage of proPO inactivates the enzyme and generates two N-terminal fragments with bacterial clearing activity . These findings show that proPO is a multifunctional protein , with a phenoloxidase in the C-terminal region and an agglutinating and antimicrobial peptide in the N-terminal region , as well as N-terminal proPO-casps peptides with distinct biological activities .
Freshwater crayfish ( P . leniusculus ) was purchased and reared in a closed system at 10°C . Only healthy animals were used for the experiments . The antibody against human procaspase-1 and p20 subunit was purchased from Invitrogen . The antibody against kazal protease inhibitor ( KPI ) was from Santa Cruz Biotechnology ( sc-46652 ) . The ECL peroxidase-linked donkey anti-rabbit IgG ( species-specific whole Ab ) was purchased from GE Healthcare . The peroxidase-linked anti-goat IgG antibody ( whole molecule ) was purchased from Sigma . The FITC-conjugated goat anti-rabbit IgG ( whole molecule , affinity isolated antigen-specific antibody ) was purchased from Sigma . To produce antibodies against the N-terminal and C-terminal peptides of proPO , proPO-N 1–76 and proPO-C 621–694 were cloned into the bacterial expression vector pGEX-4T-1 ( GE Healthcare ) . Then , these plasmids were subsequently transformed into Escherichia coli cells ( BL21 ) , and a single colony was grown in LB medium containing 100 µg/ml ampicillin to OD600 = 0 . 5 and induced with 0 . 2 mM isopropyl β-d-thiogalactoside ( IPTG ) for 6 h at 20°C . Recombinant GST-fusion proteins were purified using GSTrap FF columns ( GE Healthcare ) , and the GST tag was removed on the column by incubation with thrombin ( GE Healthcare ) at 4°C overnight . Then , the free recombinant peptides were eluted with PBS from the column . Two milligrams of recombinant proPO-N or proPO-C was used for the production of rabbit antiserum . The anti-proPO-N and anti-proPO-C antibodies were purified from the rabbit antiserum using GammaBind G-Sepharose ( GE Healthcare ) following the manufacturer's instructions . GCs were separated using a 70% Percoll gradient in 0 . 15 M NaCl . The separated cells were resuspended in 0 . 15 M NaCl , seeded on coverslips , fixed and treated as described previously [35] . The immunostaining was performed using an antibody against the N-terminus of proPO ( 5 ng/µl , 1∶160 ) and a FITC-conjugated anti-rabbit antibody ( 1∶300 ) . In addition , an antibody against a crayfish kazal protease inhibitor ( KPI ) was also used to counter stain hemocyte granules . The slides were mounted with VECTASHIELD mounting medium containing DAPI ( Vector Laboratories ) . The stained cells were then observed under a fluorescence microscope . The hemolymph was centrifuged at 1000× g for 5 min at 4°C , and the hemocyte pellet was collected and washed two times with PBS . Then , hemocytes were lysed in PBS containing 2% Triton X-100 [15] and 1× protease inhibitor cocktail ( Complete , Mini , EDTA-free , Roche ) . The cell lysate was centrifuged at 15 , 000× g for 15 min at 4°C , and the supernatant was collected . The protein concentration was determined , and 20 µg of protein was mixed with Laemmli sample buffer ( 62 . 5 mM Tris-HCl , 2% SDS , 10% ( v/v ) glycerol , 0 . 1 M DTT , 0 . 01% bromophenol blue , pH 6 . 8 ) . The presence of caspase-1-like protein was first examined in crayfish hemocyte lysate ( 20 µg protein ) by western blotting using an antibody against human caspase-1 . The hemocyte lysate was prepared as described above . In addition , the amount of the caspase-1-like protein in crayfish plasma was also determined in bacterial injected crayfish since bacterial components have been reported to be activators of inflammasomes and caspase-1 [36] , [37] . To perform this experiment , crayfish ( N = 3 for each group ) were injected with 0 . 15 M NaCl , non-virulent E . coli or highly virulent A . hydrophila B1 [38] , and hemolymph was collected at 1 h after injection . Hemocytes were removed from the hemolymph by centrifugation at 1000× g for 5 min at 4°C . Then , 200 µg of plasma protein was loaded onto 12 . 5% SDS-PAGE gels and subjected to western blotting as previously described [35] . The presence of caspase was detected using a rabbit anti-caspase antibody ( 1∶3000 ) and ECL peroxidase-linked donkey anti-rabbit IgG ( GE Healthcare ) ( 1∶7500 ) . The detection of actin was also performed as a loading control using a goat anti-actin antibody ( 1∶5000 ) and an anti-goat secondary antibody ( 1∶5000 ) . To determine caspase-1 activity , Ac-YVAD-pNA ( Santa Cruz ) , a synthetic peptide and substrate for caspase-1 , was used . Hemolymph was collected at 1 h after injection of 0 . 15 M NaCl , E . coli or A . hydrophila B1 . Cell-free plasma samples were prepared as described above . Then 50 µl of plasma was mixed with 200 nM of Ac-YVAN-pNA in the presence or absence of the caspase-1 inhibitor , Z-YVAD-FMK ( 50 µM ) . The mixtures were incubated at 37°C for 1 . 5 h , and the absorbance was determined at 405 nM . The plasma without substrate was used as a negative control for each sample . The caspase-1 activity was reported as OD405/g plasma protein . To determine if there are any potential caspase-1 cleavage sites in P . leniusculus proPO , the amino acid sequence of proPO ( GenBank: CAA58471 ) was analyzed using the bioinformatics tool PeptideCutter ( http://web . expasy . org/peptide_cutter/ ) . E . coli and A . hydrophila B1 were used to induce PO activity in vivo . Bacteria ( 1–3×107 CFU/100 µl ) suspended in 0 . 15 M NaCl were injected into the crayfish ( N = 5 ) . The hemolymph was then collected before and 1 and 3 h after injection and centrifuged at 1000× g for 5 min to remove the hemocytes . Then , 30 µl of the cell-free plasma was used in a PO activity assay . The plasma was incubated for 30 min at room temperature ( RT ) with 20 µl of 3 mg/ml L-DOPA and 50 µl phosphate-buffered saline ( PBS ) . The PO activity was determined by monitoring the absorbance at 490 nm , and a reaction mixture without substrate served as the baseline . Cell-free plasma ( 250 µl ) was centrifuged at 110 , 000× g at 4°C for 1 . 5 h to remove hemocyanin and 180 µl of the supernatant was subjected to acetone precipitation . Then , 1 . 6 µg of protein from each sample was loaded onto an SDS-PAGE gel , and the proPO-casps were detected by western blotting as described above using antibodies against the N-terminus or C-terminus of proPO . Because proPO is highly expressed in GCs , separated GCs were used in these experiments . The GCs were resuspended in 0 . 15 M NaCl and seeded into 96-well plates . After attachment for 10 min , the cells were incubated at RT in 10 mM HEPES-0 . 2 M NaCl buffer ( pH 6 . 8 ) containing different concentrations of CaCl2 ( 0 , 1 , or 10 mM ) . Then , 50 µl of buffer was collected from each well after 30 and 60 min and subjected to TCA precipitation . The protein pellets were dissolved in the same volume of Laemmli sample buffer and analyzed by western blotting using antibodies against the N-terminus ( 1∶3000 ) or C-terminus ( 1∶3000 ) of proPO as described above . To investigate the effect of the caspase-1 inhibitors , Z-YVAD-FMK ( Tocris Bioscience ) and Ac-WEHD-FMK ( Santa Cruz ) on proPO-casp release , GCs were incubated with 75 µl of HEPES-NaCl containing the inhibitor ( 0 , 1 , 10 , or 50 µM ) for 30 min before the addition of 75 µl of HEPES-NaCl containing 2 mM CaCl2 . Then , the buffer was collected at 30 and 60 min , and the samples were prepared and analyzed as described above . To knockdown crayfish caspase-1 , double-stranded RNA ( dsRNA ) was synthesized with MEGAScript Kit ( Ambion ) and the template was amplified using the following primers ( T7 promoter sequence is in italic ) ; for dsCaspase-1 5′-TAATACGACTCACTATAGGGACCTGTGGACCGACCTAGTGC-3′ and 5′-TAATACGACTCACTATAGGGGTCGGGCCTTTAGTTGGACACC-3′ , dsGFP: 5′-TAATACGACTCACTATAGGGCGACGTAAACGGCCACAAGT-3′ and 5′-TAATACGACTCACTATAGGGTTCTTGTACAGCTCGTCCATGC-3′ . The purified dsRNAs ( 1 µg/well ) were then added to granular cells and maintained in 25% L-15 medium in 0 . 15 M NaCl at 16°C for 65 h ( 50% of medium was changed once at 48 h ) . The cells were then treated with 10 mM Ca2+ for 3 h to induce caspase-1 release . Next the cells were washed with 25% L-15 medium four times and then maintained in the medium for another 24 h to allow the cells to produce new caspase-1 . Later , the cells were treated with 10 mM Ca2+ again , and same volume of medium was collected from each well at 60 min after incubation . In addition the granular cell lysate was also prepared from the 60-min Ca2+ treated cells . The samples were subjected to SDS-PAGE and the proPO-casp fragments and caspase-1 like protein were detected by western blotting . ProPO-ppA was amplified from P . leniusculus hemocyte cDNA using the following primers: proPO32EcoRI-F , ( 5′ TTTTTTGAATTCCAGGTGACCCAGAAGTTGCTGAGGA 3′ ) and proPOppA-R , ( 5′ CGCCTCGAGCTACCTGTTCACTTCAACCTGCATGCTT 3′ ) . The PCR product was visualized by agarose gel electrophoresis , extracted from the gel and purified before being ligated into the pET32a expression vector between the EcoRI and XhoI restriction sites . The protein was expressed in E . coli BL21 ( DE3 ) pLysS cells . After the IPTG induction , the proteins , which were expressed in inclusion bodies , were refolded and purified with Ni-affinity chromatography . Further , proPO-casp1 and proPO-casp2 were amplified using proPO-F ( 5′ CATGCCATGGGCCATCATCATCATCATCATCAGGTGACCCAGAAGTTGCTGAGGA 3′ ) and proPO-casp-R1 ( 5′ CGCCTCGAGCTAATCTGCCTCAAACGCGTCTCCTAAG 3′ ) for proPO-casp1 and proPOcasp-R2 ( 5′ CGCCTCGAGCTAGTCGTGACAGAATGCCAGCAGCACA 3′ ) for proPO-casp2 . Both PCR products were ligated into the pET28b expression vector between the NcoI and XhoI restriction sites and expressed in the E . coli expression system as described above . Bacterial cells were disrupted by sonication , and the recombinant proteins were purified with Ni-affinity chromatography and dialyzed against 20 mM Tris-HCl , pH 8 . 0 , at 4°C . Bacterial and protein injections were performed as follows . Briefly , wild-type E . coli were harvested at the mid-log phase , washed six times with 150 mM NaCl at 1200×g for 5 min and resuspended in 150 mM NaCl at 1×109 CFU/ml . Bacterial suspensions ( 100 µl ) were mixed with 20 µg of recombinant protein or 20 mM Tris-HCl and injected into the crayfish at the base of a walking leg . At 40 min and 3 h after injection , hemolymph was collected , serially diluted with 150 mM NaCl and plated on LB agar . The plates were incubated at 37°C overnight . The number of bacterial colonies was counted , and the number of CFU per ml was calculated for each treatment . To investigate whether the bacterial clearance activity was a direct effect of the protein fragments , we performed an in vitro bacterial clearance assay as follows . E . coli were prepared as described above , and 100 µl of resuspended E . coli was mixed with 20 µg of recombinant protein . Then , the volume was adjusted to 1 ml with 150 mM NaCl . The mixtures were incubated for 1 hour at room temperature with mild agitation , serially diluted and plated onto LB agar to calculate the CFU per ml . The plates were observed under a microscope . The minimum protein concentration for bacterial agglutination was tested as previously described [39] . The bacteria used in the experiment were Staphylococcus aureus Cowan , Micrococcus luteus M III , E . coli D21 , A . hydrophila B1 , and Pseudomonas aeruginosa OT97 . Overnight cultures of bacteria were collected and washed three times in 150 mM NaCl . Each bacterial species was resuspended , and the optical density was adjusted to 2 . The recombinant proteins were twofold serially diluted , and 50 µl of each dilution was mixed with 50 µl of bacterial suspension and incubated at room temperature for 1 hour . E . coli and protein mixtures were prepared as described for the in vitro bacterial clearance assay . After 5 min of incubation , SYTO9 ( Invitrogen ) was added to a final concentration of 50 nM , and propidium iodide ( PI ) was added to a final concentration of 1 µg/ml . Then , the samples were visualized with a fluorescence microscope . E . coli at O . D . 0 . 5 ( 100 µl ) was incubated with 20 µg of the proPO-ppA peptide for 40 min at room temperature with mild agitation . After incubation , the bacteria were harvested and fixed with glutaraldehyde following standard procedures for SEM .
|
Melanization is an important reaction in most multicellular organisms , both animals and plants . The initiation steps of this reaction in invertebrates are catalyzed by the prophenoloxidase ( proPO ) activating system a proteolytic enzyme cascade , which primary function is to recognize cell wall products from microorganisms and respond by activation of the system and generation of immune effector molecules . This cascade requires careful regulation to achieve spatial and temporal control to avoid dangerous side effects . We here show that a Caspase1-like enzyme can inactivate proPO when ppA is not activating the proPO to avoid deleterious effects and further we show for the first time that the N-terminal peptide from ppA cleavage of proPO ( activation of proPO ) has an important biological function as also the Caspase1 cleaved fragments . Our results also show that Caspase 1-induced inflammatory response is evolutionarily conserved and is linked to melanization .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology",
"and",
"life",
"sciences",
"immunology",
"evolutionary",
"biology",
"evolutionary",
"immunology",
"immune",
"response",
"immune",
"system"
] |
2014
|
Caspase-1-Like Regulation of the proPO-System and Role of ppA and Caspase-1-Like Cleaved Peptides from proPO in Innate Immunity
|
Human African trypanosomiasis ( HAT ) or sleeping sickness leads to a complex neuropsychiatric syndrome with characteristic sleep alterations . Current division into a first , hemolymphatic stage and second , meningoencephalitic stage is primarily based on the detection of white blood cells and/or trypanosomes in the cerebrospinal fluid . The validity of this criterion is , however , debated , and novel laboratory biomarkers are under study . Objective clinical HAT evaluation and monitoring is therefore needed . Polysomnography has effectively documented sleep-wake disturbances during HAT , but could be difficult to apply as routine technology in field work . The non-invasive , cost-effective technique of actigraphy has been widely validated as a tool for the ambulatory evaluation of sleep disturbances . In this pilot study , actigraphy was applied to the clinical assessment of HAT patients . Actigraphy was recorded in patients infected by Trypanosoma brucei gambiense , and age- and sex-matched control subjects . Simultaneous nocturnal polysomnography was also performed in the patients . Nine patients , including one child , were analyzed at admission and two of them also during specific treatment . Parameters , analyzed with user-friendly software , included sleep time evaluated from rest-activity signals , rest-activity rhythm waveform and characteristics . The findings showed sleep-wake alterations of various degrees of severity , which in some patients did not parallel white blood cell counts in the cerebrospinal fluid . Actigraphic recording also showed improvement of the analyzed parameters after treatment initiation . Nocturnal polysomnography showed alterations of sleep time closely corresponding to those derived from actigraphy . The data indicate that actigraphy can be an interesting tool for HAT evaluation , providing valuable clinical information through simple technology , well suited also for long-term follow-up . Actigraphy could therefore objectively contribute to the clinical assessment of HAT patients . This method could be incorporated into a clinical scoring system adapted to HAT to be used in the evaluation of novel treatments and laboratory biomarkers .
Human African trypanosomiasis ( HAT ) , commonly known as sleeping sickness , is caused by subspecies of the protozoan parasite Trypanosoma brucei ( T . b . ) and transmitted by tsetse fly bites in sub-Saharan Africa . This disease , which mainly affects rural populations , is one of the most neglected tropical diseases , and is fatal if left untreated [1]–[3] . There are two forms of the disease: the West and Central African form caused by T . b . gambiense , which represents the vast majority of cases , and the East African form caused by T . b . rhodesiense . The disease evolves in two stages: hemolymphatic ( stage 1 ) and meningoencephalitic ( stage 2 ) , which require different treatments . The arsenical compound melarsoprol has been widely used for stage 2 disease , but this therapeutic approach has severe side-effects , including fatal complications [3] . The main criterion for disease staging relies principally on the detection of elevated white blood cell ( WBC ) number and/or trypanosomes in the cerebrospinal fluid ( CSF ) [4] . The validity of this criterion is , however , under debate [5]–[8] , information on its correlation with clinical disease severity is limited , and new laboratory biomarkers for disease staging are currently under evaluation [7] , [9]–[13] . Objective clinical methods , preferably non-invasive , are therefore needed not only for the examination of HAT patients but also for follow-up , correlation with stage biomarkers , evaluation of treatment results and assessment of clinical trials . HAT leads to a constellation of neurological and psychiatric alterations , with characteristic sleep disturbances [2] , [14]–[18] . In fact , in a prospective multinational study of a large cohort of HAT patients , sleeping disorder , subjectively reported by the patients mostly as insomnia , was found to be the leading clinical symptom of the disease [17] . Sleep alterations have been repeatedly proposed as main signs of central nervous system involvement in HAT [3] , [15] , [19] , although their relationship with HAT staging is not yet fully established . The technique of polysomnography ( PSG ) has documented in detail sleep disturbances in HAT [14]–[16] , showing that the disease is characterized by disruption of the sleep-wake cycle during 24 h , with nocturnal insomnia and daytime sleepiness , and thus drawing attention also to disturbances of daily biological rhythms . The structure of sleep , and especially the sequence of the two types of sleep , namely rapid eye movement ( REM ) sleep which should be normally preceded by non-REM ( NREM ) sleep , is also frequently altered in HAT , with the occurrence of sleep-onset REM ( SOREM ) episodes [15] in which REM sleep is preceded by wakefulness . These disturbances are improved or reversed by treatment , indicating that they are disease-related signs [15] , [20] . PSG has been proposed as a non-invasive technique also for the evaluation of children affected by HAT [19] . However , PSG recordings are cumbersome , relatively expensive and difficult to perform in resource-limited health centers , and may not , therefore , represent a routine procedure for HAT in the field . This has led us to the search for less expensive and more user-friendly non-invasive technologies to allow objective assessment of day/night disturbances during the disease . A method developed in the last years is actigraphy , based on movement recordings via battery-run activity sensors , the actigraphs , wrist-watch size devices mostly worn on the wrist . Collected data are then downloaded to a portable computer and analyzed to provide an estimate of alterations of rest-activity from which sleep-wake parameters are analyzed [21] , [22] . Actigraphy has been developed and validated for human circadian rhythm disorders and sleep disturbances . Although actigraphy does not reach the level of information on sleep and wake parameters obtained by PSG , this technique provides a non-invasive , reliable tool for the ambulatory assessment of sleep disorders and the effects of treatment designed to improve sleep [21] , [23]–[25] . Actigraphy is used in a variety of clinical conditions ( see , for example , [26]–[28] ) , and has been increasingly used also in children [21] , [29] . According to WHO guidelines , second stage HAT is defined by the detection in the CSF of >5 WBC/µl [4] , but different cut-off criteria are used in some African countries where HAT is prevalent . In particular , a parameter of >5 up to ≤20 WBC/µl CSF , defining an intermediate stage of disease , is also used before second stage drug treatment is initiated [8] , [17] , [30] , [31] . Since objective tests of HAT symptoms are urgently needed for clinical assessment of the patients and treatment follow-up under field conditions , we here undertook a pilot study based on actigraphic recordings of patients infected by T . b . gambiense in different stages of disease . The investigation was conducted in two phases . First , actigraphy was performed in Cameroonian patients before and after the initiation of treatment . These findings have been previously reported in part in abstract form [32] . The study was then pursued , in the Democratic Republic of Congo ( DRC ) , with actigraphic recordings of HAT patients at admission . Nocturnal PSG was recorded simultaneously to actigraphy . All the parameters derived from actigraphy have been analyzed with a user-friendly software which can be easily applied under field conditions . Altogether the findings point to actigraphy as a useful , non-invasive tool for an objective clinical assessment and monitoring of HAT .
Two patients ( Y1 and Y2 ) were investigated and treated in the Neurology Department of the Central Hospital of Yaoundé , Cameroon ( Table 1 ) . This is one of the countries known to be endemic for HAT , classified among those with less than 100 new cases per year [6] with 5 foci of the disease [33] , [34] . The patients were referred by the National Trypanosomiasis Control Programme of the Ministry of Public Health during a screening campaign for HAT after positive screening for antibodies against T . b . gambiense in the blood with the card agglutination test for trypanosomiasis ( CATT ) . Both patients presented with histories of various complaints , including nocturnal insomnia and diurnal somnolence . CSF analysis was done according to standard methods at the Centre Pasteur Laboratory in Yaoundé . On the basis of the CSF finding of 6 WBC/µl , the patients were diagnosed as intermediate stage HAT ( see above ) , and were treated with a 10-day course of pentamidine ( 4 mg/kg daily , administered through the deep intramuscular route ) . The drug was supplied free of charge through the World Health Organization system . Actigraphic recordings were performed at admission after the initiation of treatment . The patients kept a diary of daily activities throughout the investigation . Nocturnal PSG was recorded in patient Y2 simultaneously with actigraphy before and after treatment . Actigraphy of healthy , age- and sex-matched control subjects ( CY1 and CY2 ) was recorded in parallel with the first actigraphic recordings of patients Y1 and Y2 . Seven patients ( K1–K7; Table 1 ) were investigated in DRC , which is classified to be among the countries with more than 1000 HAT cases per year [6] , [34] and with considerable underreporting [35] . All the patients had just come to observation through the National Sleeping Sickness Control Programme ( NSSCP ) of the Ministry of Health and had not yet received specific treatment . Also these patients were found to be infected by T . b . gambiense with CATT screening and disease stages were defined on the basis of CSF analysis as above ( Table 1 ) . Nocturnal PSG recordings were performed simultaneously with actigraphy , upon arrival , in the “Institut National de la Recherche Biomédicale” ( INRB ) in Kinshasa . The patients were then transferred to a hospital in Kinshasa ( “Centre Hospitalier Roi Baudouin I” or “Centre Neuro-Psycho-Pathologique” , CNPP ) , in collaboration and conformity with the procedures of the NSSCP for initiation of treatment . Actigraphic recordings were also done , in the same environments , on age- and sex-matched control subjects ( CK1–CK7 ) , who were CATT-negative . The patients had a thorough general and neurological examination . HIV serology was negative in all the patients and control subjects . For actigraphy , octagonal BASIC Motionlogger® actigraphs ( Ambulatory Monitoring , Inc . , New York , NY , USA ) were worn by the subjects on the non-dominant wrist . Throughout the recording period , the actigraphs were only removed when the subject was taking a shower to avoid damage by water to the device . For PSG , electrodes were placed to record the electroencephalogram ( EEG ) , electrooculogram , and electromyogram according to standard procedures ( see also [15] ) . The standard 10/20 system was used to place EEG electrodes on the scalp of the subjects . A computerized system equipped with Neuron Spectrum PSG software ( Neuronsoft® , Geneva , Switzerland ) was used to filter , record , and store the tracings obtained . In patient Y1 , 24 h actigraphic recordings were done upon admission and at day 55 after the initiation of treatment . In patient Y2 , 36 h actigraphic recordings were performed before treatment and 14 days after the initiation of treatment , and 12 h PSG recordings were performed simultaneously with actigraphic recording during the night . In patients K1–K7 , 36 h actigraphic recordings were performed at presentation , together with 12 h PSG recordings during the night . No difficulties were encountered during actigraphic recordings , whereas for PSG several challenges were faced at the Neurology Department of the Yaoundé Central Hospital as well as at INRB in Kinshasa , especially due to electrical power failures ( which also caused non-recording or loss of PSG data for patients Y1 and K7 ) . The study was conducted according to the principles expressed in the Declaration of Helsinki . All patients recruited received written and verbal information explaining the purpose of the study and gave informed consent . Ethical consent forms were designed in English and French in Cameroon and in French in DRC , and were also translated into local languages during administration . For the participation of the 5 year-old patient and matched control child , consent was given by the parents . The protocols received approval and ethical clearance by the Cameroon National Ethics Committee , and authorization by the Ministry of Public Health of Cameroon , as well as by the National Ethics Committee of the Democratic Republic of Congo and the Ministry of Health National Sleeping Sickness Control Program . All patients were hospitalized and cared free of charge in the Neurology Department of the Central Hospital Yaoundé for the Cameroonian patients and in the “Centre Hospitalier Roi Baudouin I” or CNPP for the Congolese patients . All hospitalization charges were paid by the research project funds . The raw data of activity over time ( actigram ) were displayed on a portable computer ( Figs . 1 and S1 ) . Analysis of the actigrams was performed with the Action 4 version 1 , a user-friendly software supplied by the manufacturer of the actigraphs ( Fig . S1 ) . Following the software steps ( Fig . S1 ) , the algorithm of Sadeh [22] , [36] was applied to the rest-activity ( RA ) signals to estimate the total sleep time during the night ( the daily suggestive period of rest: from 10 PM to 7 AM ) , and the day ( the daily suggestive period of activity: from 10 AM to 7 PM ) . In addition , using the same software ( and the steps illustrated in Fig . S1 ) , raw activity values were analyzed to obtain information on the 24 h RA rhythm using the cosinor rhythmometry method ( see supporting information S2 ) . The Action 4 software , as well as software provided by other actigraph manufacturers allow to perform such analyses in a few rapid steps which can be easily learned without extensive training . Cosinor rhythmometry analyses provide an F-ratio , which reflects the degree of fragmentation of the RA signal: the more the tracing is fragmented , the lower is the associated F-ratio . Data were also obtained on the characteristics of the RA rhythm and , in particular , the rhythm-adjusted mean ( MESOR ) , amplitude , and peak activity time or acrophase [37] , [38] ( see supporting information S2 ) . The PSG tracings were scored and analyzed using the Neuron Spectrum PSG software mentioned above , and the total sleep time during the night ( from 10 PM to 7 AM ) was determined . Statistical evaluation was conducted on data derived from the recordings performed in DRC , for which groups of subjects were available . The sleep total time during day or night , respectively , was evaluated with the Student's unpaired t test in the adult patients K2–K7 versus the matched control group of adult subjects ( CK2–CK7 ) . P values lower than 0 . 05 were considered significant .
On observation of 24 h actigrams of patient Y1 before treatment , bouts of high activity were evident during the night , indicating disturbed sleep . The actigram of patient Y2 showed more marked alterations , which consisted not only of bouts of high activity during the night but also of a fragmented activity during the day , frequently interrupted by episodes of rest indicative of episodes of diurnal somnolence ( Fig . 1B ) , as compared to the control ( Fig . 1A ) . Quantitative analyses of the actigrams recorded in these patients before treatment showed that in patient Y1 the total time spent asleep corresponded to 20% of the day ( versus 2% in the matched control subject ) and 59% of the night ( versus 81% in the control ) ( Fig . 2A , B ) . In patient Y2 the time spent asleep corresponded to 44% of the day and 60% of the night ( versus 1 . 8% and 96 . 3% , respectively , in the matched control subject ) ( Fig . 2C , D ) . The alterations revealed by the actigraphic recordings showed an improvement after the initiation of treatment . In patient Y1 the number and amplitude of bouts of activity during the night decreased with respect to the pre-treatment recording . Analysis of the actigram at 55 days after treatment showed that the total time spent asleep had decreased to 12% of the day and increased to 73% of the night ( Fig . 2A , B ) . In patient Y2 , the actigram pattern seemed to have improved especially during daytime at 14 days after treatment ( Fig . 1C ) , as confirmed by the analyses which showed a marked decrease of the time spent asleep during the day ( 23% ) and a slight increase of the sleep time at night ( 64% ) ( Fig . 2C , D ) . Nocturnal PSG recordings were also made in patient Y2 , so that the actigram and concomitant hypnogram could be compared ( Fig . 3 ) . The analysis of sleep time from PSG indicated that before treatment patient Y2 spent asleep 47% of the night , which increased to 58% at two weeks after treatment ( Fig . 2D ) . Such values indicated that in this patient actigraphy had overestimated the proportion of sleep during the night as compared to PSG , which , as it will be discussed further , is a possible limitation of this technique [22] . Nevertheless , both RA and PSG recordings showed a reduction of the patient's sleep time during the night with an improvement after treatment , though actigraphy underestimated the post-treatment improvement of this parameter . The nocturnal actigrams and hypnograms showed a good correspondence in patient Y2 , in particular revealing activity bursts and wakefulness episodes ( Fig . 3 ) . However , SOREM episodes observed in the hypnogram before and after treatment ( Fig . 3 ) were not shown by the actigrams . This is due to the fact that , as also discussed further , algorithms used to estimate sleep from RA consider short intervals of activity and are therefore not suited for the detection of brief , single epoch events [22] , such as SOREM events . Analysis of the RA signal and rhythm characteristics ( Fig . 4A , B ) provided further objective demonstration of disease-related signs in these patients . The waveform of the activity data ( fitting curve ) allowed the visualization of the daily curve of the rhythm ( Fig . 4A ) . This clearly showed an alteration in patients Y1 and Y2 at admission , which was less marked , especially in patient Y1 , after treatment ( Fig . 4C , D ) . The value of the F-ratio , which was decreased in both patients and especially in patient Y2 , also showed a post-treatment increase , indicating clinical improvement ( Fig . 4C , D ) . In particular , before treatment the F-ratio value was 228 in patient Y1 ( 446 in control subject CY1 ) , and 87 in patient Y2 ( 970 in control subject CY2 ) . During treatment this value increased to 293 in patient Y1 and 286 in patient Y2 . Evaluation of the rhythm characteristics ( acrophase , MESOR and amplitude ) showed different degrees of alteration in the patients and partial recovery after treatment ( Fig . 4C , D ) . As mentioned previously , 36 h actigraphy and simultaneous 12 h nocturnal PSG were performed in patients K1–K7 , including one child ( K1 ) , who were analyzed at admission ( Table 1 ) . The actigrams of the adult patients compared to the matched controls showed varying degrees of alterations ( Fig . 5 ) . In the 2 patients with very high WBC number in the CSF ( K6 and K7 ) , the actigrams revealed a complete disruption of the day/night cycle ( Fig . 5H , I ) . The analysis of sleep time showed that these patients had sleep episodes accounting for a considerable proportion of the day ( K6: 31 . 8% , K7: 33 . 7% of sleep time versus less than 1% in the respective controls ) , and a considerable decrease of the time spent asleep during the night ( K6: 57 . 2% , K7: 39 . 9% , versus about 80% in the respective controls ) ( Fig . 6A , C ) . The alterations were less marked in most of the other adult patients ( Figs . 5D–G , 6 ) . However , when considering the severity of alterations revealed by actigraphy and the WBC counts in the CSF , discrepancies were noted . The actigrams appeared altered in 2 patients with WBC counts corresponding to 5 and 6 cells , respectively ( Fig . 5D , F ) , while they were relatively preserved not only in another patient with a WBC count of 6 cells ( Fig . 5E ) , but also in a patient with WBC count of 27 cells ( Fig . 5G ) . Quantitative analyses of the proportion of sleep time in the adult patients K2–K4 revealed corresponding alterations especially during the night ( Fig . 6C ) , with a considerable increase of time spent asleep during the day in patient K4 ( 23 . 5% versus 0 . 6% in the control ) ( Fig . 6A ) . Analyses derived from simultaneous nocturnal PSG recordings in the adult patients K2–K6 showed a good correspondence with the RA signal in actigraphy , though the latter overestimated sleep time in one patient ( K2 ) ( Fig . 6C ) , as noted above for patient Y2 . The statistical analysis showed a significant difference in the average time spent asleep during day and night , respectively , in the group of adult HAT patients versus the control group , with significant increase of the total sleep time during the day and decrease during the night ( P<0 . 05 for both parameters ) ( Fig . 6B , D ) . In the 5-year old child infected by T . b . gambiense with a WBC count in the CSF of 3 cells , actigraphy revealed marked functional alterations compared with the healthy child of the same age and sex ( Fig . 7 ) . The actigram of this patient appeared disrupted ( Fig . 7A ) , as reflected by the quantitative evaluation of the sleep time , which corresponded to 37 . 8% of the day ( versus 0 . 76% in the control child ) ( Fig . 7B ) and 62 . 7% of the night ( versus 90% in the control ) ( Fig . 7C ) . The nocturnal PSG recording provided the same value of sleep time ( 62 . 4% ) ( Fig . 7C ) as that derived from actigraphy . SOREM episodes were observed in all DRC patients except in K6 ( Fig . S2 ) ( in K7 , as mentioned previously , PSG data were corrupted and therefore not scored ) . The analysis of the rhythm waveform showed in the patients different degrees of alterations of the daily pattern and rhythm parameters ( Fig . 8A–G ) . The statistical evaluation of the rhythm characteristics in the adult patients confirmed a significant mean decrease of the F-ratio with respect to the matched control group ( P<0 . 01 ) ( Fig . 8H ) . These analyses also showed a significant decrease of the mean values of the MESOR and amplitude ( P<0 . 05 for both parameters ) , with a preserved mean value of acrophase time , thus indicating that the time of peak activity was less affected than the other rhythm characteristics ( Fig . 8H ) .
Our pilot study shows that actigraphy during HAT can effectively reveal sleep-wake alterations characteristic of this disease and encourages further investigation . In particular , sleep-wake parameters , such as the night sleep disruption and the infiltration of episodes of sleep during the day , pointed out by previous PSG studies [15] , were well reflected by the actigrams . Furthermore , simultaneous actigraphic and PSG recordings during the night have shown in the present study a good correspondence of the two techniques in revealing sleep disturbances . The use of PSG has contributed to significant progress in the diagnosis of neurophysiological alterations during HAT [14]–[16] , [20] , and this technology remains the gold standard for the analyses of sleep-wake disorders . However , PSG may have serious limitations in HAT endemic settings . Compared to PSG , actigraphy requires much simpler and cost-effective equipment ( a dry-cell battery-run , a watch-size device and a laptop ) . The software ( 5 . 5 megabytes ) can run on a portable computer , and the files of the actigrams ( 21 kilobytes ) can be easily transferred through a cellular phone , or attached to e-mail messages even with slow internet connections , making very easy the acquisition and transfer of data for analyses . The battery-run actigraph can record and store data for several days without the risk of data loss due to electrical power failures ( which can frequently occur in field studies and did occur during our study ) . The possibility of an easy use of actigraphy in the field needs requires confirmation in future studies , but it should also be considered that can be done at home , and is thus well suited for long-term studies without interfering with the subject's routine activities [22] . In our investigation , the qualitative observation of actigraphic recordings showed in the cohort of 8 adult patients ( Y1 and Y2 , and K2–K7 ) varying degree of alterations . These ranged from well evident , high bouts of activity during the night , and therefore with considerable sleep fragmentation , to the complete disruption of the sleep-wake cycle . In patients Y1 and Y2 , actigraphy showed an improvement of the signal parameters after treatment initiation . These findings tally with previous studies based on PSG recordings [20] , which have in addition shown that complete recovery of sleep-wake parameters in HAT patients requires a very long time , and symptoms may persist for months after the end of treatment . A longitudinal long-term assessment is therefore required , actigraphy is especially suited for this purpose . The present observation in the child affected by HAT ( K1 ) is in accordance with previous evidence that actigraphy is valid for an evaluation of sleep-wake in children [29] , [36] . This finding is also in accordance with a recent report that sleep alterations reflect HAT severity in children [19] . The present preliminary evidence of the efficacy of actigraphy in a child affected by HAT is of particular interest especially considering the great difficulty in the evaluation of clinical signs ( such as inactivity and unusual behavior ) in children of 1–6 years of age suffering from this disease [17] . Although actigraphy is not a routine procedure in the evaluation of brain and/or systemic infections , data have been obtained with actigraphic recording in persons living with HIV infection [39] , [40] , given that sleep disturbance is a common complaint during this infection . These studies have been performed in HIV-infected women with CD4 cell counts between 40 and 930 mm3 , and using as exclusion criteria AIDS-dementia diagnosis , neuropathy or use of illicit drugs . The investigations have shown a moderate reduction of sleep time during the night with napping episodes during the day , providing an objective evaluation of the patients' complaints of insomnia and fatigue [39] , [40] . These alterations do not configurate , however , the fragmentation of the sleep pattern and sleep-wake cycle characteristic of HAT . Limitations of actigraphy were also evident in our pilot study . SOREM episodes , detected in the hypnograms of our HAT patients in agreement with previous findings [15] , could not be differentiated from normal rest in the actigraphic recording . Furthermore , the simultaneous actigraphic and nocturnal PSG recordings showed that in some of the adult cases the analysis derived from actigraphy tended to overestimate the total sleep time . As mentioned above , this limitation is inherent to the technique , due to the difficulty to distinguish sleep from episodes in which the subject wakes up but remains motionless [22] . However , this limitation was also present in the evaluation of sleep from actigraphy in control subjects and in multiple sessions of actigraphic recordings , and may not , therefore , influence the patients' clinical monitoring . It should also be noted that in the child affected by HAT the sleep time derived from actigraphy and nocturnal PSG showed the same values , probably due to the fact that quiet wakefulness events are unlikely to occur in children , thus further supporting the use of this tool for the clinical evaluation of children affected by HAT . An interesting issue is raised by the correlation between the signs of disease revealed by actigraphy and PSG on one hand , and the WBC counts in the CSF on the other hand . In agreement with previous investigation based on PSG [15] , [16] the patients with very high number of WBCs in the CSF showed severe sleep-wake changes ( here revealed by both actigraphy and PSG ) , while in the patients with lower numbers of WBCs ( 3–27 WBC/µl CSF in our study ) such correlation was not close . In addition , it has to be noted that SOREM episodes were detected in our study also in patients with 3–6 WBC/µl CSF , as shown in a previous study in which SOREM episodes were detected in 2 out of 4 patients with 0–7 WBC/µl CSF , although the episodes were more frequent in patients with >400 WBC/µl CSF [15] . Importantly , HAT brings about a severe neuroinflammatory condition . Cytokines and chemokines are at the core of HAT pathogenetic and clinical parameters , and inflammatory mediators have been implicated in sleep disturbances during the disease [18] , [41] , [42] . Novel experimental evidence has pointed out potential laboratory biomarkers of HAT severity and stages [7] , [9]–[13] . The correlation of actigraphic recordings and cytokine measurements has already been validated in other conditions which lead to increase of inflammatory biomarkers [43] , [44] , also in children [45] . Such correlation could thus represent a precious diagnostic and monitoring approach in HAT management . On the basis of the results here presented , and despite the above limitations , actigraphy appears as a tool well suited for objective measurements of disturbed sleep pattern and the daily distribution of sleep-wake in HAT patients , providing valuable neurophysiological information . Considering the current interest to establish scoring scales adapted to the clinical characteristics of the various types of nervous system diseases , it is foreseen that such scoring scales will be developed in the near future also for HAT to improve the assessment of new treatments . Objective data obtained by user-friendly actigraphy , standardized in a larger cohort of HAT patients , could therefore be suited to be incorporated into such novel scoring systems . In conclusion , on the basis of this first published report , actigraphy seems to be a very promising tool for obtaining objective clinical data in HAT , suited also for long-term assessment and follow-up of HAT patients . This technique could therefore also be useful for the clinical evaluation of relapses , assessment of novel treatments and correlation with disease and staging biomarkers in body fluids which are currently under investigation in many laboratories . This is of special relevance given that HAT affects populations living in environments with precarious health facilities .
|
The clinical picture of the parasitic disease human African trypanosomiasis ( HAT , also called sleeping sickness ) is dominated by sleep alterations . We here used actigraphy to evaluate patients affected by the Gambiense form of HAT . Actigraphy is based on the use of battery-run , wrist-worn devices similar to watches , widely used in middle-high income countries for ambulatory monitoring of sleep disturbances . This pilot study was motivated by the fact that the use of polysomnography , which is the gold standard technology for the evaluation of sleep disorders and has greatly contributed to the objective identification of signs of disease in HAT , faces tangible challenges in resource-limited countries where the disease is endemic . We here show that actigraphy provides objective data on the severity of sleep-wake disturbances that characterize HAT . This technique , which does not disturb the patient's routine activities and can be applied at home , could therefore represent an interesting , non-invasive tool for objective HAT clinical assessment and long-term monitoring under field conditions . The use of this method could provide an adjunct marker of HAT severity and for treatment follow-up , or be evaluated in combination with other disease biomarkers in body fluids that are currently under investigation in many laboratories .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"diagnostic",
"medicine",
"neglected",
"tropical",
"diseases"
] |
2012
|
Actigraphy in Human African Trypanosomiasis as a Tool for Objective Clinical Evaluation and Monitoring: A Pilot Study
|
Genetic analyses in Drosophila epithelia have suggested that the phenomenon of “cell competition” could participate in organ homeostasis . It has been speculated that competition between different cell populations within a growing organ might play a role as either tumor promoter or tumor suppressor , depending on the cellular context . The evolutionarily conserved Hippo ( Hpo ) signaling pathway regulates organ size and prevents hyperplastic disease from flies to humans by restricting the activity of the transcriptional cofactor Yorkie ( yki ) . Recent data indicate also that mutations in several Hpo pathway members provide cells with a competitive advantage by unknown mechanisms . Here we provide insight into the mechanism by which the Hpo pathway is linked to cell competition , by identifying dMyc as a target gene of the Hpo pathway , transcriptionally upregulated by the activity of Yki with different binding partners . We show that the cell-autonomous upregulation of dMyc is required for the supercompetitive behavior of Yki-expressing cells and Hpo pathway mutant cells , whereas the relative levels of dMyc between Hpo pathway mutant cells and wild-type neighboring cells are critical for determining whether cell competition promotes a tumor-suppressing or tumor-inducing behavior . All together , these data provide a paradigmatic example of cooperation between tumor suppressor genes and oncogenes in tumorigenesis and suggest a dual role for cell competition during tumor progression depending on the output of the genetic interactions occurring between confronted cells .
Growth regulation requires the fine tuning between the rate of cell death and cell proliferation in developing organs . Studies in Drosophila have revealed that somatic cells within a growing epithelium compete with one another for contribution to the adult organ and this phenomenon , known as “cell competition” [1] , is possibly conserved among organisms , for a review [2] . Cell competition was discovered several decades ago comparing the clonal growth parameters of Drosophila wild type cells ( +/+ ) and slow-dividing Minute/+ cells [1] . From those analyses and recent data [3] , it has been concluded that the contact between wild type and slow-growing cells , in genetic mosaics , favors the positive selection and clonal expansion of faster cells ( winners ) at the expense of slow-dividing ones ( losers ) , although eventually the final number of cells in the organs is unaffected [3] . The biological function of cell competition remains unclear but it is thought to contribute to tissue homeostasis by coordinating the rate of cell proliferation and cell death [4] , [5] . One of the best examples illustrating cell competition was obtained from the analysis of Drosophila myc [4] , [5] , opening to the speculation that this phenomenon might play a role in tumorigenesis [2] , [6] , however the basis of cell competition in tumorous situations has just begun to be investigated [7] . dmyc is an evolutionarily conserved proto-oncogene associated with different cellular processes , including cell cycle progression , cell growth and apoptosis [8]–[11] . The function of dMyc protein is both necessary and sufficient to control rRNA synthesis and ribosome biogenesis [12] . In Drosophila , cells carrying hypomorphic alleles of dmyc are viable in a homotypic context , but they are outcompeted and excluded from the epithelium when surrounded by wild type cells [5] . By contrast , dmyc overexpressing cells become “supercompetitors” able to kill wild type surrounding cells [4] , [5] . Remarkably , dMyc upregulation is related with many types of human cancers [13] and it favors the clonal expansion of cells carrying additional oncogenic mutations [14] , [15] . During the last years , the Hippo ( Hpo ) tumor suppressor pathway has emerged as a safeguard system restricting organ growth and preventing hyperplastic disease in metazoans [16] , [17] . Mutations in several members of this pathway have been associated with tumor formation both in Drosophila and in humans [18] . It has also been reported that mutations in many members of the Hpo pathway can rescue the viability of heterozygous M/+ cells in genetic mosaics [19] , suggesting that these mutant cells behave as “supercompetitors” . Therefore the detailed analysis of Hpo pathway members appears to be an attractive model in which to evaluate the relationship between cell competition and tumor growth , as well as the molecular mechanisms required for this crosstalk . Hpo , Salvador ( Sav ) and Warts ( Wts ) constitute the core of the Hpo pathway that regulates by phosphorylation the downstream transcriptional co-activator Yorkie ( Yki ) [18] , [20] . The hyperphosphorylated form of Yki is retained in the cytoplasm [21] , [22] , thereby preventing the expression of several target genes involved in cell proliferation control ( Cyclin E , E2F1 , bantam miRNA ) [16] , [23]–[25] , cell death ( dIAP1 ) [16] and cell signaling regulation ( dally and dally-like ) [26] . It has been demonstrated that Yki regulates its target genes by binding to Scalloped ( Sd ) , a TEAD/TEF family transcription factor [27]–[30] . In addition , recent data indicate that Yki is also able to bind to the homeoprotein Homothorax ( Hth ) forming a complex which regulates the transcription of bantam in the eye disc [31] . The atypical cadherins Fat ( Ft ) [26] , [32]–[37] and Dachsous ( Ds ) [20] , [26] , [33] , [38] , as well as the FERM-domain proteins Expanded ( Ex ) and Merlin ( Mer ) [39] , have also been implicated in the pathway as upstream components . Although their biochemical functions are still uncertain , it is assumed that they converge on Wts to regulate Yki activity [40] , [41] . Here we provide a detailed analysis of the autonomous and non-autonomous effects on growth of yki-expressing cells and mutations of members of the Hpo pathway . In addition we show that dmyc is a transcriptional target of Yki , able to confer competitive properties to the Hpo pathway mutant cells in the Drosophila wing . Furthermore , dmyc upregulation is essential to sustain the high rate of cell proliferation of Hpo mutant cells and to protect them from being eliminated in a competitive background . Finally , we show that the relative levels of dMyc protein between neighboring cells are critical in order to define the role of cell competition during tumor progression .
In order to analyze the competitive properties of Hpo pathway mutant cells , we used mosaic analysis to compare the size of yki overexpressing clones ( hereafter referred to as ykiover ) with their wild type twins . While clones and twins showed a comparable size in the wild type control ( Figure 1A , 1F , and 1G , and Figure S1A , S1C , S1D ) , ykiover clones were notably larger than their wild type twins in wing discs dissected either 60h ( Figure 1D , 1H , and 1I ) or 48h ( Figure S1B , S1E , S1F ) after heat-shock induction . Furthermore , ykiover wild type twins were almost disappeared from the epithelium at 120h after egg laying ( AEL ) ( Figure 1B and 1C ) . These differences in size were also prominent when discs were dissected at 96h AEL ( Figure 1D and 1E ) . Interestingly , the clonal expansion of ykiover cells was also correlated with non-autonomous apoptosis , as revealed by active Caspase 3 immunoreactivity of a subset of surrounding wild type cells ( Figure 1B–1E ) . The size advantage of ykiover clones and the induction of apoptosis in wild type cells is consistent with the broadly assumed definition of cell competition , which implies that the clonal expansion of the winner cells occurs at the expense of the juxtaposed losers , that are eliminated by apoptotic death [2] , [42] , [43] . The pattern of cell death in wild type and ykiover cells ( Figure 1B–1E ) was not confined to the interface between the two cell types; as can be seen in Figure 1E , cell death extends several cell diameters away and wild type cells tend to die massively when enclosed between nearby mutant clones ( Figure 1E , yellow arrowhead ) . A similar pattern of non-autonomous cell death was observed in wild type cells nearby mutant clones for other members of the Hpo pathway , such as ft and ex ( Figure S2A , S2B , S2C ) . Strikingly , ykiover clones and wts mutant clones grown for a longer period presented autonomous cell death ( Figure 1C , see active Caspase 3 staining , and Figure S2D ) , despite the upregulation of anti-apoptotic molecules such as dIAP1 [16]; this might be possibly due to either developmental constraints compensating for excessive proliferation of the entire organ or toxicity caused by high and constant levels of Yki . Altogether , these results confirm the previously suggested supercompetitive properties of the Hpo pathway mutant clones [19] by revealing their ability to overgrow and eliminate surrounding wild type cells . It is well documented that the confrontation of different levels of dMyc protein between two populations of cells either in vivo [4] , [5] or in cell culture [44] can trigger cell competition , however the molecular mechanism by which this occurs is unknown . In addition , myc family oncogenes are frequently overexpressed in human cancers and it contributes to tumor progression of YAP-expressing cells ( mammalian orthologue of yki ) [17] . We have previously shown that a transcriptional activation of dmyc occurs in ft mutant tissues and that ft clones fail to grow in a dmyc hypomorphic background [45] , indicating a possible regulation of this oncogene by the Hpo pathway . Moreover , the expression pattern of dMyc is complementary to that of Ds in the wing imaginal disc ( Figure 2A ) , suggesting a possible functional interaction . To validate this hypothesis , we analyzed dMyc expression in mutant clones for several members of the Hpo pathway and in ykiover cells by immunofluorescence . Noticeably , we found that dMyc was upregulated in a cell-autonomous manner in ykiover clones throughout the wing disc ( Figure 2B and Figure S3 ) , with the weakest activation in the lateral regions , and in a subset of clones mutant for several Hpo pathway members ( Figure 2C–2F ) . These differences in dMyc activation between ykiover clones and clones mutant for other members of the Hpo signaling pathway might be due to additional levels of regulation of the Hpo cascade operating on upstream members . According to our previous observations , we would predict a repression of dMyc upon Hpo pathway hyperactivation . To investigate this hypothesis , we expressed Hpo in the spalt expression domain of the developing wing disc . Since Hpo overexpressing cells die massively by apoptosis during development [25] , we coexpressed the anti-apoptotic factor p35 . As expected , cells coexpressing Hpo and p35 show reduced levels of dMyc with respect to the control ( Figure S4A ) in both late ( Figure S4B ) and early ( Figure S4C ) wing discs . Thus dMyc levels can be regulated by the Hpo pathway activity . dmyc was observed upregulated in RT-PCRs performed on ft mutant imaginal discs [45] , suggesting that it could be a transcriptional target of the Hpo pathway . In order to investigate this , we first performed an in situ hybridization in Drosophila wing discs expressing yki under the control of the decapentaplegic ( dpp ) promoter . As expected , dmyc transcript is detectable in the dpp domain both in yki and control dmyc-expressing discs ( Figure 3A ) . No signal within the dpp domain was detected in dpp>GFP control discs ( not shown ) . We were able to reproduce these data using a dmyc>lacZ line [46] which recapitulates accurately the dmyc pattern throughout the wing disc during development [7] , [47] . As can be seen in Figure 3B , the ßGal expression is increased in the dpp domain upon yki expression , indicating that Yki acts upon dmyc transcription . This result was supported using clonal analysis , both in ykiover cells , as shown in Figure 3C , and in cells mutant for ft ( Figure S5 ) . Altogether , these data demonstrate the ability of the Hpo pathway to regulate dmyc transcription in the imaginal wing disc . Yki transcriptional activity depends on the formation of tissue-specific complexes with different partners such as Scalloped and Homothorax [27]–[31] . In order to study the contribution of Sd to dmyc upregulation by Yki in the wing disc , we generated ykiover clones coexpressing either a UAS-sd or a UAS-sd-RNAi construct ( see Figure S6A for validation ) . As can be seen in Figure 3D , sdover; ykiover clones overgrew relative to ykiover clones ( compare with Figure 2B , 68% increase on average , n = 27 , P<0 , 005 ) confirming previous data [29] , but we were not able to detect significant differences in dMyc protein levels compared to ykiover clones ( n = 22 , P = 0 , 43 ) . As expected , control sdover clones did not overgrow and did not deregulate dMyc ( Figure 3E ) , demonstrating that Yki is required for dMyc upregulation . We were not able to recover sd-RNAi; ykiover clones in the wing pouch region , but clones generated in other territories of the wing disc , although large , did not upregulate dMyc ( Figure 3F ) , nor showed the same degree of hyperplasia as Yki expression alone ( Figure 1B–1E ) . sd-RNAi control clones were very small and did not deregulate dMyc ( not shown ) . These data indicate a key role for Sd in vivo in upregulating dMyc in ykiover clones , and in contributing to the ykiover tumorous phenotype . Interestingly , examination of dmyc locus revealed the existence of several CATTCCA repeats in non-coding regions of the gene , which perfectly match the mammalian [48] , [49] and Drosophila [28] , [29] TEAD/TEF family transcription factor consensus binding motifs ( mammaliam orthologues of Scalloped ) . In addition , these putative binding motifs for Yki/Sd complexes are evolutionarily conserved in D . simulans ( Figure 3G ) and relatively close to the insertion point of P elements that recapitulate the endogenous expression of the gene ( dmPL35 LacZ [50] , [51] and dmBG02383 Gal4 insertions - http://flybase . org/reports/FBti0018138 . html ) . To test the significance of these sequences in dmyc regulation , we generated a dmyc-firefly reporter containing the putative responsive elements for Yki/Sd complexes ( Figure 3H ) and performed a transient dual luciferase assay in S2 cells . As can be seen in Figure 3I , the reporter was specifically activated upon Sd and Yki cotransfection but , unexpectedly , the transfection of Yki alone was able to activate the reporter as efficiently as the cotransfection Yki/Sd ( Figure 3I ) . This result suggests that in presence of high levels of Yki alone , additional partners such as Hth [31] could bind it and co-regulate dmyc expression . Indeed , complementarily to Yki/Sd complexes , Yki/Hth complexes seemed to play the same role in the presumptive thoracic region of the wing disc . Supporting this conclusion , hth-RNAi; ykiover clones down-regulated dMyc in the notum ( 30% reduction on average , n = 15 , P<0 , 05 , Figure S6B , yellow arrows ) and did not grow as tumors in that region . By contrast , they were undistinguishable from ykiover clones in the wing pouch region ( Figure S6B , white arrowhead ) , where Hth expression is almost undetectable ( Figure S6C ) . Altogether , these latter results indicate that Sd and Hth play a role in Yki-induced tumorigenesis by regulating dmyc expression in the wing disc , with Sd playing a more critical role in the pouch and Hth acting in the presumptive thorax . With the aim to investigate the cell-autonomous contribution of dMyc overexpression to ykiover phenotypes , we first compared the size of ykiover clones with that of ykiover; dmyc-RNAi clones ( Figure 4 , see also Figure S7A , S7A′ , and [7] for RNAi construct validation ) . As expected , dmyc-RNAi clones showed a reduced number of cells with respect to that observed in wild type clones ( 21% reduction on average , compare Figure 4B and 4B′ with Figure 4A and 4A′ , P<0 . 01 ) . The reduction in cell number displayed by the ykiover; dmyc-RNAi clones with respect to the ykiover clones was even more evident ( 43% reduction on average , compare Figure 4D and 4D′ with Figure 4C and 4C′ , P<0 . 01 ) , and this percentage raised up to 65% ( n = 87 , P<0 , 001 ) when these clones were induced earlier in development ( 42–54h AEL ) , indicating a strong cell-autonomous requirement of dMyc protein for the expansion of ykiover clones . We also observed that the non-autonomous apoptosis induced by yki overexpression was reduced upon dmyc deprivation ( 32% on average , n = 28 , P<0 , 01 , Figure S7B ) . These data suggest that dMyc upregulation promotes cell proliferation of ykiover clones in an autonomous manner , and also promotes their competitive behavior . To further characterize this proliferation-promoting effect of dMyc , we compared the clonal behavior of various mutations in members of the Hpo pathway grown in two different genetic backgrounds: a wild type context and a genetic background overexpressing dmyc under the control of a hedgehog promoter in the posterior ( P ) compartment of the wing disc . We found that ft , ex and ds mutant clones were consistently larger in those territories expressing uniform levels of dMyc than in the wild-type background ( Figure 5 and Figure S8 ) . It is however described that the overexpression of dMyc is able to autonomously increase apoptosis [8]–[11] . In fact , the wild type tissue expressing high amounts of dMyc tends to die and does not overgrow ( see active Caspase 3 stainings in Figure 5A and 5D ) . Noticeably , the apoptosis mediated by dMyc overexpression seems to be extremely reduced inside ft and ex clones ( Figure 5A and 5D ) with respect to the wild type surrounding territories , likely due to the upregulation of antiapoptotic genes such as dIAP1 , a target of the Hpo pthway [20] . In addition , the dying cells in this genetic background might induce morphogens to promote compensatory proliferation [52] that may contribute to the extra-growth of ft- or ex-UAS-dmyc expressing clones . To circumvent this problem , we repeated the same experiment coexpressing dmyc and dIAP1 . As can be seen in Figure S9 , both ft ( Figure S9A ) and ex ( Figure S9D ) mutant clones grown in the P compartment were still consistently larger than those originated in the A compartment , thus confirming a specific cooperation of dmyc and Hpo pathway mutants in clonal expansion . Hpo mutant cells therefore seem to show the ability to take advantage of the cell mass accumulation boosted by dMyc overexpression to proliferate faster . To address the non-autonomous relevance of dmyc upregulation in providing ykiover cells with a supercompetitive behavior , we compared the size of ykiover clones generated in a wild type background to that of ykiover clones generated in a background ubiquitously overexpressing dmyc under the control of a tubulin ( tub ) promoter ( cell competition assay , [4] , [5] ) . In this assay cells express the endogenous dmyc gene plus an extra copy of the gene under the control of a tub promoter that ensures two-to-threefold increase of dmyc transcript [5] . This extra copy of dmyc is located in a removable cassette between the tub promoter and a Gal4 cDNA . Upon dmyc cassette excision , the tub promoter drives Gal4 expression in the clones and , as a result , those cells express lower levels of dmyc relative to the background and are rapidly eliminated from the tissue by cell competition . Only few genes have so far been found whose overexpression rescues cell viability in this context [5] . The relative difference in dMyc levels between yki-expressing cells and the surrounding tub>dmyc cells was minimized in a competitive background compared to a wild type context ( compare Figure S7C and S7C′ with Figure 2B ) . In this competitive background , ykiover clones showed a diminished ability to overgrow compared to a wild type background ( 44% reduction on average , compare Figure 6C and 6C′ and Figure 6B and 6B′; P<0 , 01 ) . Besides the reduction in size , ykiover clones showed an important reduction in clone number both in discs ( Figure 6C ) and adult wings ( compare Figure S7E to Figure S7D ) . Moreover , ykiover clones induced earlier in development ( 42–54h AEL ) were never recovered at the end of larval development ( not shown ) . These data indicate that the competitive properties of ykiover cells are extremely reduced when they are surrounded by cells expressing very high amounts of dMyc . We then performed the same competition assay as before while reducing dmyc activity inside the clones . We used the pupal lethal dmycPL35 allele [49] and , taking advantage of dmyc locus association to chromosome X , we were able to analyze both female ( heterozygous condition , the expression of dmyc is halved ) and male ( hemizygous condition , the expression of dmyc is completely removed ) larvae . In dmycPL35/+; tub>dmyc females , ykiover clones were smaller than those described in the previous assay ( 28% reduction on average , compare Figure 6D and 6D′ to Figure 6C and 6C′ , P<0 , 05 ) , whereas they were completely outcompeted by 48h after the heat shock in males ( not shown ) . Since it has been observed that a dmycPL35 heterozygous condition does not impair cell growth or proliferation rate [49] , our results reveal an important role for dmyc-induced cell competition in controlling the clonal expansion of ykiover cells , which may occur via their non-autonomous capabilities to compete with neighboring wild type cells . yki LOF clones generated in a wild type background are not able to grow [16] , [25] and the ectopic expression of the antiapoptotic proteins dIAP1 [25] or p35 ( Figure S10A ) poorly rescues their viability , whereas a Minute background [53] or bantam overexpression within yki clones has been shown to partially rescue their growth [25] . Since our results have indicated that dmyc participates in tumor growth of the Hpo pathway mutant cells , we therefore analyzed if the expression of dMyc was sufficient to prevent the death of yki mutant cells . The overexpression of dMyc failed to rescue the viability of yki−/− cells ( Figure S10B ) . Since yki mutant cells express low levels of the apoptosis inhibitor dIAP1 ( not shown ) , this result is not surprising , considering the autonomous cell death described for cells overexpressing dMyc [11] . However , yki mutant cells coexpressing dMyc and p35 also failed to grow ( Figure S10C ) . The lack of expression of additional antiapoptotic genes and cell cycle regulators [18] possibly impedes the clonal growth of yki mutant cells even though they overexpress dMyc . This result suggests that dmyc expression is able to enhance the ability of Hpo pathway mutant cells to grow , but it is not sufficient to rescue tissue growth of yki−/− clones .
dmyc upregulation has been demonstrated in many studies to provide cells with supercompetitive properties [4] , [5] , [7] . The model explaining how dMyc can confer competitive properties to cells is based on the relative levels of this protein in neighboring cell populations , transforming those cells expressing higher levels of dMyc into supercompetitors [4] , [5] . dmyc overexpression is nevertheless insufficient to drive tumorous growth; dmycover clones fail to overproliferate and show strong autonomous apoptosis [9] . Interestingly , we found that dMyc protein is overexpressed in Hpo pathway mutant clones , indicating an involvement for this cascade in dmyc regulation ( Figure 2 ) . Furthermore , the upregulation of dMyc in Yki-expressing cells correlates with an increase in the amount of mRNA , observed by in situ hybridization ( Figure 3A ) and using a dmyc>lacZ line ( Figure 3B and 3C ) . Finally , we have identified a regulatory region in the second intron of dmyc that is sensitive to Yki abundance; importantly , this regulatory region includes predicted consensus-binding motifs for Sd ( Figure 3H ) . Clonal experiments in the wing disc indicate that Sd is necessary for Yki function in vivo , since upon Sd downregulation Yki is no longer able to induce tumorous growth and does not upregulate dMyc ( Figure 3F ) . All these findings support the notion that there is a transcriptional regulation of dMyc mediated by Yki/Sd complexes in the wing pouch . Importantly , similar results were observed for dMyc regulation in the notum by Yki/Hth complexes , suggesting that tumor growth and dmyc regulation are tissue-specific . We found that dMyc upregulation is a common feature of Hpo pathway mutant cells . Since dmyc has been repeatedly associated with tumor progression and cell competition , we analyzed its role in the clonal expansion of Hpo pathway mutant cells . We observed that the reduction of dMyc expression restricts the ability of Hpo pathway mutant cells to proliferate ( Figure 4 ) , whereas its uniform overexpression strongly promotes their proliferation ( Figure 5 ) . Furhermore , while dMyc-expressing wild type cells surrounding mutant clones are rapidly eliminated by autonomous apoptosis , Hpo pathway mutant cells are able to take advantage of dMyc role in protein biosynthesis and cellular growth to divide rapidly . This is a clear example of functional cooperation between different genes in order to favor tumor progression , but it also indicates a specific role of dMyc in promoting the clonal expansion of Hpo pathway mutant cells . According to these data , we conclude that dMyc behaves as a growth-promoting factor which sustains the hyperplastic phenotype of Hpo pathway mutant cells . Importantly , this specific cooperation might be evolutionarily conserved , since c-myc appears to be upregulated in a murine model of YAP-induced carcinoma [17] . It has been suggested that cell competition may be a mechanism potentially restricting the clonal expansion of tumorous cells [7] , but it might also help faster proliferation of transformed cells . Our data indicate that Hpo pathway mutant cells are able to use high levels of dMyc to proliferate rapidly ( Figure 5 ) , but in a competitive context , where neighboring cells express high levels of dMyc , clonal expansion of ykiover cells is restrained ( Figure 6 ) , therefore suggesting a tumor suppressor role for cell competition . Conversely , dMyc upregulation in ykiover clones grown in a wild type background favors their clonal expansion promoting cell autonomous proliferation and also conferring the ability to outcompete sourrounding cells in a non-autonomous manner . These findings suggest that the phenomenon of cell competition may play a dual role in tumor progression depending on the output of the genetic interactions occurring between adjacent cells . In summary , we have shown a tumor-braking gene network in Drosophila epithelia which tightly controls cell proliferation , apoptosis and cell competition via the Hpo pathway and dMyc expression . Importantly , YAP deregulation has been reported in several types of human cancers [54]–[56] , therefore the mechanism of clonal expansion of Hpo pathway mutant cells in Drosophila might be relevant to understand tumor progression in mammals .
The fly strains used in the present work were obtained by the Bloomington Stock Center and are described at http://flybase . bio . indiana . edu . The following strains were instead obtained by: w; UAS-yki ( D Pan ) ; yw , tubFRTdmycFRTGal4 and yw , dmycPL35 , actFRTy+FRTGal4 ( P Gallant ) ; w , hs-FLP; actFRTy+FRTGal4 , UAS-GFP ( B Edgar ) ; w; FRT40A , dsD36 ( I Rodríguez ) . The UAS-RNAi constructs for dmyc , sd and hth were obtained from the VDRC . All experiments were carried out at 25°C unless otherwise indicated . MARCM UAS-yki twin-spot clones were induced at different stages of development by a 35-minutes heat shock at 37°C and larvae of the following genotype were dissected at either 84-100h AEL or 120h AEL: yw , hs-Flp , tub-Gal4 , UAS-GFP; FRT42D , tub-Gal80/FRT42D , Ubi-GFP; UAS-yki/+ . Clones of the same genotype were induced 54–66 h AEL and dissected 48h after a 20-minutes heat shock ( Figure S1 ) . For FRT-Flp twin analysis , the following hypomorphic or null alleles were used: dsD36 , ftG-rv , exE1 , wtsX1 , ykiB5 . Loss-of-function clones of ds , ft , ex and wts in either wild-type or mutant backgrounds overexpressing different transgenes in the posterior compartment were induced at 48–72h AEL by 1 hour heat shock at 37°C . Larvae of the following genotype were dissected at 120h AEL: yw , hs-Flp; FRT40A , Ubi-GFP/FRT40A , dsD36 or ftG-rv or exE1 yw , hs-Flp; FRT82B , Ubi-GFP/FRT82B , wtsX1 yw , hs-Flp; FRT40A , Ubi-GFP/FRT40A , dsD36 or ftG-rv or exE1; hh-Gal4/UAS-dmyc yw , hs-Flp; FRT40A , Ubi-GFP/FRT40A , ftG-rv or exE1; hh-Gal4/UAS-dmyc , UAS-dIAP1 The size of non-confluent clones was measured drawing each Z-stack of the confocal images using ImageJ software ( http://rsbweb . nih . gov/ij ) . Afterwards the area of the clones was normalized dividing by the area of the wing pouch , considered as the territory encircled by the first outer folding of the wing . In Figure S1 , the narrower window of clonal induction allowed us to compare clonal size without size normalization respect to the wing pouch . Statistical analysis was performed with Microsoft Excel and R ( www . r-project . org ) . Statistical significance was determined by two tailed Student's t test and reported as the associated probability value ( P ) . Flp-Out clones were induced at 60h AEL by a 8-minutes heat shock at 37°C; imaginal discs of the following genotype were dissected at 120h AEL: yw , hs-Flp; actFRTy+FRTGal4 , UAS-GFP yw , hs-Flp; UAS-dmycRNAi/+; actFRTy+FRTGal4 , UAS-GFP/+ yw , hs-Flp; actFRTy+FRTGal4 , UAS-GFP/UAS-yki yw , hs-Flp; UAS-dmycRNAi/+; actFRTy+FRTGal4 , UAS-GFP/UAS-yki . yw , hs-Flp/w , dmyc>lacZG0354; actFRTy+FRTGal4 , UAS-GFP/UAS-yki . Cell competition assays were performed at 72h AEL inducing a 40-minutes heat shock at 36°C . Larvae of the following genotype were dissected at 120h AEL: yw , tubFRTy+FRTGal4/hs-Flp; UAS-GFP/+ yw , tubFRTy+FRTGal4/hs-Flp; UAS-GFP/+; UAS-yki/+ yw , tubFRTdmycFRTGal4/hs-Flp; UAS-GFP/+; UAS-yki/+ yw , dmycPL35 , hs-Flp , tubFRTdmycFRTGal4/+-Y; UAS-GFP/+; UAS-yki/+ . MARCM yki clones overexpressing p35 , dMyc or both were generated at 48–72h AEL by a 45-minutes heat shock at 37°C and larvae were dissected 48h later . Immunostainings were performed using standard protocols . The following primary antibodies were used: mouse anti-dMyc ( 1∶5 , P Gallant ) , mouse anti-En ( 1∶50 , DSHB ) , rabbit anti-active Caspase 3 ( 1∶100 , Cell Signaling Technology ) , rabbit anti-p35 ( 1∶1000 , Stratagene ) , rabbit anti-Ds ( 1∶100 , D Strutt ) , rabbit anti-Hth ( 1∶400 , A Salzberg , [57] ) , mouse anti-dIAP1 ( 1∶100 , B Hay ) and rabbit anti-ßGal ( 1∶400 , F Graziani ) . Anti-mouse and anti-rabbit Alexa Fluor 555 ( 1∶200 ) ( Molecular Probes ) and anti-mouse Cy5 ( 1∶200 ) ( Jackson Laboratories ) against corresponding primary antibodies were used as secondary antibodies . Imaginal discs were mounted in Vectashield ( Vector Laboratories ) for confocal imaging . Single Z stacks were acquired with Leica SP2 and SP5 confocal microscopes . Images for Figure 4 and Figure 6 were captured with an epifluorescence Nikon 90i microscope . Entire images were elaborated with Photoshop CS2 ( Adobe ) and the projections along the Z axis were rebuilt starting from 35–55 Z stacks using the ImageJ public software ( NIH ) . For measurements of dMyc abundance , fluorescence intensity was calculated using the ImageJ public software ( NIH ) as the average gray value within selectioned portions of confocal Z stacks . For measurement of active Caspase 3 signal outside UAS-dmyc-RNAi; UAS-yki and UAS-yki clones , staged wing discs were chosen containing as few clones as possible and single cells positive to active Caspase 3 observed at a maximum distance of five nuclei ( counterstained with DAPI ) from the border of the clone were counted on confocal Z stacks . In situ hybridization was performed with a full length dmyc probe [9] on wing imaginal discs of L3 larvae expressing UAS-GFP , UAS-dmyc or UAS-yki under the control of dpp-Gal4 . RNA in situ hybridization was carried out using digoxigenin-labeled RNA probes [58] . Drosophila S2 cells were grown at 25°C in Schneider medium ( GIBCO ) supplemented with 10% heat-inactivated FCS and 100 units of penicillin . 1189 base pairs located in the second intron of the dmyc sequence ( Figure 3H ) were subcloned into a pGL3-firefly vector ( Promega ) and co-transfected with Sd and/or Yki-expressing pAc5 . 1/V5-HisB plasmids [28] using Effectene Qiagen Transfection Kit . The primers used for that purpose were: 5′ CAGCGGTACCAGTTTGCTGTCCTCTGC 3′ 5′GCACTCTAGAGCCATGCGGAATTGTGCG 3′ . The PCR product was first cloned in pCR 2 . 1 TOPO-TA ( Sigma ) and then subcloned in KpnI/XhoI sites of pGL3 Promoter vector . For luciferase transient expression assays , 2×104 cells were plated in 96-well dishes . Cells were harvested at 48 hours after transfection and luciferase activity was measured using the Dual-Luciferase reporter assay system ( Promega ) . Dual-Luciferase measurements were performed using a FLUOstar Optima luminometer ( BMG Labtech ) and normalized to the Renilla luciferase activity using pAct5C-seapansy as an internal control . All transient expression data reported in this paper represent the means from three parallel experiments , each performed in triplicate . Average relative luciferase activity was graphed and statistically analyzed by the Student's t-test .
|
One of the major challenges of developmental biology and cancer research is to get a better understanding of how different signals regulate proper organ growth and prevent tumor formation . Even though there is a strong correlation between tumor progression and Myc family misexpression or Hippo signaling pathway malfunction , the relationship between these organ growth regulators remains unclear . Here , we demonstrate that the Hippo signaling pathway controls the transcription of Drosophila dmyc . Furthermore , we show that the misregulated expression of dMyc in Hippo mutant cells elicits their proliferative expansion at the expense of normal surrounding cells . These findings reveal a molecular mechanism of cooperation between oncogenes and tumor suppressor genes that favors both tumor progression and wild-type tissue elimination . Additionally , our findings indicate a dual role for cell competition during the tumour progression depending on the cellular context .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"biology/cell",
"growth",
"and",
"division",
"genetics",
"and",
"genomics/cancer",
"genetics",
"developmental",
"biology",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses"
] |
2010
|
dMyc Functions Downstream of Yorkie to Promote the Supercompetitive Behavior of Hippo Pathway Mutant Cells
|
The Ypd1 phosphorelay protein is a central constituent of fungal two-component signal transduction pathways . Inhibition of Ypd1 in Saccharomyces cerevisiae and Cryptococcus neoformans is lethal due to the sustained activation of the ‘p38-related’ Hog1 stress-activated protein kinase ( SAPK ) . As two-component signalling proteins are not found in animals , Ypd1 is considered to be a prime antifungal target . However , a major fungal pathogen of humans , Candida albicans , can survive the concomitant sustained activation of Hog1 that occurs in cells lacking YPD1 . Here we show that the sustained activation of Hog1 upon Ypd1 loss is mediated through the Ssk1 response regulator . Moreover , we present evidence that C . albicans survives SAPK activation in the short-term , following Ypd1 loss , by triggering the induction of protein tyrosine phosphatase-encoding genes which prevent the accumulation of lethal levels of phosphorylated Hog1 . In addition , our studies reveal an unpredicted , reversible , mechanism that acts to substantially reduce the levels of phosphorylated Hog1 in ypd1Δ cells following long-term sustained SAPK activation . Indeed , over time , ypd1Δ cells become phenotypically indistinguishable from wild-type cells . Importantly , we also find that drug-induced down-regulation of YPD1 expression actually enhances the virulence of C . albicans in two distinct animal infection models . Investigating the underlying causes of this increased virulence , revealed that drug-mediated repression of YPD1 expression promotes hyphal growth both within murine kidneys , and following phagocytosis , thus increasing the efficacy by which C . albicans kills macrophages . Taken together , these findings challenge the targeting of Ypd1 proteins as a general antifungal strategy and reveal novel cellular adaptation mechanisms to sustained SAPK activation .
Candida albicans is the leading cause of systemic fungal infections in humans resulting in over 400 , 000 deaths each year in immuno-compromised patients [1] . The ability of C . albicans to adapt to host-imposed stresses encountered during infection is an important virulence trait [2] . Central to fungal stress responses are the stress-activated protein kinases ( SAPKs ) , which are conserved eukaryotic signalling enzymes that allow cells to adapt to environmental change [3 , 4] . In C . albicans , the Hog1 SAPK is activated in response to diverse , physiologically relevant , stress conditions , and cells lacking Hog1 are acutely sensitive to such stresses [5–7] . Consistent with the vital role of the Hog1 SAPK in stress survival , C . albicans cells lacking HOG1 display significantly attenuated virulence in systemic , commensal , and phagocyte infection models [8–11] . All SAPK activation mechanisms reported to date result in the phosphorylation of conserved threonine and tyrosine residues located within the TGY motif of the catalytic domain of the kinase [3] . Such pathways are tightly regulated as the nature of the response is dependent on the extent and period of SAPK activation . For example , in the model yeast Saccharomyces cerevisiae , transient activation of the Hog1 SAPK is vital to survive osmotic stress [4] , whereas sustained activation triggers programmed cell death [12] . Similarly in human cells transient activation of the p38 SAPK promotes stress-induced gene expression and cellular proliferation [13] , whereas sustained SAPK activation triggers apoptosis [14] . In contrast , much less is known regarding the regulation and cellular consequences of sustained SAPK activation in C . albicans . Despite the availability of several antifungal drugs , the high mortality rate associated with C . albicans systemic infections and the emergence of drug resistant strains highlights the urgent clinical need for new anti-fungal therapies [15] . Although Hog1 is an essential virulence determinant in C . albicans , the conservation with highly related SAPKs in human cells suggests that Hog1 itself may be unsuitable as an antifungal target . Instead , there has been much interest in identifying fungal-specific regulators of SAPKs as potential drug targets . Candidate targets include two-component related phosphorelay systems which constitute an important mechanism employed by fungi , but not mammals , to sense and relay specific stress signals to SAPK modules [16] . In S . cerevisiae , this system is comprised of a hybrid histidine kinase ( Sln1 ) , an intermediary phosphorelay protein ( Ypd1 ) , and a response regulator protein ( Ssk1 ) ( Fig 1 ) . Following osmotic stress , the Sln1 histidine kinase is inactivated , which halts phosphorelay through Ypd1 , and consequently leads to the rapid dephosphorylation of Ssk1 [17] . Dephosphorylated Ssk1 is a potent activator of the Ssk2/Ssk22 MAPKKKs which regulate Hog1 activation [18 , 19] . Significantly , loss of either Sln1 or Ypd1 function in S . cerevisiae is lethal [20] , due to the accumulation of unphosphorylated Ssk1 and the resulting sustained Hog1 activation which triggers apoptosis-mediated cell death [12] . It is likely that sustained Hog1 activation can also not be tolerated in the human fungal pathogen , Cryptococcus neoformans , as Ypd1 is essential for the viability of cells containing Hog1 [21] . The essential nature of Ypd1 in these fungi , and the many reports illustrating the importance of two-component proteins in fungal pathogenicity , has fuelled interest in targeting Ypd1 for antifungal drug development ( reviewed in [22] ) . C . albicans has seven two-component proteins; three histidine kinases ( Sln1 , Chk1 , Nik1 ) , three response regulators ( Ssk1 , Skn7 , Crr1/Srr1 ) , and a single phosphorelay protein ( Ypd1 ) [23] . In S . cerevisiae , Ypd1 plays a pivotal role in mediating all phosphorelay events from the upstream histidine kinases to the downstream Ssk1 and Skn7 response regulators [17 , 24] , which supports the concept that Ypd1 is an appropriate antifungal target . Hence , in this study , we investigated the impact of inactivating Ypd1 upon stress signalling and virulence of C . albicans . As reported recently , we found that C . albicans can survive deletion of YPD1 [25] . Here we extend this finding by illustrating that C . albicans survives the sustained SAPK activation following Ypd1 loss by evoking multiple mechanisms to reduce the level of phosphorylated Hog1 . Furthermore , we demonstrate that inactivation of Ypd1 during infection actually increases the virulence of C . albicans in a number of infection models , revealing that Ypd1 may not be a suitable target for anti-fungal drug development .
C . albicans contains a single homologue of the S . cerevisiae phosphorelay protein Ypd1 [26] . Although deletion of Ypd1 results in a lethal phenotype in both S . cerevisiae and C . neoformans [20 , 21] , a recent study revealed that YPD1 is not an essential gene in C . albicans , with ypd1Δ cells instead displaying a slow growth phenotype [25] . To investigate this further we created a C . albicans strain , tetO-YPD1 ( Fig 2A ) , in which one allele of YPD1 was deleted and the remaining allele placed under the control of a doxycycline-repressible promoter [27] . Northern analysis confirmed that treatment of tetO-YPD1 cells with doxycycline caused a rapid decrease in YPD1 mRNA levels ( Fig 2B ) . However , whilst repression of YPD1 expression did result in a slower growth rate ( Fig 2C , upper panel ) , the cells were viable . Furthermore , consistent with the previous study [25] , we were able to generate a viable homozygous ypd1Δ null mutant which displayed a slower growth rate compared to wild-type cells ( Fig 2C , lower panel ) . Deletion of YPD1 is lethal in S . cerevisiae due to constitutive SAPK activation . Consistent with previous findings [25] , we found that repression of YPD1 expression in tetO-YPD1 cells ( Fig 2D , upper panel; S1 Fig ) , or deletion of YPD1 ( Fig 2D , lower panel ) , also stimulated high levels of Hog1 phosphorylation in C . albicans . Together these data indicate that the inhibitory effect of YPD1 on SAPK activation in S . cerevisiae is conserved in C . albicans . However , it was possible that the phosphorylated Hog1 detected did not result in activation of Hog1-dependent downstream events . To test this possibility the expression levels of the Hog1-dependent genes , GPD2 and RHR2 [28] , important for glycerol biosynthesis were examined upon repression or deletion of YPD1 ( Fig 2E ) . Both genes were found to be up-regulated and , furthermore , as expected , increased intracellular glycerol concentrations were observed in ypd1Δ cells in the absence of stress ( Fig 2F ) . Thus collectively , these data confirm that the phosphorylated Hog1 kinase triggered by inactivation of Ypd1 in C . albicans is active . However , in contrast to S . cerevisiae and C . neoformans constitutive Hog1 activation does not result in loss of viability in C . albicans . Analyses of cells with loss of Ypd1 function , either by repression or deletion of the YPD1 gene , revealed identical morphological abnormalities and stress-resistance profiles . For example , loss of YPD1 resulted in swollen pseudohyphal-like cells ( Fig 3A ) , possibly due to the increased intracellular levels of the osmolyte glycerol that occurs upon inactivation of Ypd1 ( Fig 2F ) . In addition , as reported previously [25] , cells lacking Ypd1 were highly flocculent as demonstrated by their rapid sedimentation rate ( Fig 3A ) . Interestingly , cells lacking Ypd1 displayed acute sensitivity to sodium arsenite , increased resistance to the organic peroxide tert-butyl hydroperoxide ( t-BOOH ) , and wild-type levels of resistance to osmotic stress ( Fig 3B ) . Loss of YPD1 also resulted in increased sensitivity to the cell wall perturbing agent calcofluor white ( Fig 3B ) . This is consistent with Hog1 activation in ypd1Δ cells ( Fig 2D ) , as hog1Δ cells display significant resistance to this drug [6] . Loss of Ypd1 function is predicted to perturb phosphorelay to all three response regulator proteins in C . albicans; Ssk1 , Skn7 and Crr1/Srr1 [23 , 29 , 30] . In S . cerevisiae it is the accumulation of the unphosphorylated Ssk1 response regulator which triggers hyperactivation of the Hog1 SAPK [18] . Hence , to investigate which phenotypes associated with Ypd1 loss in C . albicans were due to Ssk1-mediated activation of Hog1 , hog1Δ ypd1Δ and ssk1Δ ypd1Δ double mutant strains were created . All of the ypd1Δ-associated phenotypes , described above , were found to be dependent on both Hog1 and Ssk1 . For example , the swollen pseudohyphal-like morphology associated with ypd1Δ mutant cells was repressed in the absence of either HOG1 or SSK1 ( Fig 3C ) . In fact , the hog1Δ ypd1Δ double mutant cells instead displayed the morphological defects characteristic of hog1Δ cells ( Fig 3C ) . The high glycerol levels characteristic of ypd1Δ cells were dependent on Hog1 ( Fig 3D ) , and the stress-phenotypes associated with deletion of YPD1 were not maintained in hog1Δ ypd1Δ or ssk1Δ ypd1Δ double mutant cells ( Fig 3E ) . Indeed , in all the conditions examined , cells lacking HOG1 and YPD1 displayed similar stress phenotypes as hog1Δ cells , and ssk1Δ ypd1Δ cells were phenotypically similar to ssk1Δ cells . Importantly , confirming the link between Ypd1 and Hog1 activity , reintegration of HOG1 or SSK1 into the hog1Δ ypd1Δ and ssk1Δ ypd1Δ mutants , respectively , resulted in cells that were phenotypically identical to the ypd1Δ strain ( Fig 3C and 3E ) . Furthermore , the hyper-phosphorylation of Hog1 detected in ypd1Δ cells was absent in ssk1Δ ypd1Δ cells , but was restored upon reintegration of SSK1 ( Fig 3F ) . This result confirms that Ssk1 is essential for the high basal level of Hog1 phosphorylation in ypd1Δ cells . Taken together , these results are consistent with the model that accumulation of unphosphorylated Ssk1 triggers the ypd1Δ-dependent sustained activation of Hog1 in C . albicans and , moreover , that this activation underlies the morphological and stress phenotypes associated with loss of Ypd1 . We next investigated the molecular mechanism underlying the ability of C . albicans to survive sustained Hog1 activation . In S . cerevisiae , the lethality associated with YPD1 loss can be by-passed by the artificial over-expression of either of the protein tyrosine phosphatases , Ptp2 or Ptp3 , which normally dephosphorylate and negatively regulate Hog1 activity [31 , 32] . Indeed , YPD1 ( tyrosine phosphatase dependent ) was initially identified in a synthetic lethal screen for S . cerevisiae mutants whose growth was dependent on the expression of PTP2 [33] . Similarly , Ptp2 in C . neoformans [34] and Ptp2 and Ptp3 in C . albicans [35] , have been reported to negatively regulate the respective Hog1 SAPK pathways in these pathogenic fungi . Hence , it was possible that the ability of C . albicans ypd1Δ cells to retain viability was linked to the activity of the Ptp2 and Ptp3 phosphatases . Strikingly , we found that the expression of PTP3 is significantly induced in ypd1Δ cells compared to wild-type cells , and PTP2 is also up-regulated albeit to a lesser extent ( Fig 4A ) . Similar findings were also observed upon doxycycline-mediated repression of YPD1 in tetO-YPD1 cells ( Fig 4B and 4C ) . Significant induction of PTP3 occurred with similar kinetics as the increase in Hog1 activation , and some induction of PTP2 was also evident ( Fig 4B and 4C ) . Previous studies revealed that arsenite is a potent inhibitor of protein tyrosine phosphatases that regulate SAPK pathways in both yeast and humans [36–38] . Interestingly , C . albicans cells lacking Ypd1 are acutely sensitive to arsenite ( Fig 3B ) , raising the possibility that arsenite-mediated inhibition of Ptp2 and/or Ptp3 causes catastrophic levels of Hog1 phosphorylation and ultimately cell death . Indeed , in agreement with this hypothesis , sodium arsenite treatment massively increased Hog1 phosphorylation in ypd1Δ but not wild-type cells ( Fig 4D ) . To confirm that arsenite primarily activates Hog1 through phosphatase inhibition , we next examined Hog1 phosphorylation in cells expressing a mutant version of the MAPKK Pbs2 ( Pbs2DD ) , where the activating phosphorylation sites of Pbs2 are mutated to phosphomimetic aspartate residues . The Pbs2DD mutant protein yields basal activation of Hog1 but prevents further upstream signalling events to the SAPK [10] . Consistent with arsenite-dependent inhibition of Hog1-specific phosphatase ( s ) , significant induction of Hog1 phosphorylation was observed in Pbs2DD cells following arsenite treatment ( Fig 4E ) . Furthermore , this arsenite-induction of Hog1 requires Pbs2 activity as arsenite-induced Hog1 phosphorylation does not occur in cells expressing an inactive Pbs2AA kinase where the activating phosphorylation sites of Pbs2 are mutated to alanine residues mimicking hypophosphorylation ( Fig 4E , compare PBS2 and PBS2DD with the PBS2AA lanes ) . Thus , these results strongly suggest that the ability of C . albicans cells to survive sustained activation of Hog1 upon loss of Ypd1 function is due to the action of Ptp phosphatases that reduce the levels of phosphorylated Hog1 . To test this directly , we sought to delete both PTP2 and PTP3 in tetO-YPD1 cells with the prediction that doxycycline-mediated repression of YPD1 in this background would result in catastrophic levels of Hog1 activation and cell death . Initially we focused on deleting PTP3 as this gene shows the greatest induction upon loss of YPD1 ( Fig 4A and 4B ) . Deletion of PTP3 in tetO-YPD1 cells resulted in notably higher levels of Hog1 phosphorylation upon repression of YPD1 expression ( Fig 4F ) . Moreover , deletion of PTP3 had a dramatic impact on the growth of tetO-YPD1 cells upon doxycycline-mediated repression of YPD1 , but not in the absence of doxycycline ( Fig 4G ) . These results support the model that induction of PTP3 promotes C . albicans survival following Ypd1 loss by limiting Hog1 phosphorylation . However , repression of YPD1 in ptp3Δ cells did not result in a lethal phenotype which is consistent with previous findings that Ptp2 and Ptp3 function redundantly to regulate C . albicans Hog1 [35] . Upon attempting to generate tetO-YPD1 ptp3Δ ptp2Δ cells we found that only one copy of PTP2 could be deleted . This was unexpected as a ptp2Δ ptp3Δ double mutant has previously been characterised [35] . Strikingly , however , tetO-YPD1 ptp3Δ ptp2/PTP2 cells exhibited a much higher basal level of Hog1 activation than that seen cells lacking PTP3 ( Fig 4F ) . Following doxycycline-mediated repression of YPD1 in ptp3Δ ptp2/PTP2 cells , further substantial increases in Hog1 phosphorylation were detected compared to that observed in tetO-YPD1 ptp3Δ cells ( Fig 4F ) . Indeed , consistent with the additive effect of deleting one copy of PTP2 , the growth of tetO-YPD1 ptp3Δ ptp2/PTP2 cells was barely detectable upon repression of YPD1 expression ( Fig 4G ) . However , it is important to note that such cells also display a slow growth phenotype under −DOX conditions when YPD1 is expressed ( Fig 4G ) . Nonetheless , taken together , these data support the model that the induction of PTP3 and PTP2 , together with the basal expression of PTP2 , facilitate C . albicans survival following loss of the Ypd1 phosphorelay protein . Although C . albicans can clearly tolerate sustained Hog1 activation , cells lacking YPD1 display reduced fitness compared to wild-type cells ( Fig 2C ) . Strikingly , however , we noted that ypd1Δ cells , when maintained on rich media plates , lost the morphological abnormalities associated with inactivation of Ypd1 function . Specifically , the swollen pseudohyphal morphology observed in freshly isolated ypd1Δ cells ( Day 1 ) , was largely replaced with normal budding cells after incubation on rich media plates for 13 days ( Fig 5A ) . Because of these findings we examined whether the stress phenotypes associated with YPD1 loss also changed over time . In agreement with our previous findings ( Fig 3 ) , freshly isolated ypd1Δ cells ( Day 1 ) exhibited a slow growth rate ( indicated by small colony size ) , and increased sensitivity to sodium arsenite and calcofluor white . However , by day 13 the slow growth phenotype was lost , and the sensitivity to sodium arsenite and calcofluor white mimicked that exhibited by wild-type cells ( Fig 5B ) . Consistent with this loss of morphological and stress sensitive phenotypes , we found that the high basal level of Hog1 phosphorylation , triggered by loss of YPD1 , decreased over a 13 day period ( Fig 5C ) . Hog1 phosphorylation levels were significantly reduced by day 10 , and by day 13 levels were similar to that seen in wild-type cells . Consistent with a reduction in Hog1 activation , the high basal level of expression of the Hog1 target gene , GPD2 , also declined over time ( Fig 5D ) . The decrease in Hog1 phosphorylation was not due to a reduction in Hog1 protein and/or HOG1 mRNA levels which remained constant over the 13 day experiment ( Fig 5C and 5D ) . This suggests that the sustained SAPK phosphorylation caused by loss of Ypd1 triggers adaptation within the cell that results in lower levels of activated Hog1 . As we had found the negative regulators PTP3 and PTP2 to be induced in ypd1Δ cells , we asked whether a further induction in their expression could contribute to the adaptation mechanism resulting in the time-dependent decline in Hog1 phosphorylation levels . In agreement with our previous findings ( Fig 4A ) , we found PTP3 and PTP2 to be up-regulated in ypd1Δ cells at day 1 and this was maintained at day 7 ( Fig 5B ) . However , by day 10 the levels of PTP3 and PTP2 in ypd1Δ cells had returned to wild-type levels . Hence , although C . albicans appears to adapt to sustained SAPK activation in the short term by triggering the up-regulation of PTP2 and PTP3 , this up-regulation is temporary . Indeed , the induction of PTP2 and PTP3 is only observed in cells in which significant levels of Hog1 activation are seen ( compare Fig 5C and 5D ) . This suggests that C . albicans adapts to Hog1 activation in the long term by via another mechanism ( s ) independent of Ptp2 and Ptp3 . Although the specific adaptation mechanism has not been identified , these results illustrate that C . albicans cells adapt to YPD1 loss over time by reducing the levels of phosphorylated Hog1 such that the negative effects of sustained SAPK activation are ablated , and ypd1Δ cells become phenotypically similar to wild-type cells . We asked whether the reduction in Hog1 phosphorylation levels following long-term sustained Hog1 activation was irreversible . Cells lacking ypd1Δ that had been maintained for 11 days on solid rich media , were then either kept on this media ( 13 day ) or patched onto fresh media in the absence ( -NaCl ) or presence ( +NaCl ) of 0 . 3M NaCl , a stress condition known to transiently activate Hog1 ( Fig 6A ) . Cells were taken from these plates after 2 days , sub-cultured in liquid media , and phosphorylation and total levels of Hog1 examined ( Fig 6A ) . As expected , a high basal level of Hog1 phosphorylation was absent in ypd1Δ cells at 13 days ( Fig 6B ) . However , passage of ypd1Δ cells over plates containing NaCl , but not lacking NaCl , resulted in a restoration of Hog1 phosphorylation ( Fig 6B ) . To determine whether NaCl treatment could fully restore Hog1 activation in ypd1Δ cells , we compared the level of Hog1 phosphorylation in ypd1Δ cells over the 13 day time course with that seen following passage over NaCl plates . Whilst exposure of ypd1Δ cells to NaCl does restore a high basal level of Hog1 phosphorylation , this is not to the same level as that seen in day 1 samples ( Fig 6C ) . Therefore , these data reveal that a transient exposure of ypd1Δ cells to osmotic stress can partially over-ride the adaptation mechanism ( s ) that prevents sustained Hog1 phosphorylation . Wild-type cells did not exhibit an increase in Hog1 phosphorylation after passage over media containing NaCl , illustrating that the phosphorylation observed in ypd1Δ cells is due to a re-establishment of sustained Hog1 activation , rather than stress-induced activation of the SAPK ( Fig 6B ) . Furthermore , the restoration of sustained Hog1 phosphorylation in ypd1Δ cells , resulted in the return of a swollen pseudohyphal morphology ( Fig 6D ) . Collectively , the results indicate that the cellular adaptation mechanisms that reduce Hog1 activation in cells lacking YPD1 can be partially over-ridden in the presence of stress that requires Hog1 function . The lethality associated with the deletion of YPD1 in S . cerevisiae and the fungal pathogen C . neoformans [20 , 21] , has led to much interest in this phosphorelay protein family as a potential prime antifungal target [22] . Hence , we next examined the impact of doxycycline-mediated repression of YPD1 on C . albicans virulence . The doxycycline-regulatable gene expression system has been successfully used to control C . albicans gene expression during infection in both a mouse model of systemic candidiasis [27 , 39] , and in a Caenorhabditis elegans infection model [40] . Furthermore , repression of YPD1 expression after infection has three clear advantages over testing the virulence of ypd1Δ cells directly . Firstly , this avoids the problems of accurately obtaining an inoculum size with the highly flocculent and filamentous ypd1Δ strain . Secondly , complications arising from lack of efficient dissemination of a highly flocculent strain are circumvented . Finally , doxycycline-mediated repression of YPD1 following infection more closely mimics the drug-induced inactivation of a particular fungal target following infection . Initially we employed the C . albicans-C . elegans liquid medium pathogenesis assay [40] to examine the impact of YPD1 repression on virulence . Worms were infected with tetO-YPD1 C . albicans cells which had been grown in the absence of doxycycline and thus expressing YPD1 ( Fig 2B ) . Subsequently , the animals were transferred to liquid medium in the presence ( +DOX ) or absence ( -DOX ) of doxycycline . Strikingly , we observed a significant increase in C . elegans killing after infection with tetO-YPD1 cells in the presence of doxycycline in which YPD1 expression is repressed ( Fig 7A ) . A total of 41% of infected animals died after 72 h in liquid medium containing doxycycline compared to 11% mortality in media lacking doxycycline ( P<0 . 001 ) . Importantly , consistent with previous reports [40] , doxycycline did not affect the pathogenicity of wild-type C . albicans strains towards C . elegans ( S2 Fig ) . Hence , taken together , these results suggest that doxycycline-mediated repression of YPD1 expression actually enhances the virulence of C . albicans towards C . elegans . To investigate whether these results could be corroborated in a distinct infection model , we employed the three day murine intravenous challenge model of C . albicans infection [41 , 42] . This model combines weight loss and kidney fungal burden measurements following 3 days of infection to give an ‘outcome score’ . A higher outcome score is indicative of greater weight loss and higher fungal burdens and thus increased virulence . Twelve mice were infected intravenously with tetO-YPD1 cells grown in medium lacking doxycycline . Following infection , one group of mice ( n = 6 ) were orally dosed with doxycycline daily ( +DOX ) to repress YPD1 expression , and the second group of placebo treated mice ( n = 6 ) given water ( -DOX ) . Significantly , inhibition of YPD1 expression during infection resulted in greater weight loss , increased kidney fungal burdens , and thus higher outcome scores compared to cells which continue to express YPD1 ( Fig 7B ) . Consistent with previous studies [27 , 39] , doxycycline treatment alone does not impact on C . albicans virulence in this mouse model of systemic candidiasis ( S3 Fig ) . Histological analysis of the kidneys revealed that there was significant inflammation associated with infection in the doxycycline treated animals ( Fig 7C panel i ) , with little inflammation seen for the placebo treated animals ( Fig 7C panel iv ) . Clusters of filamentous fungal cells were obvious in the doxycycline-treated mouse kidneys ( Fig 7C panels ii & iii ) , whereas only isolated fungal cells were found in the placebo treated kidneys ( Fig 7C panels v & vi ) . This indicates that the CFU measurements underestimate the increased fungal burden following repression of YPD1 expression , presumably as a consequence of the increased filamentation . Nonetheless statistical analysis revealed that the difference between +DOX and −DOX cells was significant for all three parameters , including fungal burden ( Fig 7B ) . Collectively , these studies illustrate that repression of YPD1 expression during infection in two distinct animal models enhances the virulence of C . albicans . These results have clinical significance as they predict that a drug designed to block two-component signalling in C . albicans could actually promote the virulence of this major human fungal pathogen . To further investigate how Ypd1 loss promotes C . albicans virulence , we employed live cell video microscopy to follow the impact of inhibition of YPD1 expression on C . albicans-macrophage interactions . Specifically , tetO-YPD1 cells were treated with ( +DOX ) or without ( -DOX ) doxycycline for 3 h prior to co-incubation with murine J774 . 1 macrophages , which were then grown in media +/-DOX , respectively . Quantitatively , there were no significant differences between the migration speed of J774 . 1 macrophages towards tetO-YPD1 cells treated or not with doxycycline , or rate of engulfment of fungal cells ( S4 Fig ) . Importantly , however , doxycycline mediated-repression of YPD1 resulted in C . albicans cells that displayed a significantly enhanced ability to kill macrophages . It was observed that 82±4 . 1% of macrophages were killed following co-incubation with tetO-YPD1 cells in the presence of doxycycline ( +DOX ) , compared to 60±3 . 0% macrophages killed in the absence of doxycycline ( -DOX ) ( P<0 . 01 ) ( Fig 7D ) . The ability of C . albicans to transition to the hyphal form following phagocytosis is pivotal in triggering macrophage death [43] . Thus , the rate of hyphae formation of tetO-YPD1 cells following phagocytosis by macrophages was measured . Hyphal growth was significantly faster in tetO-YPD1 cells grown in the presence of doxycycline ( 0 . 35±0 . 032 μm/min ) than in the absence of doxycycline ( 0 . 26 ± 0 . 023 μm/min ) ( Fig 7E ) . These results illustrate that repression of YPD1 expression promotes hyphal growth following phagocytosis which in turn likely enhances the ability of C . albicans to kill macrophages . Given the importance of macrophages in immune responses to C . albicans infections , the increased capacity of C . albicans cells lacking Ypd1 to kill macrophages likely contributes to the enhanced virulence observed in a murine model of systemic infection .
The generation of fungal pathogen-specific drugs is hindered by the conservation of many potential drug-targets in the human host . Thus the complete absence of two-component related proteins in metazoans , but their presence in fungi , has rendered such pathways attractive drug-targets [22 , 44 , 45] . In S . cerevisiae , and the human fungal pathogen C . neoformans , loss of the two-component phosphorelay protein Ypd1 causes lethality due to the sustained activation of their respective SAPK pathways [20 , 21] . However , as reported during the course of this work [25] , a major human fungal pathogen , C . albicans , can tolerate sustained activation of the Hog1 SAPK pathway triggered by loss of Ypd1 . We have significantly advanced our understanding of this observation in three main ways . Firstly we find that the constitutive activation of Hog1 in cells lacking YPD1 is mediated through two-component mediated regulation of the Ssk1 response regulator , and that the pleiotropic phenotypes associated with Ypd1 loss are dependent on Ssk1-mediated Hog1 activation . Secondly , we have provided novel insight into the mechanisms by which C . albicans survives and adapts to sustained SAPK activation . Thirdly , we show that inactivation of YPD1 promotes , rather than reduces , the virulence of C . albicans , and we provide evidence to suggest that this is mediated at least in part through effects on fungus-phagocyte interactions . A model summarising the major findings from this work is depicted in Fig 8 . One question raised by this study is why is sustained SAPK activation tolerated in some fungi but not others ? In S . cerevisiae , prolonged SAPK activation is reported to lead to cell death by causing an increase in reactive oxygen species ( ROS ) thus triggering apoptosis [12] . Although we do not know whether sustained Hog1 activation triggers an increase in intracellular ROS in C . albicans , it is worth noting that C . albicans is considerably more resistant to oxidative stress than S . cerevisiae [46] . In C . neoformans , lethality due to Hog1 hyperactivation is attributed to the over-accumulation of intracellular glycerol [47] . In C . albicans , sustained Hog1 activation in ypd1Δ cells also triggers an increase in intracellular glycerol levels , presumably contributing to the associated swollen cell phenotype . Interestingly , this is tolerated in these mutant cells and indeed it is noteworthy that C . albicans can withstand higher levels of osmotic stress ( which triggers glycerol biosynthesis ) than many other fungal species [46] . However , perhaps a key factor in maintaining the viability of ypd1Δ cells , is the significant induction of the tyrosine phosphatase encoding genes PTP2 and PTP3 which negatively regulate Hog1 phosphorylation [35] . Interestingly , the lethality associated with ypd1Δ-mediated sustained Hog1 activation in S . cerevisiae can be by-passed by artificial over-expression of either PTP2 or PTP3 [31 , 32] , suggesting that C . albicans actually adopts this strategy to prevent catastrophic levels of Hog1 activation . Consistent with this hypothesis is the observation that treatment of C . albicans ypd1Δ cells with the tyrosine phosphatase inhibitor , arsenite , triggers a significant increase in levels of Hog1 phosphorylation and cell death . Furthermore , deletion of PTP3 together with deletion of one copy of PTP2 , triggers dramatic increases in the level of phosphorylated Hog1 in tetO-YPD1 cells and furthermore , virtually abolishes cell growth following doxycycline-mediated repression of YPD1 . It is not clear why we could not generate tetO-YPD1 cells lacking both PTP3 and PTP2 , as a double ptp3Δ ptp2Δ mutant has previously been characterised [35] , although it is possible that basal YPD1 expression levels are altered in the tetO-YPD1 background . Notably , however , we find that the basal level of expression of PTP2 plays a major role in preventing Hog1 activation under non-stress conditions , as significant increases in the levels of Hog1 phosphorylation are seen upon deleting one copy of PTP2 . Similar findings have been reported in both S . cerevisiae and C . neoformans where basal levels of Hog1 are significantly increased in ptp2Δ mutant cells [31 , 34] . Taken together , the data presented in this paper strongly support the model that induction of PTP3 and PTP2 expression , together with the basal expression of PTP2 , allow C . albicans to survive loss of the Ypd1 phosphorelay protein . The viability of C . albicans cells lacking Ypd1 has allowed an investigation of how fungal cells adapt to long-term sustained SAPK activation . Remarkably , we find that cells evoke a mechanism that prevents the long-term constitutive phosphorylation of Hog1 . Moreover , this decrease in Hog1 phosphorylation is actually accompanied by a reduction in the levels of PTP2 and PTP3 , the negative regulators of Hog1 , raising the possibility that an upstream signalling branch to Hog1 is inhibited instead . Furthermore , this mechanism is reversible as it can be over-ridden following a transient exposure to stress that requires Hog1 activity . Do these observations that C . albicans can survive and adapt to sustained SAPK activation have physiological relevance ? As a human commensal , this fungal pathogen is continuously exposed to host-imposed stresses . Indeed , Hog1 is vital for C . albicans to exist commensally in the gut [11] , cause systemic infections [8 , 10] , and survive phagocytosis [9] . Thus , the ability of this pathogen to adapt to sustained SAPK activation by actively modulating Hog1 phosphorylation levels may be important to promote survival in certain host niches that continuously generate a stressful environment . Furthermore , we propose that the capacity of C . albicans to restore levels of SAPK phosphorylation , upon subsequent exposure to conditions that require Hog1-mediated stress responses , underpins the flexibility needed to allow adaptation of C . albicans to the range of environments encountered within the host [48] . Interestingly , long-term SAPK activation can be tolerated in other organisms . For example , in some human cell types sustained SAPK activation promotes either cell survival [49] or cellular differentiation [50] , rather than apoptosis . However , whether similar mechanisms are evoked in humans and C . albicans , to adapt to sustained SAPK activation is unknown . The main findings reported here in this paper focused on cells in which HOG1 is expressed from its native chromosomal locus . However , during the course of this work we noted that the cellular response of C . albicans to YPD1 loss differed depending on whether the HOG1 gene was present at its native locus , or re-integrated at the RPS10 locus in hog1Δ ypd1Δ+HOG1 cells . In both instances , cells adapted to long term SAPK activation by reducing the levels of phosphorylated Hog1 ( S5A and S5B Fig ) . However , when HOG1 was reintegrated at the RPS10 locus , this was also accompanied by a reduction in HOG1 mRNA levels ( S5C Fig ) . Consistent with the decrease in Hog1 levels , hog1Δ ypd1Δ+HOG1 cells changed dramatically over time to phenocopy hog1Δ cells ( S5D Fig ) . These results contrast significantly with those from ypd1Δ cells when HOG1 is expressed from its normal locus where the cells change over time to phenocopy wild type cells ( compare Fig 5 and S5 Fig ) . Indeed , the ypd1Δ-dependent down-regulation of HOG1 located at the RPS10 locus was independent of HOG1 promoter and terminator sequences ( S5E Fig ) , suggesting that this is a genome position-effect rather than a gene-specific phenomenon . Insertion of genes at the RPS10 locus has been employed by numerous labs to successfully generate re-integrant strains [51 , 52] . Significantly , the present study has highlighted the importance of studying a gene at its native chromosomal locus when investigating the role and/or regulation of the gene . Two-component proteins represent attractive antifungal targets as these signal transduction proteins are absent in metazoans [22 , 44 , 45] . Indeed , deletion of any one of the three C . albicans histidine kinases , or the Ssk1 response regulator , attenuate virulence in mouse systemic infection models . Furthermore , the Ypd1 phosphorelay protein has generated significant interest as an antifungal target , due to the lethality associated with its deletion in both S . cerevisiae and the human fungal pathogen C . neoformans [20 , 21] . However , in this study we report that inhibition of YPD1 expression during infection actually increases the virulence of C . albicans in two distinct infection models; a C . elegans pathogenesis model [40] and a murine model of systemic candidiasis [41 , 42] . Histology images from murine kidneys show that , similar to that seen in vitro , repression of YPD1 during infection results in clusters of highly filamentous cells . Enhanced filamentation in vivo has previously been shown to increase virulence [39] , and thus the increased virulence associated with YPD1 loss in vivo may be due to the formation of clusters of hyper-filamentous cells . Moreover , we show that drug mediated inhibition of YPD1 expression increases the efficacy by which C . albicans can kill macrophages . This is likely related to the observation that Ypd1 loss also results in increased filamentation within this infection model , as hyphae formation within the macrophage promotes C . albicans-mediated killing of macrophages by triggering pyroptosis [53] , and by mechanically rupturing the macrophage cell membrane [54] . The molecular basis underlying the increased virulence seen upon repressing YPD1 expression is unknown , but given that Hog1 is essential for C . albicans virulence in several infection models [8–11] , it is possible that the concomitant increases in Hog1 activity as a consequence of Ypd1 loss promotes C . albicans survival in the host ( Fig 8 ) . It is important to note here that the filamentous phenotype exhibited by cells lacking Ypd1 is due to the concomitant sustained Hog1 activation . To explore whether the enhanced virulence seen upon Ypd1 loss was dependent on Hog1 , we compared the virulence of wild-type , hog1Δ and hog1Δ ypd1Δ C . albicans cells in the three day murine infection model described above . Although it was not feasible to analyse ypd1Δ cells in this model due to the highly flocculent nature of this strain , hog1Δ and hog1Δ ypd1Δ cells demonstrated equally impaired virulence in the three-day murine systemic infection model ( S1 Table ) . Thus the impaired virulence exhibited by hog1Δ cells is not improved upon deletion of YPD1 and , therefore , is consistent with the hypothesis that the enhanced virulence observed upon loss of YPD1 may be dependent on the concurrent increased activation of the Hog1 SAPK to promote fungal stress resistance and/or filamentation . Importantly , the data presented here illustrating that Ypd1 loss potentiates C . albicans virulence , indicates that antifungals administered to treat C . albicans infections that target Ypd1 have the potential to actually enhance the virulence of this major pathogen . It is also noteworthy that SAPKs are important for the virulence of other human and plant fungal pathogens [55] . Consequently , any compound capable of inhibiting two-component signalling that leads to SAPK activation , may promote the survival of other fungal pathogens when encountering host-imposed stresses . Thus , in conclusion , our findings question the validity of the Ypd1 protein as a broad-spectrum antifungal target . Indeed , the significant differences in stress-signalling outputs between S . cerevisiae and C . albicans , underscores the importance of directly studying predicted ‘essential’ genes in pathogenic fungi , rather than in model yeast .
All the C . albicans strains used in this study are listed in Table 1 . Cells were grown at 30°C in YPD rich medium [56] . Addition of 20μg/ml doxycycline was used to repress expression from the tetO promoter . The strains used in this study are listed in Table 1 , and oligonucleotides used for their construction are listed in S2 Table . Glycerol concentrations were determined using The Free Glycerol Determination Kit ( Sigma-Aldrich ) , following the manufacturer’s instructions . Three independent biological replicates were performed . Protein extracts were prepared from mid-exponential phase cells and phosphorylated Hog1 was detected by western blot analysis with an anti-phospho-p38 antibody ( New England Biolabs ) as described previously [6] . Blots were stripped and total levels of Hog1 were determined by probing with an anti-Hog1 antibody ( Santa Cruz Biotechnology ) , and in some cases protein loading determined using an anti-tubulin antibody ( DSHB , University of Iowa ) . RNA preparation and Northern blot analyses were performed as described previously [6] . Gene-specific probes were amplified by PCR from genomic DNA using oligonucleotide primers specific for YPD1 , GPD2 , RHR2 , PTP2 , PTP3 , HOG1 and ACT1 ( S1 Table ) . C . albicans strains were grown at 30°C to exponential phase and then 10 fold serial dilutions were spotted onto YPD plates containing the indicated compounds . Plates were incubated at 30°C for 24 h . Differential interference contrast images were captured using a Zeiss Axioscope microscope as described previously [6] . All animal experiments were conducted in compliance with United Kingdom Home Office licenses for research on animals ( project license number PPL 60/4135 ) , and were approved by the University of Aberdeen Animal Welfare and Ethical Review Body ( AWERB ) . Animal experiments were minimised , and all animal experimentation was performed using approaches that minimised animal suffering and maximised our concordance with the 3Rs .
|
As fungi-attributed human deaths are increasing , there is an urgent need to develop new antifungal treatments . Two-component related proteins , such as the Ypd1 phosphorelay protein , have been heralded as antifungal targets as they are not found in humans and because inactivation of YPD1 in several different fungi causes sustained SAPK activation and cell death . However , we have discovered that inactivation of YPD1 in the major human pathogen , Candida albicans , actually enhances virulence . Furthermore , we reveal that this fungus adapts to the sustained activation of the Hog1 SAPK triggered by Ypd1 loss by mounting distinct mechanisms that actively reduce the level of phosphorylated Hog1 . These findings question the validity of Ypd1 proteins as broad-spectrum antifungal targets and provide insights into the cellular adaptation to sustained SAPK activation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"animal",
"models",
"of",
"disease",
"immunology",
"microbiology",
"animal",
"models",
"fungi",
"model",
"organisms",
"experimental",
"organism",
"systems",
"kidneys",
"fungal",
"pathogens",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"animal",
"models",
"of",
"infection",
"mycology",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"animal",
"studies",
"microbial",
"pathogens",
"mouse",
"models",
"yeast",
"biochemistry",
"candida",
"cell",
"biology",
"post-translational",
"modification",
"anatomy",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"cellular",
"types",
"macrophages",
"saccharomyces",
"cerevisiae",
"renal",
"system",
"organisms",
"candida",
"albicans"
] |
2017
|
Blocking two-component signalling enhances Candida albicans virulence and reveals adaptive mechanisms that counteract sustained SAPK activation
|
Central questions in regenerative biology include how stem cells are maintained and how they transition from self-renewal to differentiation . Germline stem cells ( GSCs ) in Caeno-rhabditis elegans provide a tractable in vivo model to address these questions . In this system , Notch signaling and PUF RNA binding proteins , FBF-1 and FBF-2 ( collectively FBF ) , maintain a pool of GSCs in a naïve state . An open question has been how Notch signaling modulates FBF activity to promote stem cell self-renewal . Here we report that two Notch targets , SYGL-1 and LST-1 , link niche signaling to FBF . We find that SYGL-1 and LST-1 proteins are cytoplasmic and normally restricted to the GSC pool region . Increasing the distribution of SYGL-1 expands the pool correspondingly , and vast overexpression of either SYGL-1 or LST-1 generates a germline tumor . Thus , SYGL-1 and LST-1 are each sufficient to drive “stemness” and their spatial restriction prevents tumor formation . Importantly , SYGL-1 and LST-1 can only drive tumor formation when FBF is present . Moreover , both proteins interact physically with FBF , and both are required to repress a signature FBF mRNA target . Together , our results support a model in which SYGL-1 and LST-1 form a repressive complex with FBF that is crucial for stem cell maintenance . We further propose that progression from a naïve stem cell state to a state primed for differentiation relies on loss of SYGL-1 and LST-1 , which in turn relieves FBF target RNAs from repression . Broadly , our results provide new insights into the link between niche signaling and a downstream RNA regulatory network and how this circuitry governs the balance between self-renewal and differentiation .
The balance between stem cell self-renewal and differentiation is pivotal for normal development , adult homeostasis , and regeneration . Indeed , aberrant stem cell regulation can cause disease , including human degenerative disorders and cancers [1] . Stem cell daughters can exist in a “naïve” multipotent state or a “primed” state that has been triggered to differentiate , typically via transit-amplification [2–4] . Stem cells that divide asymmetrically rely on oriented cell division to generate one naïve and one primed daughter [e . g . 5] , but the mechanism underlying stem cells that divide stochastically to generate pools of naïve and primed daughters [e . g . 6 , 7] remains largely unanswered . Challenges have included the complexity of their niches [8] and diversity of stem cell states ( e . g . quiescent vs . proliferative ) [9] . Thus , understanding how stem cell daughters are regulated to remain naïve or transition to a primed state can greatly benefit from a tractable model with well-defined niche and stem cells . The Caenorhabditis elegans gonad provides a paradigm for analyzing regulation of a stem cell pool [10] . In this system , a single-celled mesenchymal niche maintains a pool of ~225 stochastically-dividing germ cells in the “progenitor zone” ( Fig 1A ) [10] . That progenitor zone itself consists of a distal pool of 30–70 naïve germline stem cells ( GSCs ) and a more proximal pool of GSC daughters that have been triggered to begin differentiation and hence have been “primed” ( Fig 1A ) [11] . Central to GSC maintenance are two conserved regulators , Notch signaling and PUF ( for Pumilio and FBF ) RNA-binding proteins [12 , 13] . GLP-1/Notch signaling from the niche is essential for GSC maintenance [14] and two nearly identical PUF proteins , FBF-1 and FBF-2 ( collectively FBF ) , act as broad-spectrum repressors of differentiation RNAs to promote GSC self-renewal ( Fig 1B ) [15 , 16] . FBF provides one regulatory hub in the stem cell regulatory network; other hubs rely on GLD translational regulators to drive differentiation [17] . However , key questions remain . Here we focus on how Notch signaling and FBF repression are coordinated to establish a naïve GSC pool and facilitate transition to the primed state . Recently-identified GSC regulators are the sygl-1 and lst-1 genes , which are direct targets of niche signaling [18] . The lst-1 sygl-1 double mutant exhibits the same severe GSC loss as a GLP-1/Notch mutant while single mutants maintain GSCs , revealing functional redundancy [18] . However , the molecular functions of SYGL-1 and LST-1 have been a mystery . LST-1 harbors a single Nanos-like zinc finger , suggesting a possible role in post-transcriptional regulation . Yet both proteins are composed largely of low-complexity regions; neither is recognizable beyond Caenorhabditids; and the two amino acid sequences bear little resemblance to each other despite their redundancy [18] . Despite the novelty of these proteins , their striking GSC loss phenotype coupled with the restriction of their mRNAs to a region corresponding to the GSC pool [18 , 19] suggested that understanding their function and regulation would provide insights into regulation of a stem cell pool . Here we investigate SYGL-1 and LST-1 proteins to understand their roles in stem cell regulation . We find that both are cytoplasmic proteins and spatially restricted to the GSC region . Intriguingly , modest SYGL-1 expansion increases size of the stem cell pool , and vast expansion of either SYGL-1 or LST-1 drives formation of a germline tumor . The SYGL-1 and LST-1–dependent tumors form in the absence of GLP-1/Notch signaling , reinforcing their key roles in stem cell maintenance . However , SYGL-1 and LST-1 no longer drive tumor formation in the absence of FBF . Consistent with the idea that SYGL-1 and LST-1 drive stem cell self-renewal in a complex with FBF , SYGL-1 and LST-1 interact physically with FBF and are required for repression of an FBF target RNA . We suggest that SYGL-1 and LST-1 are FBF partners and function to ensure repression of FBF target RNAs within the stem cell pool .
To visualize SYGL-1 and LST-1 proteins , we generated epitope-tagged versions of sygl-1 and lst-1 , including single-copy transgenes using MosSCI [20–22] and endogenous alleles using CRISPR-Cas9 [23 , 24] ( Fig 1C and 1D ) . Importantly , these epitope-tagged SYGL-1 and LST-1 proteins were functional: they maintain GSCs when tested in appropriate mutant backgrounds ( S1D and S1E Fig ) . Therefore , they mimic their wild-type counterparts and we refer to them henceforth as SYGL-1 and LST-1 . By immunostaining , both proteins were expressed in the cytoplasm of the distal-most germ cells within the progenitor zone: SYGL-1 was largely punctate while LST-1 was enriched in perinuclear granules ( Figs 1E–1H and S1A–S1C ) . Using the conventional metric for position along the gonadal axis , germ cell diameters ( gcd ) from the distal end ( Fig 1A ) , we found SYGL-1 enriched from 1-~12 gcd , and LST-1 from 1-~5 gcd ( Fig 1K and 1L , see legend for how we determined extents ) . These protein extents correspond well to the distributions of their respective wild-type mRNAs , assayed by single-molecule FISH [19] , and were reproducible regardless of epitope tag . We counted the number of germ cells stained for each protein and found SYGL-1 in ~125 cells and LST-1 in ~45 germ cells ( Fig 1K and 1L ) . Strikingly , high SYGL-1 and LST-1 levels were correlated with low GLD-1 expression ( Fig 1I and 1J ) , consistent with their opposing functions ( see Introduction ) . We conclude that SYGL-1 and LST-1 are restricted within the progenitor zone to the GSC region , consistent with their pivotal roles in GSC self-renewal . The spatial restriction of SYGL-1 and LST-1 proteins suggested that their distribution might govern size of the GSC pool . Previous studies reported that progenitor zones ( PZ ) were smaller in sygl-1 and lst-1 single mutants than in wild type [18 , 25] , but GSC pool size was not analyzed . We first confirmed the decreased PZ size in mutants used previously , sygl-1 ( tm5040 ) and lst-1 ( ok814 ) . We also generated additional mutants: sygl-1 ( q828 ) deletes the entire open reading frame plus all introns ( Fig 1C ) and lst-1 ( q826 ) harbors a premature stop codon ( Fig 1D ) . PZ sizes were essentially the same for the various alleles of each gene ( S2A and S2B Fig ) , as were other measures ( e . g . brood size , fertility , embryonic lethality ) ( S2C and S2D Fig ) , suggesting that all are strong loss-of-function . We call them sygl-1 ( 0 ) and lst-1 ( 0 ) henceforth . Consistent with previous results [18 , 25] , the PZ size was affected differently for the two genes: the sygl-1 ( 0 ) PZ was about half the size of wild type , while the lst-1 ( 0 ) PZ was only marginally smaller than wild type ( S2A and S2B Fig ) . We therefore focused on the SYGL-1 extent and its relationship to GSC pool size . The onset of SYGL-1 expression relies on Notch signaling from the niche , which activates sygl-1 transcription [18 , 19] , but we thought its distribution might be refined post-transcriptionally since genome-wide studies identified RNA regulatory proteins binding to the sygl-1 3’UTR [26 , 27] . To test this idea , we replaced the sygl-1 3’UTR with a 3’UTR that supports expression throughout the germline , the tubulin tbb-2 3’UTR [28] . The transgene carrying this 3’UTR replacement was otherwise identical to the sygl-1 transgene described above ( Fig 1C ) , including insertion into the same chromosomal site and rescue of lst-1 sygl-1 double mutants from sterility to fertility ( S3A Fig ) . For simplicity , we refer in this section to the wild-type version as the “sygl-1 3’UTR” transgene , and to the replacement version as the “tbb-2 3’UTR” transgene ( Fig 2A ) . The tbb-2 3’UTR transgene , assayed in the absence of endogenous SYGL-1 , produced both an expanded distribution of SYGL-1 ( ~15 gcd or ~1 . 4-fold more extended than normal ) ( Fig 2B–2D ) and more abundant SYGL-1 ( ~2-fold more than normal ) ( Fig 2E ) . We conclude that the wild-type sygl-1 3’UTR restricts SYGL-1 distribution and lowers its abundance . We first found that PZ size was 1 . 3-fold larger in tbb-2 3’UTR transgenic animals than in either sygl-1 3’UTR transgenic animals or wild type ( Fig 2F ) . To test the idea that GSC pool size might also be enlarged , we used the emb-30 assay [11] . Briefly , this assay arrests cell divisions with a temperature-sensitive allele of emb-30 ( tn377 ) , which encodes a component of the anaphase promoting complex [29] . This arrest stops proximal movement of germ cells through the progenitor zone and resolves them into two discrete pools: cells in the distal GSC pool remain naïve and acquire an M-phase marker ( PH3 ) , while cells in the proximal pool are primed to differentiate and acquire a differentiation marker ( GLD-1 ) [11] ( Fig 2G ) . With this assay , we estimated GSC pool sizes in strains carrying emb-30 and either the wild-type sygl-1 locus ( normal SYGL-1 ) , the sygl-1 null mutant ( no SYGL-1 ) or the tbb-2 3’UTR transgene ( expanded SYGL-1 ) . GSC pools with wild-type SYGL-1 contained ~35 naïve cells; those with no SYGL-1 contained ~21 , and those with expanded SYGL-1 had ~43 on average ( Fig 2K ) . Indeed , the 1 . 4-fold increase in SYGL-1 extent ( from ~11 to ~15 gcd , on average ) corresponds well with the estimated 1 . 3-fold increase in GSC number ( from 35 to 43 , on average ) and PZ germ cell number ( from 229 to 298 , on average ) . Importantly , the extent of LST-1 expression along the gonadal axis ( gcd ) and number of LST-1–expressing cells in the distal gonad were essentially the same in sygl-1 ( + ) and sygl-1 ( 0 ) germlines as well as those harboring the tbb-2 3’UTR transgene ( S3B–S3F Fig ) . The simplest explanation is that LST-1 expression is likely independent of SYGL-1: The smaller LST-1 expression domain establishes a smaller GSC pool size in sygl-1 mutants , but that extent of SYGL-1 expression establishes GSC pool size in wild-type and tbb-2 3’UTR animals . We conclude that GSC pool size correlates with SYGL-1 extent and suggest that GSC pool size correlates with LST-1 extent in the absence of SYGL-1 . To extend the idea that distributions of SYGL-1 and LST-1 govern GSC pool size , we tested the effect of expressing SYGL-1 or LST-1 throughout the germline . To this end , we made single-copy transgenes driven with a mex-5 germline promoter and the tbb-2 3’UTR , which supports ubiquitous expression throughout the germline [28] ( Fig 3A ) . For brevity , we refer to the transgenes as sygl-1 ( ubiq ) and lst-1 ( ubiq ) , respectively ( Fig 3B and 3C ) . Because ubiquitous germline expression of SYGL-1 or LST-1 might render animals sterile , we created transgenes on sygl-1 or lst-1 feeding RNAi , and scored effects after RNAi removal , waiting 2–3 generations to minimize transgenerational RNAi inheritance ( Fig 3A ) . Strikingly , ubiquitous germline expression of either SYGL-1 or LST-1 created extensive germline tumors ( Fig 3E–3H ) . The penetrance of tumor formation depended on both temperature and number of generations after removal from RNAi , but was close to 100% for both sygl-1 ( ubiq ) and lst-1 ( ubiq ) after two or three generations at 15°C ( S4A and S4B Fig ) . About half of these tumors were proliferative throughout the gonad , while the other half included cells in the meiotic cell cycle , perhaps due to incomplete release from RNAi inheritance . Control animals harboring a GFP::H2B transgene driven with the same regulatory elements ( Fig 3D ) had no tumors ( Fig 3I and 3J ) , demonstrating that the tumors are specific to SYGL-1 or LST-1 . We next used markers to determine the state of cells in sygl-1 ( ubiq ) and lst-1 ( ubiq ) tumors . REC-8 localizes to the nucleus of germ cells in the mitotic cycle [30] and REC-8 was nuclear throughout the tumor ( Figs 3E and 3G , S4I and S4J ) ; PH3 marks M-phase [31] and was seen in dividing cells throughout the tumor ( Fig 3F and 3H ) ; and PGL-1 marks germ cells [32] and also was found throughout the tumor ( S4C and S4D Fig ) . Therefore , sygl-1 ( ubiq ) and lst-1 ( ubiq ) tumors are composed of germ cells that are mostly in the mitotic rather than the meiotic cell cycle . In addition , FBF-1 was abundant and GLD-1 was low throughout the tumors , consistent with germ cells being in an undifferentiated state ( S4F and S4G Fig ) . As expected , all markers behaved like wild type in the GFP::H2B control ( Figs 3I and 3J , S4E , S4H and S4K ) . We also assessed sygl-1 ( ubiq ) and lst-1 ( ubiq ) tumors for features reported in other mutants with germline tumors . The sygl-1 ( ubiq ) and lst-1 ( ubiq ) tumors formed in both XX hermaphrodites ( Fig 3E and 3G ) and XO males ( S4I and S4J Fig ) , in contrast to hermaphrodite-specific gld-1 tumors [33] . They formed in animals making only sperm ( males ) or only oocytes ( XX fog-3 females [34] ) , in contrast to spermatogenic-specific puf-8 germline tumors [35] . Finally , they did not rely on Notch signaling ( see below ) , in contrast to tumors arising from inappropriate soma/germline interactions or ectopic Notch activation [e . g . 36–39] . Thus , the most likely explanation of SYGL-1 and LST-1 tumors is that each regulator is sufficient to promote stemness in a germ-cell autonomous fashion and to do so in both sexes . The sygl-1 ( ubiq ) and lst-1 ( ubiq ) strains provide new reagents to explore how SYGL-1 and LST-1 function within the GSC regulatory pathway . Previous analyses placed sygl-1 and lst-1 downstream of , or parallel to , GLP-1/Notch signaling and upstream of GLD differentiation regulators , but their relationship with FBF was unresolved [18 , 25] ( Fig 1B ) . We first asked whether sygl-1 ( ubiq ) and lst-1 ( ubiq ) can bypass GLP-1/Notch signaling . Whereas glp-1 ( 0 ) mutants have no GSCs and make only a few sperm [14] ( Fig 4A ) , glp-1 ( 0 ) mutants develop germline tumors when either SYGL-1 or LST-1 is expressed ubiquitously ( Fig 4B and 4C ) , confirming that sygl-1 and lst-1 function downstream of Notch signaling . We next asked if sygl-1 ( ubiq ) and lst-1 ( ubiq ) can drive germline tumors in double mutants lacking both sygl-1 and lst-1 endogenous loci . Whereas lst-1 sygl-1 double mutants have no GSCs and only a few sperm ( Fig 4D ) [18] , they become tumorous when either SYGL-1 or LST-1 is expressed ubiquitously ( Fig 4E and 4F ) . Therefore , their tumor-forming activities are independent of each other , as expected . Finally , we asked if sygl-1 ( ubiq ) and lst-1 ( ubiq ) can drive germline tumors in fbf-1 fbf-2 double mutants . Previous experiments relying on loss-of-function mutants suggested that sygl-1 and lst-1 might function at the same position as fbf-1 and fbf-2 in the genetic pathway [25] ( Fig 1B ) . Here , using gain-of-function sygl-1 ( ubiq ) and lst-1 ( ubiq ) , we sought to clarify the relationship between sygl-1 , lst-1 and fbf . Because the GSC loss phenotype of fbf-1 fbf-2 is the most severe at 15°C [15 , 40] and sygl-1 ( ubiq ) and lst-1 ( ubiq ) are the most penetrant at 15°C ( S4A and S4B Fig ) , our initial analysis focused on 15°C . At this temperature , fbf-1 fbf-2 adults cannot maintain GSCs ( Fig 4G ) [15]; remarkably , they also cannot maintain GSCs even when either SYGL-1 or LST-1 is expressed ubiquitously ( Fig 4H and 4I ) . We confirmed that sygl-1 ( ubiq ) and lst-1 ( ubiq ) were expressed in fbf-1 fbf-2 mutants ( S5A–S5C Fig ) and that they made functional proteins ( S5D Fig ) . Therefore , the fbf-1 fbf-2 GSC loss is epistatic to sygl-1 ( ubiq ) and lst-1 ( ubiq ) tumors , which we interpret as sygl-1 and lst-1 acting either upstream or in parallel to FBF . In other words , SYGL-1 and LST-1 require FBF to drive self-renewal at this temperature . Although sygl-1 and lst-1 require FBF for tumor formation at 15°C , they unlikely drive stemness exclusively via FBF for two reasons: GSC loss is more severe in lst-1 sygl-1 double mutants than in fbf-1 fbf-2 double mutants [15 , 18] , and GSC loss in fbf-1 fbf-2 double mutants can be enhanced by removal of either lst-1 or sygl-1 [25; this work] . In an attempt to see their FBF-independent function , we tested fbf-1 fbf-2 sygl-1 ( ubiq ) and fbf-1 fbf-2 lst-1 ( ubiq ) animals for tumor formation at 25°C , because at this temperature , the FBF requirement is relieved in that fbf-1 fbf-2 mutants can maintain a small GSC pool [40] . Again at 25°C , both sygl-1 ( ubiq ) and lst-1 ( ubiq ) failed to generate germline tumors in the absence of FBF: fbf-1 fbf-2 sygl-1 ( ubiq ) maintained a progenitor zone comparable in size to fbf-1 fbf-2 double mutants while fbf-1 fbf-2 lst-1 ( ubiq ) were more variable , with only 10% maintaining a progenitor zone and differentiation extending to the distal end in the other 90% ( S5E–S5J Fig ) . Nonetheless , from lines of evidences noted above , SYGL-1 and LST-1 must have an FBF-independent role in stem cell maintenance . In summary , GLP-1/Notch signaling from the niche is dispensable for SYGL-1 and LST-1 tumors , and SYGL-1 and LST-1 do not need each other for their activity ( Fig 4J ) . In contrast , SYGL-1 and LST-1 rely on FBF to form tumors ( Figs 4J and S5E–S5J ) . Therefore , our results are consistent with a genetic model in which sygl-1 and lst-1 act downstream of Notch but upstream or parallel to fbf ( Fig 4K ) . The reliance of SYGL-1 and LST-1 on FBF to promote tumor formation suggested two ideas for their molecular function . One possibility was that SYGL-1 and LST-1 regulate FBF expression . To test this notion , we compared FBF expression in germlines with and without SYGL-1 and LST-1 , using a genetic background to circumvent the SYGL-1 and LST-1 requirement for GSC maintenance: gld-2 gld-1 mutants make germline tumors independently of sygl-1 and lst-1 [18] . To detect FBF-1 and FBF-2 , we used epitope-tagged transgenes , which are expressed and function biologically like their endogenous counterparts [27] . By staining , FBF-1 and FBF-2 proteins were expressed robustly both with and without SYGL-1 and LST-1 ( S6A–S6F Fig ) , and Western blots confirmed the result ( S6G Fig ) . We conclude that SYGL-1 and LST-1 are not required for FBF expression . An alternate idea posits that SYGL-1 and LST-1 act together with FBF , perhaps by enhancing FBF activity in a molecular complex . To ask if SYGL-1 and LST-1 physically interact with FBF , we first turned to the yeast two-hybrid assay ( Fig 5A ) . Briefly , SYGL-1 or LST-1 was fused to the Gal4 activation domain ( AD ) , and the PUF repeats of FBF-1 or FBF-2 were fused with the LexA DNA binding domain ( BD ) . Binding was assayed by monitoring growth on minimal media lacking histidine , as a measurement of HIS3 gene expression level . We imposed a stringent threshold by adding a competitive inhibitor of the HIS3 enzyme ( 50 mM 3-AT ) to minimize false positives . Robust growth was observed when either SYGL-1-AD or LST-1-AD was co-transformed with either FBF-1-BD or FBF-2-BD but not in controls ( Fig 5B and 5C ) . We conclude that SYGL-1 and LST-1 both interact with FBF-1 and FBF-2 in yeast . We next set out to ask if SYGL-1 and LST-1 might associate with FBF in nematodes . Immunoprecipitation of SYGL-1 and LST-1 from nematodes had been technically difficult because both proteins are normally expressed at low abundance and in only a subset of cells . To circumvent this problem , we attempted immunoprecipitation from sygl-1 ( ubiq ) and lst-1 ( ubiq ) tumorous animals . Immunoprecipitation was successful with SYGL-1 ( Fig 5D ) , and subsequent biochemistry therefore focused on SYGL-1 . To ask if SYGL-1 associates with FBF in nematodes , we generated strains harboring a sygl-1 ( ubiq ) transgene plus epitope-tagged 3xV5::FBF-2 . Our experimental and control strains made germline tumors with 3xFLAG::SYGL-1 and 3xMYC::SYGL-1 , respectively . The 3xV5::FBF-2 protein is functional and expressed ( S7A–S7D Fig ) , as previously described [41] . We used FLAG antibodies to immunoprecipitate ( IP ) protein from both experimental and control strains; RNase A was added to all IPs to exclude RNA dependence of interactions . 3xFLAG::SYGL-1 co-immunoprecipitated with 3xV5::FBF-2 from the experimental but not the control strain , and this interaction was not dependent on RNA ( Fig 5D ) . We conclude that SYGL-1 and FBF-2 associate with each other in nematodes and suggest that they form a complex . FBF regulates many target mRNAs ( see Introduction ) . If SYGL-1 works in a complex with FBF , then SYGL-1 protein might co-IP with FBF targets . To test this idea , we used the same strains described above and performed quantitative PCR of two established FBF targets , gld-1 and fem-3 mRNAs [15 , 27 , 42–44] . The experimental IP was enriched for both target mRNAs over the control IP , but it was not enriched for eft-3 mRNA ( Fig 5E ) , an mRNA not detected as a potential FBF target in genomic studies [27 , 42] . We conclude that SYGL-1 associates specifically in nematodes with both FBF protein and with FBF target mRNAs . The primary function of FBF in stem cell regulation is mRNA repression [16] . A crucial prediction of the idea that SYGL-1 and LST-1 work with FBF in a complex is that SYGL-1 and LST-1 should be required for repression of an FBF target mRNA . To test this idea , we examined gld-1 mRNA , a well-established FBF target required for differentiation [15] . Previous experiments detected a subtle increase in GLD-1 expression in GSCs of sygl-1 and lst-1 single mutants [25] . To explore this further , we again used gld-2 gld-1 mutants to remove both sygl-1 and lst-1 without changing cell fate . This time , however , we used gld-1 ( q361 ) , a missense mutant that abrogates GLD-1 protein function but produces detectable gld-1 mRNA and GLD-1 protein [30 , 45 , 46] ( Fig 6A ) . In this fashion , repression of gld-1 mRNA was uncoupled from complications of GLD-1 function in the germline . We first assayed expression of GLD-1 ( q361 ) protein . When either wild-type sygl-1 or wild-type lst-1 was present , GLD-1 ( q361 ) was expressed normally: barely detectable in distal-most germ cells and gradually increasing more proximally ( Fig 6B–6D ) . However , when both sygl-1 and lst-1 were removed , GLD-1 ( q361 ) protein increased dramatically in the distal germline ( Fig 6E ) , with quantitation revealing a three-fold increase on average ( Fig 6F ) . We next assayed expression of gld-1 ( q361 ) mRNA using single molecule fluorescence in situ hybridization ( smFISH ) . Our probe was specific to gld-1: transcripts were patterned as described previously in wild type [41 , 46] and cytoplasmic gld-1 mRNAs were undetectable in gld-1 ( q485 ) , a deletion mutant that likely renders transcripts subject to non-sense mediated decay [45] ( S8 Fig ) . Similar to the result with GLD-1 protein , gld-1 mRNAs were barely detectable distally when either sygl-1 or lst-1 was present , but became easily detectable distally when both sygl-1 and lst-1 were removed ( Fig 6G–6J ) . By contrast , nascent transcripts were seen in distal germ cell nuclei regardless of sygl-1 and lst-1 ( Fig 6K and 6L ) . We conclude that SYGL-1 and LST-1 function post-transcriptionally to repress gld-1 mRNA expression in the distal germline , a role that is strongly reminiscent of FBF activity . Collectively , our data support the idea that SYGL-1 and LST-1 partner with FBF to repress FBF target mRNAs in GSCs .
We have found that ubiquitous expression of either SYGL-1 or LST-1 protein drives formation of extensive germline tumors , and that their tumor-forming activities do not require GLP-1/Notch signaling from the niche . The significance of this result is three-fold . First , SYGL-1 and LST-1 are not only required for GSC maintenance , albeit redundantly [18] , but each on its own also drives stemness in the form of a tumor when ubiquitously expressed . This sufficiency underscores the importance of SYGL-1 and LST-1 as key stem cell regulators . Second , SYGL-1 and LST-1 are the primary targets of niche signaling for GSC maintenance: GLP-1/Notch signaling does not induce other regulators that must work with either SYGL-1 or LST-1 to maintain GSCs . Third , spatial restriction of SYGL-1 and LST-1 prevents tumor formation , making them prototypes for a new class of oncogenes . Central to understanding the niche regulation of stem cells is the identification and characterization of key downstream effectors . Advances have been made in several model systems [e . g . 47–49] , but examples of niche effectors with validated in vivo significance are rare . Perhaps the most striking parallels to SYGL-1 and LST-1 are Ascl2 and LgR5 , which encode niche signaling effectors in Wnt-regulated intestinal stem cells . Similar to SYGL-1 and LST-1 , Ascl2 and LgR5 expression is limited to stem cells [50 , 51] , and ectopic expression promotes hyperplasia [52] . However , in stark contrast to SYGL-1 and LST-1 , Ascl2 and LgR5 functions are not independent of niche signaling: LgR5 enhances Wnt signaling and Ascl2 works with Wnt-dependent transcription factors to induce a stem cell transcriptional signature [53] . Therefore , SYGL-1 and LST-1 stand out as direct targets of niche signaling that promote self-renewal by an intrinsic signaling-independent mechanism . Normally , SYGL-1 and LST-1 are spatially restricted to a region that correlates with estimates of the GSC pool ( Fig 7A ) . We confirmed the biological significance of this spatial restriction in two ways . First , a moderate expansion of SYGL-1 expression led to a similar moderate expansion of pool size . Second , a major expansion of either SYGL-1 or LST-1 led to the formation of massive germline tumors . The simple conclusion is that the presence of either SYGL-1 or LST-1 promotes the stem cell fate , while their absence is critical for the transition towards differentiation . Logical corollaries are that spatial distributions of SYGL-1 and LST-1 govern the size of the GSC pool and that their loss facilitates the transition to a cell state primed for differentiation . A key question is how their spatial restriction is regulated . GLP-1/Notch signaling from the niche activates sygl-1 and lst-1 transcription in distal germ cells [18] , but what regulates their disappearance ? A partial answer is RNA regulation: the sygl-1 3’UTR restricts SYGL-1 protein expression compared to the tbb-2 ( tubulin ) 3’UTR . In addition to RNA regulation , we suggest that SYGL-1 and LST-1 protein stabilities are also regulated . Despite the rapid kinetics of germ cell movement ( ~1 gcd per hour [54] ) , the distributions of sygl-1 mRNA and protein are similar , as are those of lst-1 mRNA and protein [19; this work] . Therefore , the SYGL-1 and LST-1 proteins must turn over as germ cells move proximally within the progenitor zone . Others have found that the proteolytic machinery is critical for progression from a stem cell state to a differentiated state in the progenitor zone [55 , 56] . We suggest that SYGL-1 and LST-1 are likely targets of such proteolysis . The C . elegans gonad therefore provides a new paradigm for how niche signaling can act through spatially restricted regulators to not only ensure the existence of stem cells but also to govern the size of a stem cell pool and facilitate the transition to a primed state . Spatial regulation is a common theme in animal development [57 , 58] and extends to stem cell regulators . In addition to Lgr5 and Ascl2 ( described above ) , the Escargot/Snail transcription factor follows a similar principle in intestinal stem cells in Drosophila and mouse models [59 , 60] . More relevant to this work is the Drosophila PUF protein , Pumilio , which promotes GSC self-renewal [61 , 62] . Pumilio is spatially restricted to GSCs and its ectopic expression generates germline tumors [63] . The clarifying advances of our work are an application of this theme to the maintenance of a stem cell pool , which is likely a broadly-used mechanism , and to a PUF protein partner rather than a PUF protein per se ( see below ) . When this work began , the molecular functions of SYGL-1 and LST-1 were unknown ( see Introduction ) . A first clue from this work was their cytoplasmic localization , which is consistent with a role in post-transcriptional regulation but can be explained in other ways . A more significant clue was that SYGL-1 and LST-1 cannot drive germline tumors in the absence of the FBF RNA-binding protein . One explanation might have been that SYGL-1 and LST-1 promote FBF expression , but that possibility was not confirmed: FBF-1 and FBF-2 were expressed in the absence of SYGL-1 and LST-1 . An alternative idea was that SYGL-1 and LST-1 might work with FBF to promote mRNA repression . In support of that explanation , SYGL-1 and LST-1 interact with FBF-1 and FBF-2 in yeast two-hybrid assays; SYGL-1 co-immunoprecipitates from nematodes with both FBF-2 protein and with FBF target mRNAs; and SYGL-1 and LST-1 post-transcriptionally repress expression of one of those FBF targets in GSCs . These multiple lines of evidence support the model that SYGL-1 and LST-1 partner with FBF to repress mRNAs in GSCs ( Fig 7B ) . We emphasize that SYGL-1 and LST-1 must also have FBF-independent functions , because the lst-1 sygl-1 phenotype is more severe than the fbf-1 fbf-2 phenotype [15 , 18] , and because single sygl-1 and lst-1 mutants enhance the fbf-1 fbf-2 phenotype [25; this work] . The fog-1 gene , which encodes a cytoplasmic polyadenylation element binding ( CPEB ) related protein [64 , 65] , redundantly promotes GSC self-renewal with FBF in that fog-1 fbf-1 fbf-2 triple mutants contain a GSC loss similar to that of glp-1 null [66] . We speculate that the FBF-independent functions of SYGL-1 and LST-1 may involve regulation of FOG-1 protein or key FOG-1 mRNA targets . But of course , other possibilities exist . Regardless , this work shows conclusively that SYGL-1 and LST-1 have an FBF-dependent function and that they likely operate with FBF in a complex . SYGL-1 and LST-1 stand out among PUF partners as the first to be essential for GSC maintenance , the first to be spatially restricted to the stem cell region , the first to affect size of a stem cell pool , the first to be tumorigenic when overexpressed , and the first to be essential for mRNA repression in GSCs . Previously identified FBF partners include NOS-3 , a Nanos homolog which is expressed throughout the germline [67 , 68] , and CPEB/CPB-1 , which is expressed and functions in spermatocytes [64 , 69] . Two other FBF partners , GLD-2 and GLD-3 , activate mRNAs and promote germ cell differentiation [70–72] , a function opposite that of SYGL-1 and LST-1 . The molecular mechanisms by which SYGL-1 and LST-1 repress RNAs await future studies . The simplest possibility is that they enhance FBF recruitment of the Not1 deadenylase complex , a conserved mode of PUF repression from yeast to humans [73–75] . Another idea is that SYGL-1 or LST-1 influences the sequence specificity and kinetics of FBF binding to target mRNAs , analogous to reports for other PUF partners such as CPB-1 for FBF [76] and Nanos or Brat for Drosophila Pumilio [77 , 78] . A third thought is that SYGL-1 and LST-1 repress RNAs by recruiting them to sites of repression in RNP granules . The emerging view of low complexity proteins as RNA granule scaffolds [e . g . 79 , 80] coupled with the punctate or granular appearance of SYGL-1 and LST-1 make this third possibility attractive , but it remains speculative . Given that several mechanisms remain plausible , we note that SYGL-1 and LST-1 may employ distinct biochemical mechanisms , despite their biological redundancy in GSC maintenance and their molecular redundancy in gld-1 mRNA repression . A tantalizing future direction is to ask if similar counterparts of SYGL-1 or LST-1 exist in other vertebrate stem cell models to enhance the repressive activity of PUF proteins , Pum1 and Pum2 . Our findings together with previous studies support a model for how niche signaling is coordinated with intrinsic stem cell regulators to establish a GSC pool with stem cells in their naïve state and then facilitate the transition to a state primed for differentiation ( Fig 7C ) . Essentially , Notch signaling localizes the GSC pool by activating expression of key intrinsic stem cell regulators , SYGL-1 and LST-1 , which partner with FBF to repress differentiation mRNAs and thereby promote the naïve state ( Fig 7C , left ) [14 , 15 , 18 , 19; this work] . Pool size is established roughly by Notch signaling , which activates sygl-1 transcription in a steep gradient across the pool [18 , 19] . However , sygl-1 mRNAs are less graded and therefore transform the steep transcriptional gradient into a markedly less steep RNA gradient [19] . Here , we show that SYGL-1 protein abundance disappears in a pattern closely mirroring loss of its mRNAs . We propose that removal of these key FBF partners drives the transition from a naïve to a primed state ( Fig 7C , middle ) , and that loss of SYGL-1 and LST-1 triggers entry into a primed state by releasing gld-1 and likely other RNAs from repression ( Fig 7C , right ) . We note that FBF is present not only in the GSC pool but also in primed cells and cells beginning overt differentiation ( entry into meiotic prophase ) [15 , 41 , 81] . However , repression of FBF target mRNAs occurs in the distal germline [15 , 40 , 75 , 82–84] and is strongest in the distal-most region or the naïve GSC pool [11] . This pattern suggests that FBF in primed cells is becoming less repressive as SYGL-1 and LST-1 are lost; indeed , FBF may be transitioning to an activating mode in this primed region [10 , 75] . Two other FBF partners , GLD-2 and GLD-3 , activate FBF-bound RNAs [75] , suggesting the possibility of a partner exchange during the transition in primed cells . One can imagine that SYGL-1 and LST-1 might be displaced by competition of other FBF partners or they might be removed by spatially regulated proteolysis . Although our model is surely oversimplified , it provides a heuristic framework for future explorations of stem cell pool regulation . For example , the model poises our thinking for analysis of both the mechanism and kinetics of transition from a naïve state to a primed state , which are likely to have profound consequences on pool regulation . Regardless , this model provides critical insights into how niche signaling is coordinated with downstream intrinsic effectors to govern the existence of a stem cell pool and its size .
Most strains were maintained and characterized at 20°C under standard conditions [85] , except as follows: strains containing emb-30 ( tn377ts ) were maintained at 15°C; strains harboring sygl-1 ( ubiq ) tumor transgenes ( qSi235 , qSi297 ) were maintained on sygl-1 ( RNAi ) feeding bacteria , and strains with lst-1 ( ubiq ) tumor transgenes ( qSi267 ) were maintained on lst-1 ( RNAi ) ( see RNAi section of Methods ) . The wild type was N2 Bristol strain . Alleles are as follows: LGI: gld-2 ( q497 ) [86] , gld-1 ( q485 ) [33] , gld-1 ( q361 ) [45] , fog-3 ( q520 ) [34] , lst-1 ( ok814 ) [87] , lst-1 ( q826 ) ( this work ) , sygl-1 ( tm5040 ) [18] . LGII: fbf-2 ( q704 ) [15] , fbf-2 ( q738 ) [81] , fbf-1 ( ok91 ) [15] . LGIII: glp-1 ( q46 ) [14] , emb-30 ( tn377ts ) [29] , unc-119 ( ed3 ) [88] . Balancers are as follows: LGI: hT2[qIs48] [89] , LGII: mIn1[mIs14 dpy-10 ( e128 ) ] [90] , LGIII: hT2[qIs48] [89] . Transgenes are as follows: LGII: weSi2[Pmex-5::GFP::his-58::tbb-2 3’end , unc-119 ( + ) ] [91] , qSi22[Plst-1::lst-1::1xHA::lst-1 3’end , unc-119 ( + ) ] ( this work ) , qSi49[Psygl-1::3xFLAG::sygl-1::sygl-1 3’end , unc-119 ( + ) ] ( this work ) , qSi69[Plst-1::lst-1::3xFLAG::lst-1 3’end , unc-119 ( + ) ] ( this work ) , qSi75[Pfbf-2::3xFLAG::fbf-2::fbf-2 3’end , unc-119 ( + ) ] [27] , qSi150[Psygl-1::3xFLAG::sygl-1::tbb-2 3’end , unc-119 ( + ) ] ( this work ) , qSi232[Pfbf-1::3xFLAG::fbf-1::fbf-1 3’end , unc-119 ( + ) ] [27] , qSi235[Pmex-5::3xFLAG::sygl-1::tbb-2 3’end , unc-119 ( + ) ] ( this work ) , qSi267[Pmex-5::lst-1::3xFLAG::tbb-2 3’end , unc-119 ( + ) ] ( this work ) , qSi297[Pmex-5::3xMYC::sygl-1::tbb-2 3’end , unc-119 ( + ) ] ( this work ) . LGIV: qSi93[Plst-1::lst-1::1xHA::lst-1 3'end , unc-119 ( + ) ] ( this work ) . Alleles generated using CRISPR-Cas9 are as follows: LGI: lst-1 ( q1004 ) [lst-1::3xV5] ( this work ) , lst-1 ( q1008 ) [lst-1:: 3xOLLAS] ( this work ) , sygl-1 ( q828 ) ( this work ) , sygl-1 ( q964 ) [3xMYC::sygl-1] ( this work ) , sygl-1 ( q983 ) [3xOLLAS::sygl-1] ( this work ) , sygl-1 ( q1015 ) [sygl-1::1xV5] ( this work ) . LGII: fbf-2 ( q931 ) [3xV5::fbf-2] ( this work ) , fbf-2 ( q932 ) [3xV5::fbf-2] ( this work ) . A complete list of strains used in this study is summarized in S1 Table . Single-copy transgenes were generated using the Mos-1 mediated single-copy insertion method ( MosSCI ) [20–22] . Briefly , plasmids containing the gene of interest were constructed using the Gibson assembly method [92] and microinjected at 50 ng/μl along with transposase and co-injection markers to target the ttTi5605 or cxTi10816 sites . Several transgenes were generated and maintained on RNAi feeding bacteria . Those requiring sygl-1 ( RNAi ) were qSi235[Pmex-5::3xFLAG::sygl-1::tbb-2 3’end , unc-119 ( + ) ] and qSi297[Pmex-5::3xMYC::sygl-1::tbb-2 3’end , unc-119 ( + ) ] . That requiring lst-1 ( RNAi ) was qSi267[Pmex-5::lst-1::3xFLAG::tbb-2 3’end , unc-119 ( + ) ] . At least two independent lines for each construct were analyzed , and results of one representative line are reported . A complete list of generated alleles and plasmids used to generate MosSCI transgenes can be found in S2 Table and S4 Table respectively . sygl-1 ( q828 ) was generated using CRISPR/Cas9 gene editing [93] . Briefly , three 25 ng/μl sygl-1 sgRNAs , a 50 ng/μl repair template designed to substitute the sygl-1 coding region with Caenorhabditis briggsae unc-119 , and 50 ng/μl pDD162 encoding Cas-9 were microinjected into the unc-119 ( ed3 ) strain with co-injection markers , and progeny were screened for the Unc movement rescue . The substitution of the sygl-1 gene with the unc-119 gene resulted in deletion of the sygl-1 coding region and was verified by sequencing . The alleles fbf-2 ( q931 ) , fbf-2 ( q932 ) , sygl-1 ( q964 ) , sygl-1 ( q983 ) , lst-1 ( q1004 ) , lst-1 ( q1008 ) , and sygl-1 ( q1015 ) were generated by RNA protein complex ( RNP ) CRISPR [23 , 24] . Briefly , injection mix containing gene-specific crRNAs ( 10 μM , IDT Alt-RTM ) , dpy-10 or unc-58 co-CRISPR crRNAs ( 4 μM , IDT Alt-RTM ) , tracrRNAs ( 14 μM , IDT Alt-RTM ) , gene-specific repair oligo ( 4 μM ) or repair plasmid ( 50 ng/μl ) , dpy-10 or unc-58 co-CRISPR repair oligo ( 1 . 4 μM ) , and Cas-9 protein ( 25 μM ) was prepared . Strains were microinjected and the progeny were screened using PCR for edits . All CRISPR alleles were verified by sequencing and outcrossed 2–4 times with wild type prior to analysis . A complete list of reagents used to generate CRISPR alleles can be found in S3–S5 Tables . To obtain lst-1 ( q826 ) , a sygl-1 enhancer screen was performed with EMS mutagenesis as described [85] , with minor modifications . Briefly , sygl-1 ( tm5040 ) hermaphrodites of the fourth larval stage ( L4 ) were mutagenized with 25 mM Ethyl methanesulfonate ( Sigma #M0880 ) for 4 hours at room temperature . F1 progeny were singled and maintained at 15°C , and F2 self-progeny were screened for germline proliferation defective ( Glp ) [14] mutants . Details of this mutagenesis screen are available upon request . The lst-1 locus was sequenced from DNA extracted from Glp animals to identify the lst-1 ( q826 ) allele , which was outcrossed 10 times with wild type prior to analysis . The sygl-1 and lst-1 gene structures reported here are based on 5’ rapid amplification of cDNA ends ( RACE ) , genome-wide mRNA sequencing data ( WormBase release 255 ) , and ribosome profiling data [94] . Specifically , the sygl-1 5’UTR , the lst-1 5’UTR , and the lst-1 start codon have been re-annotated . Most importantly , our reported lst-1 start codon removes 70 amino acids from the previously mis-annotated versions [18] and is consistent with evolutionary data from C . briggsae . For 5’ RACE , total RNA was extracted from young adult wild type ( 24 hours after L4 at 20°C ) using TRIzol ( Invitrogen #15596026 ) . 1 μg of total RNA was converted to cDNA with SuperScript III ( Invitrogen #18080051 ) using sygl-1_RT_primer ( 5’-AGCGACGAGTTGAAGAGACTC-3’ ) or lst_RT_primer ( 5’-GGTGCGACATGTCTCGTGGATC-3’ ) . cDNAs were purified ( QIAquick PCR purification kit , Qiagen #28106 ) , tailed with cytosines using Terminal Deoxynucleotidyl Transferase ( Invitrogen #EP0161 ) , and then PCR amplified using the following primers: for sygl-1 , primary PCR used Anchor_Primer ( 5’-GGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG-3’ ) with sygl-1_primary ( 5’-TCGACGAGCGAGTCAGTCTC-3’ ) ; secondary PCR used Universal_amplification_primer ( 5’-GGCCACGCGTCGACTAGTAC-3’ ) with sygl-1_secondary ( 5’-CGCCTCCGGTTGACGATGATG-3’ ) ; and tertiary PCR used Universal_amplification_primer with sygl-1_tertiary ( 5’-AGACGATGAGGTGGACATG-3’ ) . An additional tertiary reaction was carried out to improve the signal to noise ratio . For lst-1 , primary PCR used Anchor_Primer ( 5’-GGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG-3’ ) with lst_primary ( 5’-GAGTTGAAGCAGTTGCTTCGG-3’ ) and secondary PCR used Universal_amplification_primer ( 5’-GGCCACGCGTCGACTAGTAC-3’ ) with lst_secondary ( 5’-gtgttgcgacttcgagtagg-3’ ) . All amplified products were analyzed by Sanger sequencing . L4 hermaphrodites were placed onto individual plates at 20°C . At 6 to 12 hour intervals , the hermaphrodite was moved to a new plate and the embryos were counted for sterility and brood counts . Several days later , hatched progeny on each plate were counted to determine embryonic lethality . All characterization of progenitor zone ( PZ ) size was done in animals raised at 20°C until 24 hours after L4 , except in S5J Fig where animals were raised at 25°C until 16–18 hrs after L4 . To visualize nuclear morphology , gonads were dissected , fixed , and stained with DAPI ( see immunostaining and DAPI staining section below for dissection and fixation methods ) . To determine the PZ size , gonads were imaged using a confocal microscope with a z-stack depth of 0 . 4–0 . 5 μm . Next , the boundary between PZ and Transition Zone ( TZ ) was determined by conventional criteria [95] . Briefly , many germ cells in the TZ have entered meiotic prophase and hence have a crescent-shaped nuclear morphology . The PZ/TZ boundary was scored as the distal-most cell row with at least two crescent-shaped nuclei . Finally , the cells within the progenitor zone were counted manually using the cell-counter plug-in in FIJI/Image J , with each DAPI-stained nucleus scored as a single cell . To estimate the number of germ cells in fbf-1 fbf-2 gonads reported in S5D Fig , compact nuclei typical of mature sperm in a gonadal arm were counted manually using the cell counter tab in Openlab 5 . 5 . 2 ( PerkinElmer ) . Next , the number of sperm was converted to the number of germ cells ( four sperm are made from one germ cell ) . To estimate the number of distal germ cells expressing SYGL-1 or LST-1 , JK4996 , JK5073 , JK5205 , JK5263 , JK5893 , JK5929 and JK6002 were raised at 20°C until adulthood ( 24 hours after L4 ) , along with appropriate wild-type controls . Gonads were dissected , fixed , and stained with anti-FLAG , anti-OLLAS , anti-V5 , or anti-HA ( see immunostaining section below ) and imaged using the confocal microscope . Next , the number of distal germ cells that contained positive V5 or OLLAS signal ( SYGL-1 ) or positive HA , FLAG , or V5 signal ( LST-1 ) above the background level was manually scored , using the cell-counter plugin in FIJI/Image J . The assay was performed as described [11] with minor modifications . DG627 , JK5233 , JK5235 animals were raised at 15°C until 36 hours past mid-L4 , then moved to plates pre-incubated at 25°C and maintained at 25°C for 12 . 5 hours . We chose 12 . 5 hours because germ cell counts became unreliable with longer times ( nuclear morphology became increasingly compromised after incubations of 13 hours and longer ) . Next , gonads were dissected , fixed , and stained for anti-PH3 , anti-GLD-1 and DAPI ( see staining section below ) . To estimate the number of cells within the distal pool , we manually counted the number of M-phase arrested germ cells distal to the GLD-1 boundary ( as assessed by DAPI morphology and PH3 staining ) using the cell counter tab in Openlab 5 . 5 . 2 ( PerkinElmer ) . Scoring was done blind to genotype . We excluded samples with abnormal , fragmented nuclei that made cell counting unreliable ( 22–49% per genotype ) . We note that not every nucleus distal to the GLD-1 boundary was arrested in M-phase in some gonads but these few nuclei were included in the “distal pool” counts . Feeding RNAi was performed as described [96] using sygl-1 or lst-1 clones from the Ahringer RNAi library [97] . Bacteria were grown overnight at 37°C in 2xYT media containing 25 μg/μl carbenicillin and 50 μg/μl tetracycline . Cultures were concentrated , seeded onto Nematode Growth Medium ( NGM ) plates containing 1mM IPTG , then induced overnight before plating worms . To induce sygl-1 ( ubiq ) and lst-1 ( ubiq ) germline tumors , L4 P0 animals were transferred from RNAi bacteria to OP50-seeded NGM plates , and subsequent generations were monitored using a dissecting scope for germline tumor formation . In some experiments , gravid adults were bleached between generations to synchronize populations . All experiments with sygl-1 ( ubiq ) and lst-1 ( ubiq ) were carried out at 15°C to maximize tumor penetrance , except those in S4A and S4B Fig , where tumor penetrance was tested with different temperature regimens , and in S5E–S5J Fig , where epistasis with fbf-1 fbf-2 was assayed at 25°C . For most sygl-1 ( ubiq ) tumors , data were obtained in the F3 generation after removal from RNAi , and for most lst-1 ( ubiq ) tumors , data were obtained in F2 after removal from RNAi . Two exceptions were: ( 1 ) For epistasis experiments requiring a balancer for strain maintenance ( JK5401 , JK5403 , JK5538 , JK5585; see Fig 4B , 4C , 4E , 4F ) , tumors were scored in F1 , because all F1 balancer-carrying animals were tumorous so additional generations could not be obtained . ( 2 ) For 25°C epistasis experiments with fbf-1 fbf-2 ( see S5E–S5J Fig ) , we scored in F8 ( sygl-1 ) and F7 ( lst-1 ) after removal from RNAi to maximize tumor penetrance . Staining followed established protocols [98] with minor modifications . Briefly , staged animals were dissected in PBStw ( PBS + 0 . 1% ( v/v ) Tween-20 ) with 0 . 25 mM levamisole to extrude gonads . Tissues were fixed in 2% ( w/v ) paraformaldehyde diluted in 100 mM K2HPO4 ( pH 7 . 2 ) for 10 minutes when using anti-FBF-1 and anti-PGL-1 antibodies . For all other antibodies , tissues were fixed in 3% ( w/v ) paraformaldehyde diluted in 100 mM K2HPO4 ( pH 7 . 2 ) for 30 minutes . Post fixation , all samples were permeabilized with ice-cold methanol or PBStw + 0 . 2% ( v/v ) Triton-X for 5–10 minutes . Next , they were blocked with either 30% ( v/v ) goat serum diluted in PBStw ( for anti-FLAG ) or 0 . 5% ( w/v ) bovine serum albumin diluted in PBStw ( all other antibodies ) for 1 hour . For primary antibodies , samples were incubated overnight at the following dilutions in the blocking solution: Mouse anti-FLAG ( 1:1000 , M2 clone , Sigma #F3165 ) , Rabbit anti-GLD-1 ( 1:100 , Gift from E . Goodwin ) , Rat anti-HA ( 1:100 , 3F10 clone , Roche #11867423001 ) , Rabbit anti-REC-8 ( 1:100 , [30] ) , Rat anti-FBF-1 ( 1:5 , [15] ) , Mouse anti-SP56 ( 1:200 , [99] ) , Mouse anti-PH3 ( 1:200 , Cell Signaling #9706 ) , Rabbit anti-PGL-1 ( 1:100 [32] ) , Mouse anti-V5 ( 1:1000 , Bio-Rad #MCA1360 ) , Rat anti-OLLAS ( 1:2000 , L2 clone , Novus Biologicals #NBP1-96713 ) . For secondary antibodies , samples were incubated for 1 hour at room temperature at the following dilutions: Donkey Alexa 555 anti-mouse ( 1:1000 , Invitrogen #A31570 ) , Goat Alexa 555 anti-rabbit ( 1:1000 , Invitrogen #A21429 ) , Goat Alexa 488 anti-rabbit ( 1:1000 , Invitrogen #A11034 ) , Donkey Alexa 488 anti-rat ( 1:500 , Invitrogen #A21208 ) , Donkey Alexa 647 anti-mouse ( 1:500 , Invitrogen #A31571 ) . To visualize DNA , DAPI was included at a final concentration of 0 . 5–1 ng/μl during the last 10 minutes of secondary antibody incubation . Vectashield ( Vector Laboratories #H-1000 ) was used as mounting medium . Single molecule FISH ( smFISH ) was performed as described [19 , 41 , 100] . Custom Stellaris FISH probes were designed by utilizing the Stellaris FISH probe designer ( Biosearch Technologies , Inc ) available online at www . biosearch . com/stellarisdesigner . The gld-1 probe set contains 48 unique probes labeled with CAL Fluor Red 610 and was used at a final concentration of 0 . 25 μM . For the compound microscopy data shown in Fig 4 , images were taken using a Zeiss Axioskop with Hamamatsu CCD or ORCA cMOS camera equipped with 63x 1 . 4NA Plan Apochromat oil immersion objective . Carl Zeiss filter sets 49 , 38 , and 43HE were used for the visualization of DAPI , Alexa 488 , and Alexa 555 respectively . An X-Cite 120Q lamp ( Lumen Dynamics ) was used as the fluorescence light source . Openlab 5 . 5 . 2 ( PerkinElmer ) and Micromanager [101 , 102] were used as acquisition software . For all other figures , a Leica TCS SP8 confocal microscope driven by LAS software version 3 . 3 . 1 or X was used . This laser scanning confocal microscope was equipped with Photomultiplier ( PMT ) and Hybrid detectors ( HyD ) . For all images , a 63x 1 . 4NA HC Plan Apochromat oil immersion objective was used with 100–200% zoom for immunostaining , and 300% zoom for single molecule FISH , using the standard scanner with 400Hz scanning speed . For figure preparation , contrast was linearly adjusted in Adobe Photoshop identically across all samples . In some cases , images were merged using the stitching plugin in FIJI/Image J [103] to generate whole gonad images . All images used for quantitation were acquired using the sequential scan mode on the Leica TCS SP8 , under the same conditions across all samples . Next , average intensity of multiple z-slices was projected onto a single plane . To eliminate signal intensities outside of the gonad ( i . e . intestine ) , a separate binary mask was created by thresholding Nomarski images of the gonad taken at the same time; the binary mask was then multiplied to other channels such that only signals within the gonad would be considered for quantitation . Next , intensity at a given distance “x” from the distal tip of the gonad was averaged over five-micron intervals ( “moving average” ) . For simplicity , distance from the distal end was converted to conventional germ cell diameters , using a conversion ratio of 4 . 55 μm for one germ cell diameter [19] . A custom MATLAB script was used to process steps described above . Modified yeast two-hybrid assays were performed as described [104] . Briefly , sygl-1 cDNA encoding full-length SYGL-1 ( a . a . 1–206 ) or lst-1 cDNA encoding full-length LST-1 ( a . a . 1–328 ) was cloned into the Nco I and Xho I sites in pACT2 ( Gal4 activation domain plasmid ) to generate pJK1580 and pJK2015 , respectively . Regions encoding FBF-1 ( a . a . 121–614 ) and FBF-2 ( a . a . 121–632 ) were cloned into the EcoR I and Sal I sites in pBTm116 ( LexA binding domain plasmid ) to generate pJK2019 [67] and pJK2017 , respectively . Plasmids were co-transformed into a L40-ura strain using the Te-LiAc method [105] . His3 reporter activity was assayed on synthetic defined medium ( SD ) supplemented with –Leu–Trp–His containing 50 mM 3-Amino-1 , 2 , 4-triazole ( Sigma #A8056 ) , or –Leu–Trp plates as controls for 4 days at 30°C . JK5366 , JK5574 , JK5783 , and JK5844 animals were raised at 15°C until they developed germline tumors as young adults ( 12 hours after L4 ) ( see tumor assay above ) . Animals were washed twice with M9 buffer [3 g/L KH2PO4 , 6 g/L NaHPO4 , 5 g/L NaCl , and 1 mM MgSO4] and cross-linked with 1% ( w/v ) formaldehyde for 10 minutes at room temperature ( RT ) . Pellets were resuspended in 1 ml lysis buffer [50 mM HEPES pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , 1% ( v/v ) Triton-X , complete Protease inhibitor cocktail ( Roche ) ] , frozen in liquid nitrogen , and pulverized with mortar and pestle for 10 minutes . Lysates were cleared twice by centrifugation ( 12 , 000g , 10 minutes ) , and the total protein concentration was measured by Direct Detect Spectrophotometer ( Millipore ) . To prepare antibody conjugated beads , 30 μg anti-FLAG ( M2 clone , Sigma #F3165 ) was incubated with 4 . 5 mg protein G Dynabeads ( Novex , Life Technologies , #10003D ) for 30 minutes at RT . Next , 20 mg lysates were incubated with the antibody-bead mixture for 4 hours at 4°C , with the presence of RNase A at 10 μg/ml . RNA degradation was confirmed by isolating total RNA from post-IP lysates using TRIzol LS ( Invitrogen #10296028 ) and analyzing on agarose gels . Beads were pelleted , washed four times with lysis buffer , and then two times with wash buffer [50 mM HEPES pH 7 . 5 , 0 . 5 M NaCl , 1 mM EDTA , 1% ( v/v ) Triton X-100] . Samples then were eluted with elution buffer [1% ( w/v ) SDS , 250 mM NaCl , 1 mM EDTA , 10 mM TRIS pH 8] for 10 minutes at 65°C , and analyzed using SDS-PAGE on an 8% or 12% acrylamide gel . To probe FBF abundance in S6G Fig , N2 , JK5181 , JK5182 , JK5600 , JK5602 , JK5603 , and JK5604 animals were raised at 20°C to young adulthood ( 12 hours after L4 stage ) . 40 animals were boiled in 2x Laemmli buffer and then analyzed by SDS-PAGE on a 4–20% gradient gel ( Lonza #58527 ) . For primary antibodies , blots were incubated overnight at 4°C at the following dilutions: Mouse anti-FLAG ( 1:1000 , M2 clone , Sigma #F3165 ) , Mouse anti-V5 ( 1:1000 , Bio-Rad #MCA1360 ) , Mouse anti-actin ( 1:40 , 000 , C4 clone , Millipore #MAB1501 ) , Mouse anti-α-tubulin ( 1:20 , 000 , Sigma #T5168 ) . For secondary antibody , blots were incubated for 1 hour at RT with Donkey HRP-conjugated anti-mouse ( 1:10 , 000 , Jackson ImmunoResearch ) . Immunoblots were developed using SuperSignalTM West Pico/Femto Sensitivity substrate ( Thermo Scientific #34080 , #34095 ) and imaged using an ImageQuant LAS4000 ( GE Healthcare ) . FIJI/Image J was used to calculate blot intensity . For final figure preparations , contrast of the blot was linearly adjusted in Adobe Photoshop . JK5366 and JK5574 were raised at 15°C until they developed germline tumors as young adults ( see tumor assay above ) . Immunoprecipitation was done as above except that formaldehyde cross-linking and RNase treatment of lysates were omitted . Instead , lysis buffer contained 1 U/μl SUPERase·In RNase inhibitor ( Ambion #AM2694 ) . Successful IP was confirmed by analyzing 10% of elution by Western blot , and RNA was eluted from the rest of the beads with 0 . 5 ml TRIzol ( Invitrogen #15596026 ) . RNA was purified by RNeasy Micro kit ( Qiagen #74004 ) including DNase I treatment on column . Purified RNA was checked for integrity , and converted to cDNA with Superscript III first strand synthesis kit ( Invitrogen #18080051 ) using random-hexamers as primers . Quantitative PCR was carried out using a Roche Lightcycler 480 with TaqMan gene expression assays ( Applied Biosystems ) . Enrichment was calculated by ΔΔ CT method [106] . Taqman probes used are as follows: gld-1 , Ce02409901_g1; eft-3 , Ce02448437_gH; rps-25 , Ce02464216_g1; fem-3 , Ce02457444_g1 . Statistical tests are indicated in figure legends with sample sizes . In most cases , one-way ANOVA and post-hoc Tukey multiple comparison tests were performed to calculate p-values . In cases where equal variance assumption of ANOVA was not established at p<0 . 01 ( Levine’s test ) , Welch’s one-way ANOVA ( modified ANOVA with heteroskedastic data ) and post-hoc Games-Howell multiple comparison tests were performed to calculate p-values . All statistics were performed in R .
|
Stem cells lie at the heart of metazoan development , regeneration , and tissue homeostasis , but the molecular basis of their regulation is poorly understood in their natural context within an animal . Here we investigate this problem in the nematode gonad , where germline stem cells are maintained by Notch signaling from the niche and PUF RNA binding proteins in stem cells . Yet the link between Notch and PUF has been elusive . The two Notch target genes essential for GSC maintenance encode novel proteins with few clues to function . Here we report that these mysterious proteins are cytoplasmic and function post-transcriptionally as PUF partners to ensure RNA repression . We also show that the restricted spatial distribution of these newly identified regulators governs the size of the stem cell pool and prevents tumor formation . In sum , our results demonstrate how niche signaling is linked with downstream regulators to govern the stem cell fate and establish a stem cell pool .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] |
[
"tumor",
"stem",
"cells",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"gonads",
"nuclear",
"staining",
"population",
"genetics",
"gene",
"pool",
"cell",
"differentiation",
"germ",
"cells",
"developmental",
"biology",
"stem",
"cells",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"animal",
"cells",
"stem",
"cell",
"niche",
"dapi",
"staining",
"cell",
"biology",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"evolutionary",
"biology",
"genital",
"anatomy"
] |
2017
|
SYGL-1 and LST-1 link niche signaling to PUF RNA repression for stem cell maintenance in Caenorhabditis elegans
|
Replicating viruses have broad applications in biomedicine , notably in cancer virotherapy and in the design of attenuated vaccines; however , uncontrolled virus replication in vulnerable tissues can give pathology and often restricts the use of potent strains . Increased knowledge of tissue-selective microRNA expression now affords the possibility of engineering replicating viruses that are attenuated at the RNA level in sites of potential pathology , but retain wild-type replication activity at sites not expressing the relevant microRNA . To assess the usefulness of this approach for the DNA virus adenovirus , we have engineered a hepatocyte-safe wild-type adenovirus 5 ( Ad5 ) , which normally mediates significant toxicity and is potentially lethal in mice . To do this , we have included binding sites for hepatocyte-selective microRNA mir-122 within the 3′ UTR of the E1A transcription cassette . Imaging versions of these viruses , produced by fusing E1A with luciferase , showed that inclusion of mir-122 binding sites caused up to 80-fold decreased hepatic expression of E1A following intravenous delivery to mice . Animals administered a ten-times lethal dose of wild-type Ad5 ( 5×1010 viral particles/mouse ) showed substantial hepatic genome replication and extensive liver pathology , while inclusion of 4 microRNA binding sites decreased replication 50-fold and virtually abrogated liver toxicity . This modified wild-type virus retained full activity within cancer cells and provided a potent , liver-safe oncolytic virus . In addition to providing many potent new viruses for cancer virotherapy , microRNA control of virus replication should provide a new strategy for designing safe attenuated vaccines applied across a broad range of viral diseases .
Viruses have a highly successful history as prophylactic vaccines and are also being developed for their intrinsic anticancer activities [1] . In both settings the ability to undergo restricted replication is highly desirable . Attenuated ( but not killed ) viral strains often represent the most effective viral vaccines , affording the possibility of persistent low level infection without significant pathology [2] , [3] . Unfortunately many viruses are not suitable for production of attenuated forms , and reversion to wild-type represents a significant risk . Equally the field of cancer ‘virotherapy’ relies on selective replication of lytic viruses within cancer cells , leading to cell death and spread of infection to adjacent cancer cells . Several ‘conditionally-replicating’ viruses have been engineered for activation by tumour-associated changes , showing greater potency in cancer cells than in normal cells . Unfortunately these agents are generally attenuated compared to the equivalent wild-type virus even in cancer tissues , and have so far shown little therapeutic activity in clinical trials [4] , [5] . For both vaccination and cancer virotherapy it would be attractive to produce viruses that show wild-type replication activity at therapeutic sites ( eg . within tumours or at sites of antigen presentation ) but are specifically attenuated at sites of potential pathology . The network of naturally-occurring non-coding microRNA molecules [6] negatively regulates cellular gene expression post-transcriptionally through a number of mechanisms that all involve binding of microRNA to complementary regions within a messenger RNA ( mRNA ) [7] , [8] leading to decreased protein production [9] . Tissue-selective microRNA expression is now well characterised [10] , and it provides an opportunity to regulate transgene expression from therapeutic nucleic acids and viruses . This principle was originally developed by Brown et al . [11] , who showed that inclusion of microRNA mir-142-3p binding sites within 3′UTR of retrovirally-encoded transgenes prevented expression in antigen presenting cells , preventing stimulation of an immune response and allowing long term transgene expression in other cells without rejection . The diversity of tissue-specific microRNAs now identified should enable this approach of selective attenuation of viral expression to be developed in several different contexts [12] . It is also possible to use the microRNA system to regulate replication of vaccines or conditionally-replicating ‘oncolytic’ viruses . The small size of the required insertion ( microRNA binding sites are generally only 21–24 bp ) provides considerable flexibility in the design of capacity-restricted therapeutic viruses . Equally the negative regulatory principle of using microRNA regulation to inhibit viral replication in sites of toxicity , could allow significant therapeutic potency by avoiding attenuation in target sites . This approach has previously been used to generate microRNA-controlled conditionally-replicating RNA viruses , including Polio virus for vaccination and coxsackie virus for cancer virotherapy . These viruses were engineered to contain binding sites for neural and muscle specific microRNAs respectively . The neural-restricted polio virus showed good vaccine potential , while the muscle-restricted coxsackie virus showed decreased myositis and improved anticancer efficacy [13] , [14] . In this study we have explored the use of this approach in engineering a microRNA-controlled wild-type adenovirus , a DNA virus , by expressing binding sites for microRNA mir-122 within the 3′ UTR of E1A . Mir-122 is highly and selectively expressed in hepatocytes [10] , [15] , and this modification might prevent expression of E1A within hepatocytes , thereby reducing adenovirus replication and hepatotoxicity whilst maintaining its therapeutic replication within tumour cells .
To assess the repression capabilities of mir-122 , CMV promoter-driven luciferase plasmids containing 0 , 4 and 8 sense or 4 anti-sense microRNA binding sites ( representative structures shown in Figure 1 ) were transfected into HEK-293 , OVCAR-3 and HUH7 cell lines using DOTAP ( Roche ) and luciferase activity was measured by luminometry after 24 h . The presence of the microRNA binding sites had no effect on luciferase levels detected in the mir-122 negative cell lines HEK-293 and OVCAR-3 ( Figure 2 ) . In contrast , in mir-122-positive HUH7 cells , luminescence was decreased from 7 . 9×105 RLU/µg ( anti-sense control plasmid ) to 9 . 9×104 RLU/µg ( 4 microRNA binding sites , P = 0 . 001 ) ) and 3 . 4×104 RLU/µg ( 8 microRNA binding sites , P = 0 . 001 ) . The inclusion of 4 anti-sense microRNA binding sites did not effect luciferase activity compared to the unmodified control plasmid in any cell type . Whilst the inclusion of 8 microRNA binding sites did show improved repression in comparison to 4 binding sites , we decided to use 4 binding sites for future use in view of the repetitive nature of the insertion and to minimise the likelihood of viral recombination . Luciferase expression from the microRNA-controlled plasmids shown in Figure 1 was assessed in murine livers in vivo , using an Ivis100 imaging system . Plasmid vectors were delivered at equimolar amounts by hydrodynamic delivery and imaging was performed at 8 , 24 and 48 h post injection . Control CMV promoter-driven plasmids gave high levels of transgene expression after 8 h ( 2 . 7×1011 RLU ) while inclusion of 4 microRNA binding sites in the same plasmid decreased expression to 5 . 7×109 RLU , a 47-fold decrease in expression ( Figure 3A ) . Total levels of luciferase expression fell substantially over the next 40 h , although the differential expression increased up to 129-fold ( P = 0 . 0064 ) after 48 h ( Figure 3C ) . Plasmids containing the E1A promoter and E1A coding sequence were engineered to generate an E1A-luciferase fusion transcription cassette . This vector was then further modified to contain four binding sites to mir-122 to allow in vivo imaging of E1A expression ( Figure 1 ) . Following hydrodynamic delivery of equimolar amounts of both vectors , expression from the plasmid producing the E1A-luciferase fusion protein ( with no microRNA sites ) was much lower than from the equivalent pCMV vector , probably reflecting relatively weak activity of the E1A promoter in murine cells , however the inclusion of four microRNA sites within this plasmid again mediated a significant decrease in expression ( 86-fold after 8 h , P = 0 . 01 , Figure 3B ) . It was noticeable that luciferase expression from the fusion protein decreased more rapidly with time than from the pCMV-driven vectors , perhaps reflecting the ability of E1A to negatively regulate its own promoter ( Figure 3C ) . Adenoviruses containing E1A-luciferase fusion constructs on a background of wild-type Ad5 ( Figure 4 ( iii ) and 4 ( iv ) ) were used to infect mir-122-negative A549 and OVCAR-3 cell lines in vitro . Luciferase activity from both Ad-E1A-Luc ( containing no mir-122 binding sites ) and Ad-E1A-Luc-mir122 ( containing 4 mir122 binding sites ) increased slowly between 8 and 24 h and then showed a more rapid rise that was sustained up to at least 72 h ( Figure 4B and 4C ) . This profile of luminescence may reflect initial transcription from the input viral genomes that increases rapidly following viral genomic replication . The microRNA insertion into the 3′ UTR did not affect the profile of luciferase expression in these cells , suggesting the modification did not influence the stability of mRNA encoding the E1A-luciferase fusion protein , nor did it inhibit virus replication in these mir-122-negative cells . To ascertain whether the microRNA insertion would also be inactive in mir-122- negative cells in vivo , viruses ( 1×1010 v . p . ) were injected subcutaneously in Balb/C mice ( n = 3 ) and animals were imaged after 24 h . Results demonstrated no significant difference between the expression from the two viruses ( data not shown ) suggesting no effects of the microRNA at the subcutaneous site . Adenoviruses encoding the E1A-luciferase protein with and without four microRNA binding sites were used to infect a monolayer of the mir122 positive cell line Huh7 . E1A-luciferase expression was monitored by luminometry from 6 h to 72 h post-infection ( Figure 4E ) . Luciferase expression from the Ad5-E1A-Luc showed a small but significant rise between 0 and 24 h ( reaching 1 . 1×105 RLU/µg protein ) and then increased rapidly , rising to 1 . 7×106 RLU/µg protein by 72 h . This suggests E1A transcription and replication proceeded similarly to the situation in A549 and OVCAR cell lines . In contrast , Ad5-E1A-Luc-mir122 virus showed significantly less luciferase expression at all time points , reaching only 6 . 3×104 RLU/µg after 72 h ( P = 0 . 0001 for both 48 and 72 h ) . The differential in luciferase expression between the viruses with and without microRNA binding sites increased over time , suggesting decreased genome replication of Ad5-E1A-Luc-mir122 compared to Ad5-E1A-Luc . In order to confirm that this differential in luciferase expression was due to mir-122 knockdown of E1A , a precursor RNA mimic of mir-122 ( Ambion ) was introduced into A549 cells to simulate hepatocyte expression . Ad-E1A-Luc-mir122 and either the mir122 pre-cursor , or negative control pre-mir ( Ambion ) were added to cells and luciferase readings performed after 24 h . Results showed that the introduction of the pre-mir122 reduced luciferase , and therefore E1A , expression from 9 . 2×104 RLU ( negative control pre-mir ) to 3 . 4×103 RLU ( P = 0 . 0001 , Figure 4B ) . To assess the in vivo activity of these viruses and to observe the effects of time on E1A expression over 96 h , 5×1010 vp of Ad5-E1A-Luc and Ad5-E1A-Luc-mir122 were injected intravenously into Balb/C mice . Animals were imaged at 6 , 24 , 48 , 72 & 96 h ( Figure 5 ) . After 6 h , Ad-E1A-Luc showed a luminescence signal of 1 . 6×108 RLU whilst Ad-E1A-Luc-mir122 showed only 3 . 0×106 , a differential of 52-fold . Interestingly , the signal from the Ad5-E1A-Luc treated mice increased by 2 . 5×109 RLU between 48 and 72 h ( Figure 5 , time course ) possibly reflecting a wave of virus replication . At the same time the microRNA regulated virus showed only a relatively small increase ( a rise of 3 . 4×107 RLU ) . After 96 h the differential expression between the viruses with and without microRNA sites had reached 80 fold . In order to assess the effects on hepatic toxicity of including mir-122 binding sites within wild-type adenovirus , 5×1010 v . p of Ad5WT and Ad5-mir122 were injected intravenously to Balb/C mice . One mouse in the study which received Ad5WT became hunched and immobile , and was sacrificed after 60 h with visible hepatic pathology . Remaining mice were exsanguinated under anaesthesia 72 h post-injection and blood was allowed to clot . Serum from both groups was tested for Alanine Aminotransferase ( ALT ) levels and Aspartate Aminotransferase ( AST ) to assess hepatic damage . Mice administered wild-type Ad5 showed significantly increased ALT levels ( 90 times higher than control mice treated with PBS , P = 0 . 0001; Figure 6A ) suggesting substantial liver damage had occurred . Mice administered Ad5-mir122 showed approximately 15-fold less serum ALT ( 5 times normal ) demonstrating that less liver toxicity had occurred with this virus . AST readings demonstrated similar results with a 17 fold decrease in AST in serum from mice administered Ad5-mir122 compared to serum from mice receiving Ad5WT ( P = 0 . 0002 , Figure 6A ) . To evaluate viral replication and tissue damage , livers were divided for histological analysis and QPCR . Livers taken from mice administered wild-type Ad5 showed an average of 2×109 genomes/mg liver ( wet weight; Figure 6B ) . In the total liver this represents approximately 60-fold more genome copies compared to the total amount of virus originally injected , suggesting significant genome replication . In contrast , livers from mice administered Ad5-mir122 showed only 8×107 virus genomes/mg liver , representing less than a doubling compared to the input dose ( P = 0 . 0001 for Ad5-mir122 compared to Ad5WT ) . These data confirm that the microRNA suppression of E1A is capable of significantly reducing replication of the virus genome in murine liver in vivo . Histological analysis showed a dramatic difference between animals administered wild-type Ad5 and those administered Ad5-mir122 . Wild-type Ad5 induced vacuolation , haemorrhaging and abnormal nuclear morphology , while livers from mice administered Ad5-mir122 showed very little pathology , with some mice showing no aberrant morphology in any liver section ( Figure 6C ) . Histological images of liver from a mouse administered 5×1010 vp of a non-replicating adenoviral vector are presented for comparison , showing similar or slightly greater liver pathology than was induced by Ad5-mir122 . The maximum tolerated dose of Ad5WT given i . v . is reported as about 1×109 PFU [16] , and this was confirmed in studies using nude mice bearing HepG2 human hepatocellular carcinoma xenografts ( data not shown ) . Animals were found to tolerate higher levels of Ad-mir122 ( 6×1010 v . p . , 9×109 PFU ) with only mild weight loss , although when this dose of Ad-mir122 was administered on two consecutive days , all mice were showing signs of virus-related toxicity by day 4 following the first injection . These mice were put down and the livers demonstrated macroscopic signs of viral liver damage . It therefore appears that , in tumour bearing animals , the maximum tolerated dose of Ad-Mir122 lies between 6×1010 and 1 . 2e1011v . p/mouse ( 9×109–1 . 8×1010 pfu ) .
Molecular engineering of replicating viruses to avoid pathology whilst maintaining potency in therapeutic sites would provide an important new platform for design of viral vaccines and oncolytic treatments . In this study we explored the possibility of achieving this using a DNA virus , wild-type Ad5 , engineered to avoid its major toxicity in murine liver by including four binding sites for hepatocyte-specific mir-122 within the 3′UTR of E1A . To measure E1A expression non-invasively we introduced a luciferase coding region 3′ to the E1A coding region of wild-type virus in order to produce a contiguous E1A-luciferase expression cassette , where E1A splicing would produce a series of E1A-luciferase fusion proteins . This novel virus ( including a modified version containing 4 mir-122 binding sites in the E1A 3′ UTR ) produced strong luciferase activity in vitro and in vivo that reported E1A protein levels clearly , enabling non-invasive real-time assessment of protein translation including the effects of virus genome replication . Measuring E1A protein in this way is a more reliable indicator of microRNA activity than measuring E1A mRNA , since microRNA regulation is known to affect protein translation via multiple pathways . However , given that our microRNA target sites are precisely complementary to mir-122 it is likely that argonaut 2 -mediated RNA cleavage is responsible for the majority of the knockdown observed . While the presence of the luciferase sequence slightly decreased the rate of cell killing in vitro , compared to the corresponding virus without luciferase , a complete cytopathic effect was still achieved in permissive cells after one extra day . This suggests that the fusion proteins retain all essential E1A functions . This is perhaps unsurprising given that E1A protein has been shown to still operate despite significant deletions and insertions , lacking both enzyme activity and significant secondary structure [17] . Wild-type Ad5 is normally capable of an abortive genome replication cycle in murine liver in vivo , where it mediates considerable and sometimes lethal hepatotoxicity [16] , [18] . It was unclear whether microRNA regulation could successfully control Ad5 , since the DNA genome is not a direct target for microRNA recognition and it is known that even small amounts of E1A translation can lead to genomic replication , which will then provide a template for more transcription providing a greater challenge for microRNA control . Nevertheless , although E1A production in mir-122-positive Huh7 cells in vitro was decreased only about 95% following introduction of 4 mir-122-binding into the E1A-luciferase reporter virus , in vivo luciferase imaging suggested a greater suppression of E1A expression by mir-122 , showing a 50-fold differential after 6 h that rose to 80-fold after 96 h . This may reflect a higher expression of mir-122 in murine hepatocytes in vivo than in human Huh7 cells . To complement the E1A reporter luciferase data , hepatic replication and toxicity was also assessed using wild-type Ad5 and compared with a ‘wild-type’ modified to contain four mir-122 sites ( Ad-mir122 ) . After 72 h the serum ALT was decreased 15-fold for the microRNA-containing version compared to wild-type , hepatic morphology showed far less evidence of toxicity ( most sections appearing normal ) and the number of viral genomes found in liver was decreased by a factor of 25 . These findings are consistent with those using the E1A-luciferase reporter viruses , and suggest that inclusion of the mir-122-binding sites had a dramatic effect on hepatic activity and toxicity of the virus . It is worth noting that in this study the viruses were applied at dose in vivo ( 5×1010 vp/mouse ) , well above the lethal dose for wild-type Ad5 [16] , hence this regulatory strategy appears capable of controlling the activity of significant quantities of virus . Also of note is the ability of mir-122 in mouse liver to tightly regulate the very high levels of E1A-luciferase fusion protein achieved following hydrodynamic plasmid delivery ( Figure 3 ) , some 10-fold higher than those shown by the viruses . This suggests that the doses of virus used in this study do not even come close to exceeding the regulatory capacity of mir-122 . When even higher virus doses were applied , the maximum tolerated dose of Ad5-mir122 was estimated at between 6×1010 and 1 . 2e1011v . p/mouse ( 9×109–1 . 8×1010 pfu ) , and such doses presumably allow the virus to break through regulation by hepatic mir-122 . Nevertheless these doses are high , affording a range of doses where the virus may be applied therapeutically . A similar approach , using three microRNA binding sites , has recently been evaluated for regulating activity of CR2-deleted adenovirus in vitro [19] . However , the authors concluded that further attenuation was required in order to prevent CPE in Huh7 cells . The superior performance of the virus reported here may reflect the presence of four microRNA binding sites ( rather than three ) in the 3′ UTR although it is also possible that Huh7 cells have insufficient mir-122 to achieve the level of virus control seen in primary hepatocytes . MicroRNA-based virus regulation strategies should find a variety of applications in biotechnology . Their small size ( an individual site is typically 22 bp ) allows insertion of multiple binding sites , recognising diverse microRNAs , without compromising virus packaging efficiencies . In addition the small insertion size and typical proximity to essential virus genes and regulatory regions ( e . g the E1A poly A signal ) decreases the likelihood of propagating deletions . Hence a range of stable and versatile agents may be produced using this approach . Tissue-selective abrogation of virus replication to prevent unwanted pathology should find important applications in cancer virotherapy and also in the production of a new generation of attenuated vaccines for viral diseases . For example , introduction of binding sites for mir122 into the Hepatitis A , B or Hepatitis E genome should prevent replication in hepatocytes and abrogate the main viral toxicity , whilst maintaining infection and possible replication at other cellular sites . Such an approach could yield an important new range of therapeutic vaccines , with applications across the broad sphere of viral diseases .
All animal experimentation was performed in accordance with the terms of UK Home Office guidelines and the UKCCCR Guidelines for the Welfare of Animals in Experimental Neoplasia . Luciferase reporter plasmids sensitive to mir-122 were prepared by introducing concatamers of binding sites for mir-122 ( 4 or 8 sense or 4 antisense binding sites ) into the 3′UTR of the luciferase transcription cassette . A CMV-driven luciferase-expressing plasmid vector pCIKLux ( a kind gift from Dr Deborah Gill ) was cleaved with NotI , oligonucleotides were annealed at 95°C , cooled and ligated into dephosphorylated vector . This produced vectors pCMV-Luc-mir ( shown in Figure 1 ) , pCMV-Luc-mir122X8 and pCMV-Luc-mir122anti , together with the control ( hereafter referred to as pCMV-Luc ) which contained no mir-122 binding sites . The coding region for the C terminal half of E1A was PCR amplified using Accuprime PFX ( Invitrogen ) and primers ( forward ATT ATA AGA TCT GGA TAG CTG TGA CTC CGG TCC TTC , reverse TAT TCC ATG GAT GGC CTG GGG CGT TTAC ) using a plasmid containing wild-type Ad5 as template . These primers introduced unique BglII and Nco1 restriction sites to the 5′ and 3′ termini respectively . The purified PCR product was cleaved with BglII and Nco1 and cloned into pCMV-Luc and pCMV-Luc-mir described above , using the same enzymes , producing a fusion between the C terminal half of E1A and luciferase , including zero or four microRNA sites in the 3′ UTR . These products were subcloned using PshA1 and Hpa1 into a plasmid pAd5-Kpn1 ( produced by restriction of wild-type Ad5 , see below ) to produce plasmids ( pE1A-Luc and pE1ALuc-mir122 ) in which E1A was C-terminally fused to the luciferase coding sequence . The overall scheme of plasmid cloning is shown in Figure 1 . Wild-type Ad5 plasmid containing kanamycin resistance ( a kind gift from Dr Reuben Hernandez ) was cleaved with BstZ17I and recircularised by blunt ended ligation . This vector ( Ad5-BstZ17I ) was then further cleaved and re-ligated using Kpn1 to increase the number of unique restriction sites available for further cloning . This vector is referred to as Ad5-Kpn1 . The 4 microRNA binding sites for mir122 were PCR amplified from pCMV-Luc-mir ( described above ) to introduce Dra1 sites to each end . The purified PCR product was cleaved with Dra1 and blunt end ligated into Ad5-Kpn1 which was cleaved with Hpa1 in the E1A 3′ UTR . Insertion of microRNA binding sites downstream of E1A was confirmed by DNA sequencing . Ad5-Kpn1-mir122 was reconstituted to Ad5-BstZ17I using the Kpn1 gel-extracted fragment from Ad5-BstZ17I . To generate full size adenovirus genome Ad5-BstZ17I-mir122 was cleaved with BstZ17I , dephosphorylated and subject to homologous recombination with full size wild-type Ad5 ampicillin resistant vector ( a kind gift from Dr Peter Searle ) and selected on kanamycin . Insertion of microRNA binding sites was confirmed by sequence analysis . Restriction digestion of the resulting vector confirmed full size adenovirus had been recovered . pE1A-Luc-mir and pE1A-Luc ( which are modified forms of Ad5-Kpn1 described above ) were reconstituted to Ad5-BstZ17I using the Kpn1 gel-extracted fragments from Ad5-BstZ17I . To generate full size adenovirus genome Ad5-BstZ17I-E1ALuc-mir122 and Ad5-BstZ17I-E1ALuc were cleaved with BstZ17I , dephosphorylated and subject to homologous recombination with full size wild-type Ad5 vector ( a kind gift from Dr Peter Searle ) and selected on kanamycin . Insertion of microRNA binding sites and luciferase was confirmed by sequence analysis . Restriction digests of the resulting vectors confirmed full size adenoviruses had been recovered . Genomic structures and sizes of the viruses are shown in Figure 4A . All adenoviruses were grown in A549 cells , purified by double banding in CsCl gradients with benzonase treatment after the first banding . Viral particle ( vp ) number was determined by measuring DNA content using a modified version of the PicoGreen assay ( Invitrogen , Paisley , UK ) [20] . TCID50 calculated with the KÄRBER statistical method [21] was used to estimate the adenovirus titer ( TCID50 units/ml ) and corrected to determine plaque forming units/ml ( pfu/ml ) . Adenovirus preparations characteristics are as follows: Ad5 wild-type: 1 . 13×1012 vp/ml , 1 . 98×1011 pfu/ml and particle∶infectivity ( P∶I ) ratio of 5 . 6; Ad5-mir122: 1 . 29×1012 vp/ml , 2 . 01×1011 pfu/ml and particle∶infectivity ( P∶I ) ratio of 6 . 4 . All virus preparations were screened for endotoxin and verified negative prior to use . Human hepatocellular carcinoma HUH7 cells , A549 lung carcinoma cells , OVCAR3 ovarian cancer cells and HEK293 human embryonic kidney cells were obtained from the European Collection of Cell Cultures ( Porton Down , UK ) , and maintained in DMEM with 10% foetal bovine serum ( FBS ) ( PAA Laboratories , Yeovil , UK ) including penicillin ( 25 U/ml ) and streptomycin ( 10 mg/ml ) . Cells were seeded in triplicate in 12 well plates . After 24 h plasmid DNA ( 0 . 5 µg ) was added to 50 µl of HBS buffer and mixed with 2 . 5 µl DOTAP reagent ( Roche ) also in 50 µl sterile HBS . The complex was incubated at room temperature for 30 min . 100 µl of transfection mixture was added to each well and incubated at 37°C for 4 h . Cells were washed with PBS and incubated with DMEM containing 2% FBS . 24 h following transfection media were removed and 150 µl reporter cell lysis buffer ( Promega ) was added to the cells . Cells were then frozen at −80°C for >1 h before thawing . Luciferin ( 25 µl ) ( Promega , Southampton , UK ) was added to 25 µl aliquots of cell lysate and relative luminescence was measured by luminometry ( Lumat LB 9507 , Berthold Technologies , Redbourn , UK ) . A549 cells were seeded at 5×104 cells per well and incubated overnight . Pre-mir122 ( Ambion ) and pre-mir negative control ( Ambion ) were re-suspended to 50 µM and then further diluted 10 fold . 3 µl per well of this dilution of each pre-mir was added to 22 µl Opti-MEM medium ( Invitrogen ) . 2 µl per well of NeoFx transfection reagent ( Ambion ) was added to 23 µl Opti-MEM solution . Pre-mir/Opti-MEM was mixed with the NeoFx/Optimem and allowed to complex for 10 minutes . A549 cells were washed with PBS and the transfection mixture added to cells at a total volume of 50 µl . Total amount of pre-mir is 15 pmol/well . Immediately following transfection Ad-E1A-Luc-mir122 was added at 10 vp/cell in 450 µl DMEM media ( 10% FCS ) . 18 h later , 30 pmol/well of pre-cursor mir122 and negative control pre-cursor microRNAs were added to each well in addition to the 500 µl described above . Luciferase readings were performed at 24 h . The Q-PCR methodology for measurement of adenoviral particles has been previously described [22] . Viral DNA from infected cell or tissue samples were extracted using a mammalian genomic DNA miniprep kit ( Sigma ) . Reactions were performed using Applied Biosystems master mix following the manufacturer's protocol . The cycles were as follows: 94°C 10 min; 40 times ( 94°C 30 s , 60°C 1 min ) . Primers sequences for targeting Ad5 fiber are: FW- TGG CTG TTA AAG GCA GTT TGG ( Ad5 32350–32370 nt ) and RV- GCA CTC CAT TTT CGT CAA ATC TT ( Ad5 32433–32411 nt ) and the TaqMan probe- TCC AAT ATC TGG AAC AGT TCA AAG TGC TCA TCT ( Ad5 32372–32404 nt ) , dual labeled at the 5′ end with 6-carboxyfluorescein and the 3′ end with 6-carboxytetramethylrhodamine . The results were analyzed with the Sequence Detection System software ( Applied Biosystems ) . Standard curves for tissues and cells were prepared by spiking samples of cell lysate or tissue homogenate with serial dilutions of known concentrations of virus particles and then extracting and analysing each sample separately by Q-PCR as described above . Blood was taken from mice by cardiac puncture and allowed to clot ( 15 min , room temperature ) and spun at 1200 g for 10 min . Serum was isolated and immediately frozen at −20°C ) . Samples of thawed serum ( 5 µl ) were added to ALT reagent ( 995 µl , Microgenics ) or AST reagent ( 995 µl , Microgenics ) in a 1 ml quartz cuvette , incubated at 37°C and the change in absorbance ( 340 nm ) per minute was monitored . Units of ALT and AST activity were calculated according to the manufacturer's instructions . Plasmids were administered by hydrodynamic injection ( 0 . 8 pmole/mouse , using a 10% body volume of PBS administered over 5–10 s with a 27 gauge needle ) into the tail veins of Balb/c mice . Non-invasive measurement of luminescence was performed after 8 , 24 and 48 h using an IVIS 100 system ( Xenogen , MA ) under isofluorane anaesthetic . Luciferin was administered by intraperitoneal injection ( 15 . 8 mg/ml in PBS , 100 µl/mouse ) 4 min prior to imaging . Flux levels were analyzed with Living Image Software ( Xenogen , MA ) . Clodronate was a gift of Roche Diagnostics GmbH , Mannheim , Germany . It was encapsulated in liposomes as described previously [23] . Viruses were administered intravenously ( unless otherwise indicated ) and all animals were pretreated with bisphosphonate liposomes ( 100 µl/mouse , obtained from Dr Nico van Rouijen ) 24 h before . For imaging expression of E1A encoded within replication-competent Ad5 , E1A-luciferase reporter viruses with and without 4 binding sites for mir122 ( Ad5-E1A-luc and Ad5-E1A-luc-mir122 ) were injected intravenously to Balb/c mice ( 5×1010 v . p . /mouse ) . Animals were imaged after 6 , 24 , 48 , 72 and 96 h as described above . To study the ability of mir-122-binding sites included within wild-type A5 to decrease hepatic replication of virus genomes and tissue damage , 5×1010 v . p . /mouse of Ad5WT and Ad5-mir122 were injected i . v . Animals were monitored twice daily and sacrificed after 72 h for measurement of genome replication ( by QPCR ) and assessment of pathology ( by histological analysis ) . The left liver lobe from each mouse was immersed in 10% buffered formalin overnight at room temperature , embedded in wax and sectioned using a vibratome . Sections were stained with haematoxylin and eosin and analysed by light microscopy at ×40 magnification . In vitro data are expressed as the mean of 3 replicates ±standard deviation unless otherwise stated . In vivo data are expressed as the mean of four replicates ±standard deviation , except using the plasmid pCMV-Luc-Mir for which n = 3 . Significance was evaluated using t-test and denoted on the graphs as * P<0 . 05 , ** P<0 . 005 , *** P<0 . 0005 .
|
Attenuated viruses have found important applications in medicine , including their use as vaccines ( notably for measles , mumps , polio , influenza , and chicken pox ) and their experimental development as selective cancer-killing agents , so-called “virotherapy . ” Wild-type versions are often most effective in both of these settings; however , attenuated viruses have usually been developed to decrease the risk of significant viral pathology . Recent advances in understanding regulation of gene expression by microRNA now afford the possibility to design viruses that are “selectively attenuated” in sites of potential pathology , by engineering them for inhibition by microRNA molecules that are expressed there . Here we have engineered wild-type adenovirus for recognition by a microRNA expressed in hepatocytes , producing a virus that retains wild-type infection and replication at sites of therapeutic activity ( such as cancer cells ) but is severely attenuated in hepatocytes , both in vitro and in vivo . This virus caused no significant liver toxicity to mice even when applied at ten times the lethal dose of wild-type virus . The ability to produce replication-competent viruses with key toxicities removed should provide a new platform for development of improved cancer treatments and better vaccines for a broad range of viral diseases .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"microbiology/applied",
"microbiology"
] |
2009
|
Use of Tissue-Specific MicroRNA to Control Pathology of Wild-Type Adenovirus without Attenuation of Its Ability to Kill Cancer Cells
|
Rodents are reservoirs and hosts for several zoonotic diseases such as plague , leptospirosis , and leishmaniasis . Rapid development of industry and agriculture , as well as climate change throughout the globe , has led to change or increase in occurrence of rodent-borne diseases . Considering the distribution of rodents throughout Iran , the aim of this review is to assess the risk of rodent-borne diseases in Iran . We searched Google Scholar , PubMed , Science Direct , Scientific Information Database ( SID ) , and Magiran databases up to September 2016 to obtain articles reporting occurrence of rodent-borne diseases in Iran and extract information from them . Out of 70 known rodent-borne diseases , 34 were reported in Iran: 17 ( 50% ) parasitic diseases , 13 ( 38% ) bacterial diseases , and 4 ( 12% ) viral diseases . Twenty-one out of 34 diseases were reported from both humans and rodents . Among the diseases reported in the rodents of Iran , plague , leishmaniasis , and hymenolepiasis were the most frequent . The most infected rodents were Rattus norvegicus ( 16 diseases ) , Mus musculus ( 14 diseases ) , Rattus rattus ( 13 diseases ) , Meriones persicus ( 7 diseases ) , Apodemus spp . ( 5 diseases ) , Tatera indica ( 4 diseases ) , Meriones libycus ( 3 diseases ) , Rhombomys opimus ( 3 diseases ) , Cricetulus migratorius ( 3 diseases ) , and Nesokia indica ( 2 diseases ) . The results of this review indicate the importance of rodent-borne diseases in Iran . Considering notable diversity of rodents and their extensive distribution throughout the country , it is crucial to pay more attention to their role in spreading infectious diseases for better control of the diseases .
Rodents are the largest order of living mammals , comprising approximately 42% of global mammalian biodiversity [1 , 2] . With almost 2 , 277 known species in 33 families , rodents have nearly a worldwide distribution , being absent only from Antarctica and some isolated islands [2] . They are characterized by a peculiar dentition consisting of a single pair of continuously growing incisors in each of the upper and lower jaws and a set of chewing teeth [2 , 3] . Brandt ( 1855 ) on the basis of the zygomasseteric structure divided rodents into 3 suborders: Sciuromorpha , Hystricomorpha , and Myomorpha . This classification , even though widely accepted , has been also a matter of dispute [2 , 4 , 5] . Several years later , Wilson and Reeder adopted a 5-suborder system , i . e . , Sciuromorpha , Castorimorpha , Anomaluromorpha , Hystricomorpha , and Myomorpha , among which the last suborder is the biggest in terms of species richness and population numbers [2] . This suborder contains more than half of rodents’ species and almost a quarter of the identified mammalian species [1] . Rodents are small- to medium-sized mammals , with short reproductive cycle and large litters , as well as morphological and biological adaptations to different lifestyles ( e . g . , terrestrial , subterranean , gliding , etc . ) and environments ( e . g . , semiaquatic , aquatic , or dry biotopes ) . This high compatibility makes rodents one of the best suited mammals for living in various habitats [1 , 3] . In spite of rodents’ beneficial activities such as soil aeration , mineral nutrient cycling , increase in water absorption , facilitation of biotic recovery , and control of insect populations , they can cause significant economic losses ( primarily through feeding on stored food ) and increase health risk by transmitting various infectious agents to humans [6] . Indeed , rodents are well-known reservoirs and hosts for a number of infectious diseases ( e . g . , plague , leptospirosis , leishmaniasis , salmonellosis , and viral hemorrhagic fevers ) and play an important role in their transmission and spreading [7] . Rodent-borne diseases fall into one of two main categories: directly or indirectly transmitted diseases . In the former category , diseases are transmitted by being bitten or by inhaling the germ in feces of rodents , whereas in the latter category , humans are infected as the result of consuming food and water contaminated by rodent feces or urine . Likewise , rodents could act as amplifier hosts in the case of diseases transmitted by arthropod vectors from rodents to humans . Furthermore , rodents accidentally eaten by livestock could mediate disease transmission to humans if products of these livestock were not treated properly prior to consumption [8] . Iran is located between 25 to 40 degrees northern latitude and 44 to 63 degrees eastern longitude . Due to low latitude , aridity , and high fluctuation of daily and annual temperature , Iran has a variety of climate systems . Namely , the major mountain ridges ( Alborz and Zagros ) [9 , 10] along with sever climatic influences of the Caspian Sea in the north and the Persian Gulf in the south have led to the formation of four major climatic regions in the Iranian Plateau: mild and humid ( the southern beaches of the Caspian Sea ) , cold ( western mountains ) , hot and dry ( central part of the Iranian Plateau ) , and hot and humid ( southern seashores: Persian Gulf and Sea of Oman ) [11] . The diverse topography and different ecological conditions of Iran made it a corridor for the faunal exchanges between Asia and Europe [12 , 13] and at the same time a speciation zone for a number of rodents ( e . g . , Allactaga , Microtus , Mus ) [14–16] . Several studies have shown that rodents , particularly the suborder Myomorpha , have scattered widely in Iran [12 , 17 , 18] . So far , about 79 species of rodents have been identified in the country , with the following species being the most widespread: Allactaga elater , Jaculus blanfordi , Microtus socialis , Gerbillus nanus , Meriones crassus , M . libycus , M . persicus , Rhombomys opimus , Tatera indica , Apodemus witherbyi , Mus musculus , Nesokia indica , Rattus norvegicus , R . rattus , Dryomys nitedula , and Hystrix indica [18 , 19] . House mice ( M . musculus ) and rats ( R . rattus , R . norvegicus ) occupy various habitats at greater density than the other species and pose considerable problems [20 , 21] . Dipodidae and Gerbilinae dominate the arid and semiarid regions ( e . g . , A . elater , Jaculus spp . , T . indica , Gerbillus spp . , Meriones spp . , and R . opimus ) . Arvicolinae is the dominant group in the mountains , grasslands , cultivated areas , and river valleys in western Iran ( e . g . , species of Microtus , Arvicola , and Chionomys ) . The genus Apodemus occupies different habitats , reaching highest diversity in northern parts of the country [6 , 22–24] . On the other hand , occurrence of rodent-borne diseases have been documented in virtually all provinces of Iran [22] . Despite this , no attempt has been made to compare occurrence of these diseases together and assess the risk of each of these diseases . Therefore , the aim of this review is to assess the risk of rodent-borne diseases in Iran by reviewing the Iranian and international publications on occurrence of these diseases in rodents and humans throughout the country .
This study is a review article in which the articles indexed in Google Scholar , PubMed , Science Direct , Scientific Information Database ( SID ) , and Magiran databases were searched up to September 2016 . First , we browsed the databases to obtain articles that indicate which infectious diseases are rodent-borne , using keywords like “rodent-borne diseases , ” “rodent-borne pathogens , ” “mouse-borne diseases , ” and “rat-borne diseases . ” Then , rodent-borne disease names were extracted from identified literature [7 , 8 , 25–34] . Afterwards , we browsed the databases to obtain articles reporting occurrence of rodent-borne diseases in rodents and humans in Iran . Keywords were “extracted rodent-borne disease names , Iran , ” “extracted rodent-borne disease names , rodents , Iran , ” “bacteria , rodent , Iran , ” “parasite , rodent , Iran , ” “virus , rodent , Iran , ” “bacteria , mouse , Iran , ” “parasite , mouse , Iran , ” and “virus , mouse , Iran . ” In addition , references of the selected articles were also reviewed to increase the scope of search and to cover all the related articles . Rodent-borne diseases in terms of infectious agents of diseases are divided into 3 groups of bacterial , viral , and parasitic diseases; concerning each disease , data related to infectious agents of disease and history of the disease report in human and rodents in Iran were extracted from the articles . Eventually , data on reported and unreported diseases in Iran were written in tables , separately .
Results of our review showed that among 70 worldwide known rodent-borne diseases , 34 were reported from Iran , out of which 17 ( 50% ) , 13 ( 38% ) , and 4 ( 12% ) were parasitic , bacterial , and viral , respectively ( Tables 1 , 2 , 3 , and 4 ) . Among these diseases , 21 were reported in both humans and rodents , including Escherichia coli enteritis , salmonellosis , plague , yersiniosis , leptospirosis , campylobacteriosis , tularemia , tick-borne relapsing fever , tuberculosis , Crimean–Congo hemorrhagic fever , cryptosporidiosis , toxoplasmosis , leishmaniasis , hepatic capillariasis , trichinellosis , gongylonemiasis , hymenolepiasis , taeniasis , alveolar echinococcosis , trichuriasis , and moniliformiasis . Bartonellosis , babesiosis , and plagiorchiasis have been reported only in rodents , while 10 diseases—listeriosis , Lyme disease , Q fever , hepatitis E , rabies , hemorrhagic fever with renal syndrome , toxocariasis , giardiasis , schistosomiasis , and fascioliasis—have only been documented in humans . For 8 out of 34 diseases , rodents are known to be the primary or definitive host , including in plague , leptospirosis , tick-borne relapsing fever , Lyme disease , hemorrhagic fever with renal syndrome , leishmaniasis , hymenolepiasis , and moniliformiasis; meanwhile , in other reported diseases , rodents act as the secondary host . Of these 34 diseases , 11 of them—plague , tularemia , tick-borne relapsing fever , bartonellosis , Lyme disease , Q fever , Crimean–Congo hemorrhagic fever , leishmaniasis , babesiosis , schistosomiasis , and fasciolosis—not only are rodent-borne but also are vector-borne diseases . The other 23 diseases are only rodent-borne . Except plague , which has been only reported from the western part of Iran ( Kurdistan , Hamadan , East Azerbaijan ) , and Schistosomiasis , which has been only reported from the southwestern region of the country ( Khuzestan ) , the rest ( 32 diseases ) were reported from various regions of Iran . Out of 31 diseases reported from humans in Iran , 20 were reported frequently ( more than 10 reports ) , and 11 were scarcely reported ( fewer than 10 reports ) . The first category includes plague , E . coli enteritis , salmonellosis , leptospirosis , campylobacteriosis , tick-borne relapsing fever , Q fever , tuberculosis , hepatitis E , rabies , Crimean–Congo hemorrhagic fever , cryptosporidiosis , toxoplasmosis , leishmaniasis , hymenolepiasis , taeniasis , toxocariasis , schistosomiasis , giardiasis , and fasciolosis . The second category consists of yersiniosis ( 2 reports ) , tularemia ( 3 reports ) , listeriosis ( 2 reports ) , Lyme disease ( 5 reports ) , hemorrhagic fever with renal syndrome ( 1 report ) , hepatic capillariasis ( 1 report ) , trichinellosis ( 2 reports ) , alveolar echinococcosis ( 2 reports ) , moniliformiasis ( 4 reports ) , trichuriasis ( 3 reports ) , and gongylonemiasis ( 1 report ) . Out of 24 reported diseases among rodents of Iran , the 3 diseases plague , leishmaniasis , and hymenolepiasis were the most frequently reported . These diseases had more than 10 reports , while other diseases—E . coli enteritis , salmonellosis , yersiniosis , leptospirosis , campylobacteriosis , tularemia , tick-borne relapsing fever , tuberculosis , bartonellosis , Crimean–Congo hemorrhagic fever , cryptosporidiosis , toxoplasmosis , hepatic capillariasis , trichinellosis , taeniasis , alveolar echinococcosis , moniliformiasis , trichuriasis , gongylonemiasis , babesiosis , and plagiorchiasis—had fewer than 10 reports among rodents of Iran . Overall , based on the reviewed databases , 10 species of rodents in Iran are categorized as high-index infectious regarding the number of pathogens and diseases reported on them: R . norvegicus ( 16 diseases ) , M . musculus ( 14 diseases ) , R . rattus ( 13 diseases ) , M . persicus ( 7 diseases ) , Apodemus spp . ( 5 diseases ) , T . indica ( 4 diseases ) , M . libycus ( 3 diseases ) , R . opimus ( 3 diseases ) , C . migratorius ( 3 diseases ) , and N . indica ( 2 diseases ) .
This review showed that almost half of the known rodent-borne diseases ( 34 out of 70 ) so far have been reported in Iran , and out of the 34 diseases , 21 diseases were reported in both rodents and humans . Three diseases ( i . e . , bartonellosis , babesiosis , and plagiorchiasis ) were only reported from rodents and may be listed as hazardous for human communities , too . Ten diseases—rabies , hemorrhagic fever with renal syndrome , listeriosis , Lyme disease , Q fever , hepatitis E , toxocariasis , giardiasis , schistosomiasis , and fasciolosis—were reported only from humans . However , since both infectious agents and rodent hosts of these diseases are present in Iran [19 , 22 , 23] , rodents may act as possible mediators . Therefore , all of the reported rodent-borne diseases in Iran are important . Nevertheless , plague and leishmaniasis are of greatest concern because of their repeated reports in rodent and human populations , complicated transmission and maintenance cycle , and their pathogenesis that causes sever diseases in humans . In the case of plague , although in recent years occurrence of the disease has not been reported in human populations , the disease was reported repeatedly in the past in Iran , and so far , several big epidemics of human plague occurred in Iran . During 1772–1773 , one of the largest epidemics of plague in the world occurred in Iran and the area under control of Iran at that time . This epidemic led to the deaths of around 2 million people . Moreover , outbreaks of Plague were identified in rodents reservoirs , including in M . persicus , M . libycus , M . vinogradovi , and M . tristrami , in active foci of the plague in west Iran ( Kurdistan Province ) every 2–3 years [188] . Therefore , this disease should be monitored continuously in rodent populations of Iran , especially in western and northwestern areas , so that possible occurrence and emergence in human populations can be prevented . In the case of leishmaniasis , occurrence of the disease for the first time was reported in visceral form in 1949 , and since then the disease has been reported increasingly in different provinces of Iran . Currently , both forms of the disease ( Cutaneous Leishmaniasis [CL] and Visceral Leishmaniasis [VL] ) are endemic to Iran , and approximately 20 , 000 cases due to CL and 100–300 cases due to VL annually are recorded in Iran , although the actual number of CL may be 4 or 5 times higher [122 , 189] . In addition , rodents are the main reservoir for the wet ( rural ) CL in Iran , and this infection has been reported frequently and circulated among rodents of Iran , especially R . opimus , M . libycus , M . hurrianae , and T . indica [188] . Therefore , the rodent reservoir of the wet CL in each endemic province should be tackled carefully so that the prevalence of the diseases in humans can be decreased effectively . Although some of the rodent-borne diseases have not been thus far reported from Iran ( Table 4 ) , the rodent hosts [19 , 22 , 23 , 190 , 191] or intermediate hosts ( vectors ) of some of them ( e . g . , Rhipicephalus sanguineus , vector of Boutonneuse or Mediterranean spotted fever ) exist in Iran [192–194] . In addition to this , some of these diseases ( e . g . , lymphocytic choriomeningitis ) are occurring in Iran’s neighboring countries [195 , 196] . Also , nowadays human activity and climate change can affect spatial distributions; annual/seasonal cycles; incidence and severity of many infectious diseases , particularly zoonotic ones; and thus they can lead to occurrence of emerging infectious diseases throughout the globe [8 , 197] . It is , therefore , possible that those diseases eventually occur in rodents of Iran and can be transmitted to humans . Rodent-borne viral diseases include approximately one-third of all known rodent-borne diseases [8 , 198] , and it was shown that their transmission and prevalence vary in different regions depending on various virus–host systems; environmental regulators; and anthropogenic , genetic , behavioral , and physiologic factors [199] . In fact , some of these diseases are more frequently reported in certain regions of the world . For example , hantavirus pulmonary syndrome ( HPS ) in the Latin America region has been reported repeatedly both in humans and rodents , while hemorrhagic fever with renal syndrome is more prevalent in Eurasia . Also , lymphocytic choriomeningitis is endemic to Europe and has been reported in humans and rodents of this region , especially in M . musculus [28 , 200–204] . In Iran , most of the reported rodent-borne diseases have been parasitic or bacterial . Indeed , only 4 out of 34 diseases ( hepatitis E , Rabies , Crimean–Congo hemorrhagic fever , hemorrhagic fever with renal syndrome ) mentioned in this review have the viral agents . It seems that our current knowledge suffers from lack of relevant research of viral rodent-borne diseases in the country . Moreover , the tough effect of the global climate changes on the algorithm of viral zoonotic disease occurrence should be considered in future studies . The present review also showed that 11 diseases in Iran were vector-borne , including plague , tularemia , tick-borne relapsing fever , bartonellosis , Lyme disease , Q fever , Crimean–Congo hemorrhagic fever , leishmaniasis , babesiosis , schistosomiasis , and fasciolosis . In other words , almost one-third of the reported rodent-borne diseases in Iran are transmitted by vectors . Most of the vectors are ticks , fleas , mosquitoes , and sand flies [205–208] . Snails are mediators for schistosomiasis and fasciolosis [188 , 209] . It is therefore important to pay attention to ectoparasites of rodents , along with studying their own species . M . musculus , R . rattus , and R . norvegicus hold apparently the first rank regarding the rodent-borne diseases in Iran . Indeed , more than half of the reported diseases in Iran are connected to these commensal species . However , one should keep in mind that insufficient surveys have been performed on noncommensal rodent species in Iran [6] . M . persicus , Apodemus spp . , T . indica , M . libycus , R . opimus , C . migratorius , and N . indica held the second rank regarding reported diseases . Therefore , performing comprehensive studies on the prevalence of the diseases in all rodent species of the different regions of Iran is unavoidable . Indeed , it is necessary to study other rodent species as much as the commensals M . musculus , R . rattus , and R . norvegicus . Studying the evolution of host–parasite interactions at spatiotemporal scales is a necessary complement to the study and control of infectious diseases . For this purpose , it is crucial to accurately identify the species of both parasites and their hosts . Indeed , cryptic species may vary in their habitats and host preferences [210] , and misidentification of the species may lead to wrong interpretation of host–parasite interactions and their coevolution , with serious implications for human health [211] . Unfortunately , some of the previous studies on rodent-borne diseases in Iran suffer from species misidentification . For example , Apodemus sylvaticus , Apodemus flavicollis , and Mesocricetus auratus were repeatedly mentioned in the research performed on the rodent-borne studies of Iran [51 , 145] , while recent molecular studies revealed that these 3 species do not exist in Iran [13 , 212 , 213] . Moreover , not all the entities of undetected cryptic species complexes are harmful or cause trouble for human . Taxonomic identification is , therefore , economically important so that the resources will not be spent on nontarget species [210] . This purpose implicates the engagement of biosystematic integrative approaches such as molecular standard tools and morphological analyses . For instance , DNA barcoding has proven to be a good cut-off for many vectors’ delimitation of leishmaniasis [214] . Moreover , new advances in phylogenomic-scale sequence data ( e . g . , whole genome sequencing [WGS] , short-read sequence [SRS] , etc . ) are remarkably increasing in the case of clinical or taxonomical perspectives that produce an extraordinary resolution for difficult problems [215 , 216] . Recent papers indicate how much taxonomy and public health benefit from type-sequence analyses , and it seems that research can go much farther with full genomes of reference taxonomy isolates [217] . For instance , applying WGS approaches allows taxonomists to understand what makes each species unique , with important consequences for unbiased species delimitation . From a medical perspective , clinicians will be allowed to anticipate clinical symptoms of the cases and to screen for genes that are responsible for antimicrobial resistance ( AMR ) in human pathogens [217] . Therefore , new technologies will evidently revolutionize many disciplines ( e . g . , microbiology and taxonomy ) in the near future . On the other hand , it has been shown that phenotypic variation is associated with parasite infection [218] . It implies that morphological studies and the assessment of the evolutionary consequences of phenotypic trait variation on host populations should be included in the integrative studies of host–parasite interactions . Vast and rapid developments of industry and agriculture in the past century have caused an increase in food products and provided suitable shelters in urban regions for populations of urban mice in many developed and developing countries [6] . Moreover , as the ancient climate changes have left strong imprints on modern ecosystems , human-dominated recent climate changes will also affect the composition and distribution of biota in the future . Taking into account this possibility , any shift of the host species’ range in response to climate fluctuations may change the distribution of the parasites and the agents , which may tend to prominent alterations in trends of rodent-borne disease occurrence [8] . Thus , multidisciplinary collaboration among a vast range of experts ( e . g . , epidemiologist , rodentologist , entomologist , ecologist , and microbiologist ) is unavoidable in rodent-borne disease research for better understanding the current and future hazardous places for these diseases . In a nutshell , this review showed the importance of rodent-borne diseases in Iran , some of which have the potential to cause huge epidemics in human populations , such as plague . Therefore , it is necessary to seriously consider the role of rodents in spreading infectious diseases in Iran for their better control . Also , it is necessary to conduct more detailed and multidisciplinary studies on these diseases to better understand the occurrence of these diseases in Iran . Finally , the impact of climate change on the prevalence and distribution of the diseases , especially vector-borne diseases , should be studied so that the occurrence of emerging or re-emerging infectious diseases such as plague can be predicted and hopefully prevented .
|
This review showed that approximately half of the known rodent-borne diseases have been reported in Iran , half of which were reported both in humans and rodents . Most of the diseases were bacterial and parasitic . Plague , leishmaniasis , and hymenolepiasis were the most frequent diseases among rodent populations . Also , this review showed that among the rodent species , three commensal ones—R . norvegicus , M . musculus , and R . rattus—play an important role in the transmission of diseases to humans in Iran . Considering repeated reports of many of these diseases in humans and rodents , and the notable diversity and extensive distribution of rodents throughout Iran , it is crucial to pay adequate attention to rodents as a source of zoonotic infectious diseases in the country .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"plagues",
"atmospheric",
"science",
"geographical",
"locations",
"tropical",
"diseases",
"vertebrates",
"parasitic",
"diseases",
"animals",
"mammals",
"bacterial",
"diseases",
"review",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"climate",
"change",
"infectious",
"diseases",
"zoonoses",
"protozoan",
"infections",
"iran",
"people",
"and",
"places",
"rodents",
"eukaryota",
"climatology",
"hemorrhagic",
"fevers",
"asia",
"earth",
"sciences",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"amniotes",
"organisms"
] |
2018
|
Rodent-borne diseases and their public health importance in Iran
|
Shigella is the leading cause for dysentery worldwide . Together with several virulence factors employed for invasion , the presence and length of the O antigen ( OAg ) of the lipopolysaccharide ( LPS ) plays a key role in pathogenesis . S . flexneri 2a has a bimodal OAg chain length distribution regulated in a growth-dependent manner , whereas S . sonnei LPS comprises a monomodal OAg . Here we reveal that S . sonnei , but not S . flexneri 2a , possesses a high molecular weight , immunogenic group 4 capsule , characterized by structural similarity to LPS OAg . We found that a galU mutant of S . sonnei , that is unable to produce a complete LPS with OAg attached , can still assemble OAg material on the cell surface , but a galU mutant of S . flexneri 2a cannot . High molecular weight material not linked to the LPS was purified from S . sonnei and confirmed by NMR to contain the specific sugars of the S . sonnei OAg . Deletion of genes homologous to the group 4 capsule synthesis cluster , previously described in Escherichia coli , abolished the generation of the high molecular weight OAg material . This OAg capsule strongly affects the virulence of S . sonnei . Uncapsulated knockout bacteria were highly invasive in vitro and strongly inflammatory in the rabbit intestine . But , the lack of capsule reduced the ability of S . sonnei to resist complement-mediated killing and to spread from the gut to peripheral organs . In contrast , overexpression of the capsule decreased invasiveness in vitro and inflammation in vivo compared to the wild type . In conclusion , the data indicate that in S . sonnei expression of the capsule modulates bacterial pathogenesis resulting in balanced capabilities to invade and persist in the host environment .
Shigellosis , or bacillary dysentery , is an acute human inflammatory disease of the large intestine , characterized by watery diarrhea , fever , abdominal pain , and bloody and mucus stools , caused by Gram-negative Shigella enterobacteria [1] . This disease is a major global health concern , responsible for more than 7 million Disability-Adjusted Life Years and 100 , 000 deaths per year [2] , predominantly affecting children under 5 years of age from developing countries [3] . Deaths caused by shigellosis have been linked to intestinal but also systemic complications , including pneumonia , hypoglycemia , and hemolytic-uremic syndrome [4] . Shigella bacteremia is generally rare in adults and in individuals with no underlying condition but has been described in young children with frequencies up to 7% of cases [5 , 6] , with malnutrition being an important risk factor , or in immunocompromised individuals [7] and has been associated with high mortality rates [5 , 6 , 7 , 8] . Fifty Shigella serotypes belonging to the four serogroups of the genus ( S . dysenteriae , S . flexneri , S . boydii , S . sonnei ) are distinguished based on the structure of the O antigen ( OAg ) polysaccharide of the lipopolysaccharide ( LPS ) [9] . Among them , S . flexneri and S . sonnei are endemic and have been linked to most infections [10] . For both species , similar mortality rates [11] and frequencies of bacteremia per number of cases [5] have been reported . While S . flexneri is the most common cause of shigellosis , S . sonnei is replacing S . flexneri in locations where socio-economic conditions are improving and thus has become an important pathogen in developing countries [12 , 13] . In addition , S . sonnei bacteremia is likely to be underestimated as it is usually detected within the first 24 h of onset of disease when patients do not always seek medical attention [8] . S . sonnei comprises a single clonal group , characterized by low genetic variability and antigenic homogeneity [14 , 15] . All S . sonnei isolates have Phase I O somatic antigen , the immunodominant and protective antigen [16] . Phase I polysaccharide has an OAg repeating unit of two uncommon sugars not present in other Shigella serogroups , 2-acetamido-2-deoxy-L-altruronic acid ( L-AltNAcA ) and 2-acetamido-2-deoxy-L-fucose ( FucNAc4N ) [17] . For Shigella and genetically related E . coli species , LPS OAg biosynthesis is a Wzx/Wzy-dependent process , encoded by genes for synthesis of sugars of the repeating unit ( called wbg cluster in S . sonnei [18] and rfb cluster in S . flexneri [19] ) , and for OAg unit transport ( wzx ) and polymerization ( wzy ) on the independently synthesized LPS core-lipid A moieties [9] . Unlike in other Shigella , the S . sonnei wbg OAg synthesis cluster is not located on the chromosome but on the S . sonnei large virulence plasmid ( pSS ) [20] . To invade and colonize the intestinal epithelium and to survive the strong inflammatory host response , Shigella requires expression of protein factors encoded by the virulence plasmid , such as the Type III Secretion System ( T3SS ) and its secreted effectors [1] . The LPS is also a key virulence determinant [21] . S . flexneri ( 2a and 5a ) has LPS OAg with a bimodal chain length distribution which is regulated in a growth-dependent manner [22] and is important for bacterial mobility and serum resistance [23] . Moreover , phage-encoded glucosylation of the OAg is essential for optimized LPS and T3SS functions in S . flexneri 5a M90T [24] . Less is known about S . sonnei LPS: Phase I bacteria possess a single modal OAg with a predominant chain length of 20–25 units [18] . Expression of Phase I polysaccharide and virulence are strongly interconnected and loss of the pSS virulence plasmid in vitro results in the S . sonnei Phase II cell type , lacking both the OAg and the virulent phenotype [25 , 26] . Besides the OAg side chain of LPS , Gram-negative exopolysaccharides generally include other structures , e . g . capsules . These enhance the bacterial survival in the environment and their fitness within hosts , by avoiding elimination by innate immune killing [27] . E . coli capsules have historically been classified into four groups based on genetic and biochemical criteria [28] . Group 4 capsules ( G4C ) are comprised of a high molecular weight surface polysaccharide and are also known as ‘O antigen capsules’ due to their structural similarity to the OAg side chain of the LPS [28] . In a more recent classification distinguishing two major groups based on the primary mechanisms of biosynthesis , i . e . the ABC transporter-dependent and the Wzy-dependent capsular polysaccharides [29] , group 4 capsules , together with group 1 capsules , belong to the Wzy-dependent group . OAg capsules have been found in intestinal pathogenic E . coli , such as enteropathogenic ( EPEC ) [30] and enterohemorrhagic E . coli ( EHEC ) [31] , as well as in Salmonella enterica serovar Enteritidis [32] , and have been shown to confer enhanced colonization [31] and environmental persistence [32] . G4C share the Wzy-dependent synthesis cluster for OAg units with the LPS , but require an additional g4c operon for secretion and assembly of the OAg polysaccharide from the periplasm into the capsular structure [28] . All seven genes of the E . coli g4c transcriptional unit ( ymcDCBA , yccZ , etp , etk ) were required for capsule production in EPEC serotype O127 [30] . Despite Shigella being classified as uncapsulated bacteria , genes homologous to the E . coli g4c operon are present in different strains [30] , but expression of G4C as component of Shigella exopolysaccharide and its potential contribution to pathogenicity have not been described . In this study , we show that S . sonnei , but not S . flexneri 2a , possesses a g4c operon-dependent OAg capsule . Deletion of the capsule in S . sonnei results in substantially increased invasiveness of HeLa cells in vitro and triggers increased inflammation in the intestine but decreases resistance to complement-mediated killing and spreading ability in the rabbit model . Thus , the S . sonnei G4C is an important factor in the regulation of pathogenesis and persistence .
By SDS-PAGE ( Fig . 1A ) , the phenol-water extract of S . sonnei WT had a typical LPS ladder with a predominant chain length of 20 to 25 OAg repeating units [18] and additional lower mobility material above the position of 25 units . S . sonnei strain lacking the pSS OAg-encoding virulence plasmid ( S . sonnei -pSS ) had only the low molecular weight band corresponding to the LPS core-lipid A moieties . S . sonnei ΔgalU is a deep rough LPS mutant with a defect in the pathway that transfers the OAg onto the LPS core region [21] . Its extract had slowly migrating material in addition to the low molecular weight band of the LPS inner core-lipid A molecules , but no LPS ladder . A Western blot probed with a monovalent anti-S . sonnei Phase I typing serum displayed a band with similar low mobility in phenol-water extracts of S . sonnei WT and S . sonnei ΔgalU but not in S . sonnei -pSS extract ( Fig . 1A ) . To immunize mice , we generated outer membrane particles , called Generalized Modules for Membrane Antigens ( GMMA ) , from S . sonnei WT , S . sonnei ΔgalU and S . sonnei-pSS strains genetically modified to induce high level shedding of these particles ( hyperblebbing ) by deletion of the tolR gene [33] . GMMA are highly immunogenic and present surface antigens in their natural context . By flow cytometry ( Fig . 1B ) , sera raised against S . sonnei GMMA with unmodified LPS reacted with S . sonnei WT and S . sonnei -pSS bacteria . Sera raised against S . sonnei -pSS GMMA reacted with S . sonnei -pSS but not with S . sonnei WT , indicating that the OAg on the target S . sonnei WT bacteria shields the surface antigens from the antibodies raised against OAg-negative GMMA . Sera raised against S . sonnei ΔgalU GMMA reacted like sera raised against S . sonnei GMMA with unmodified LPS , staining S . sonnei WT and S . sonnei -pSS bacteria , suggesting that they had anti-OAg activity despite lacking LPS-linked OAg . This immunoreactivity was not an artifact of GMMA immunization , as sera raised against whole formalin-fixed S . sonnei WT , S . sonnei with a knockout of the wbg OAg biosynthesis cluster in the pSS virulence plasmid ( S . sonnei ΔOAg ) , and S . sonnei ΔgalU bacteria gave similar results ( S1 Fig ) . Unlike S . sonnei ΔgalU , sera raised against whole inactivated S . flexneri 2a ΔgalU did not react with its homologous S . flexneri 2a WT strain , indicating the lack of OAg-specific immunogenicity in this background ( S1 Fig ) . We performed a competitive staining experiment ( Fig . 1C ) to test if OAg-specific antibodies in S . sonnei ΔgalU GMMA antisera were responsible for the binding to S . sonnei WT . Pre-absorption of anti-S . sonnei ΔgalU GMMA sera with phenol-water extract from S . sonnei WT but not from S . sonnei -pSS bacteria abolished reactivity ( Fig . 1C ) . In addition , flow cytometry of formalin-fixed S . sonnei WT , ΔgalU and -pSS bacteria stained with the S . sonnei Phase I typing antiserum gave strong signals with the WT and the ΔgalU strains , while no binding was revealed on the control S . sonnei -pSS strain ( Fig . 1D ) . These results confirm that S . sonnei ΔgalU expresses an OAg material on the surface . However , the ΔgalU OAg may be loosely attached to the outer membrane since the anti-Phase I surface staining of live S . sonnei ΔgalU target bacteria was variable and much less pronounced ( S1 Table ) . Genomic analysis showed a gene cluster in S . sonnei 53G and 046 with ≥ 99% identity to the E . coli g4c operon in the coding regions and the same genetic orientation ( ymcDCBA , yccZ , etp , etk ) ( S2 Fig ) . Homologous genes are also present in S . flexneri 2a 2457T and 301 genomes , but a deletion of 14 bases ( TGTCGCTTACTCGC ) in the etk locus ( position 135 to 148 ) causes a frame-shift mutation and thus inactivation of the operon ( S2 Fig ) . A S . sonnei g4c mutant strain ( S . sonnei Δg4c ) was generated to assess the functionality of the cluster . Furthermore , we inserted a selectable marker into the OAg-encoding pSS ( replacement of virG by a resistance gene ) to avoid loss of the plasmid during in vitro culture , and thereby obtained S . sonnei strains with stable OAg expression . These modifications were also introduced into hyperblebbing strains with the aim to use GMMA as source of exopolysaccharides for biochemical analyses . GMMA mainly contain outer membrane components and have negligible amounts of nucleic acids and cytoplasmic impurities [33] . We found acid-cleaved phenol-water exopolysaccharide extracts from GMMA to be suitable for polysaccharide analysis without need for further purification . Exopolysaccharide molecular weight distribution was examined by HPLC-SEC ( Fig . 2A ) . The Phase I exopolysaccharide ( Fig . 2A , solid line ) from S . sonnei with unmodified LPS possessed a trimodal distribution of low , medium and high molecular weight ( LMW , MMW , HMW , respectively ) polysaccharides . A single LMW polysaccharide population was obtained from S . sonnei ΔOAg exopolysaccharide ( Fig . 2A , dashed line ) . Exopolysaccharide from S . sonnei Δg4c had similar MMW and LMW populations as the S . sonnei Phase I exopolysaccharide , but no HMW polysaccharide ( Fig . 2A , dotted line ) . Silver stained SDS-PAGE of the corresponding phenol-water GMMA extracts ( Fig . 2B ) correlated with the HPLC-SEC analysis but did not differentiate between the patterns from S . sonnei with trimodal WT exopolysaccharide and from S . sonnei Δg4c deficient for the HMW polysaccharide . In S . sonnei ΔgalU exopolysaccharide , only the HMW polysaccharide and a very LMW polysaccharide were present ( S3 Fig ) . The Phase I exopolysaccharide was fractionated by HPLC-SEC and eluted fractions were analyzed by 1H-NMR ( Fig . 2C ) . Spectra of the HMW and the MMW polysaccharides had peaks from the FucNAc4N and the L-AltNAcA residues of the S . sonnei OAg [34] . Integration of these signals confirmed the 1:1 ratio between FucNAc4N and L-AltNAcA expected for the S . sonnei OAg units . Anomeric signals of terminal and internal α-galactose residues from the outer core of S . sonnei LPS ( 5 . 82 ppm and 5 . 62 ppm , respectively ) [34] were detected in the MMW polysaccharide . From the ratio between signals of the N-acetyl groups of the OAg residues ( 2 . 00–2 . 04 ppm ) and signals of the outer core α-galactose residues ( 5 . 82 and 5 . 62 ppm ) , we calculated that the MMW polysaccharide has about 23–30 OAg units , in line with the SDS gels and literature values for the main LPS population [18] . The size of the HMW polysaccharide was estimated by HPLC-SEC in comparison with a dextran standard curve at 206–269 OAg units . To assess the presence of LPS core in the different polysaccharide populations , we analyzed the content of core residue 3-deoxy-D-manno-octulosonic acid ( KDO ) after acid hydrolysis at the reducing end [35] . KDO was detected in the LMW and MMW polysaccharide , but not in the capsular HMW polysaccharide ( S4 Fig ) . Thus , the g4c operon is functional in S . sonnei and encodes for the formation of an LPS-unlinked high molecular weight OAg polysaccharide , i . e . a group 4 capsule . Transmission electron microscopy of alcian blue stained S . sonnei WT displayed a dark layer of electron-dense material corresponding to the exopolysaccharides ( Fig . 3 ) . Immunogold staining using anti-Phase I serum localized the Phase I antigens on the external surface of the outer membrane , protruding about 10–20 nm , and within the periplasmic space . In the absence of the G4C , S . sonnei Δg4c bacteria possessed an OAg layer of about half of the thickness of the S . sonnei WT layer . G4C expression in S . sonnei Δg4c was restored by complementing the knockout in trans through the insertion of a functional operon with its own promoter on a low copy number vector . In these S . sonnei Δg4c ( g4c ) bacteria , the thickness of the Phase I layer was augmented , extending from the outer membrane about 20–30 nm . The control S . sonnei -pSS OAg-deficient strain was negative for both the alcian blue and the immune labelling , confirming staining specificity . The invasiveness of S . sonnei WT and S . sonnei Δg4c was investigated in vitro by evaluating the number of intracellular bacteria after 1 h infection in HeLa cells ( Fig . 4A ) . Uncapsulated S . sonnei Δg4c had about 100-fold more colony forming units ( CFU ) compared to the WT strain . Complementation of the g4c knockout in S . sonnei Δg4c ( g4c ) reduced invasiveness of S . sonnei Δg4c by about 400-fold , resulting in 4-fold less intracellular bacteria than with S . sonnei WT . S . sonnei -pSS lacking the virulence plasmid did not invade HeLa cells , as expected [25] . These results show that the expression level of the capsule polysaccharide affects S . sonnei cell invasion in vitro . Thus , we investigated if the G4C masks critical elements of the invasion complex . Using flow cytometry ( Fig . 4B ) we examined the ability of a monoclonal antibody to bind to the invasion plasmid antigen B ( IpaB ) located at the tip of T3SS [36] . IpaB-dependent fluorescence on the surface of S . sonnei WT and S . sonnei Δg4c ( g4c ) was only detected at a level comparable to the negative control . The anti-IpaB signal was stronger on the capsule-deficient S . sonnei Δg4c strain . By Western blot , similar amounts of IpaB protein were detected in S . sonnei WT and S . sonnei Δg4c ( S5 Fig ) . Thus , the T3SS tip is more accessible in the absence of the capsule . IpaB was not detected by Western blot in S . sonnei Δg4c ( g4c ) suggesting an interference of the excess of capsule material either with IpaB detection or with IpaB expression . The anti-IpaB staining on S . sonnei ΔOAg , lacking both the LPS OAg side chain and the G4C but still possessing the virulence plasmid , was further increased over the staining on uncapsulated S . sonnei Δg4c . S . sonnei -pSS was negative , as expected . The impact of the capsule on S . sonnei virulence was tested in the rabbit ligated ileal loop model . Separate intestinal loops were infected with equal numbers ( 3x109 ) of S . sonnei WT , capsule-deficient S . sonnei Δg4c , complemented S . sonnei Δg4c ( g4c ) , or S . sonnei -pSS per loop ( 3 replicate loops per animal for each strain ) . Eight hours after infection , local sites were examined for Shigella-dependent pathology , while peripheral sites were examined for the relative presence of the different S . sonnei strains . In the intestine , fluids and blood were present in the S . sonnei WT-infected loops . S . sonnei Δg4c caused more fluid and blood accumulation , with loops having pale outer surface and intense inflammation . The pathology observed with the complemented S . sonnei Δg4c ( g4c ) strain was less severe than that with the WT , whereas the non-invasive S . sonnei -pSS barely caused fluid production and tissue alteration . Histopathology supported the qualitative data ( Fig . 5A ) . Hematoxylin and Eosin ( H&E ) stained slices showed alterations of villi from infected tissues and bacteria were detected by immunostaining with a polyclonal anti-S . sonnei GMMA serum . After infection by S . sonnei WT , villi were shortened and enlarged , with an average length to width ratio ( L/W ) of 5 . 3 ( Fig . 5B ) , with several indentations . Numerous regions of tissue disruption were observed and infiltration of inflammatory cells was detected within the lamina propria and in the edematous submucosal tissues ( Fig . 5C ) . These observations resulted in an inflammation score of 4 . 6 , according to the Ameho criteria with grading scores from 0–6 [37] ( Fig . 5D ) . Infection by S . sonnei Δg4c led to dramatic alterations of mucosal tissues with extensive zones of rupture and destruction of the intestinal epithelium , including epithelial detachment and loss of villi with tissue necrosis . Remaining villi had a lower L/W ratio ( 3 . 6 ) than villi in S . sonnei WT-infected loops ( Fig . 5B ) . Submucosal tissues were strongly edematous ( Fig . 5C ) , with a large area infiltrated by inflammatory cells between the residual mucosa and the muscular layer . These observations were graded as a very high 5 . 4 Ameho score ( Fig . 5D ) . In accordance with the profound epithelial changes , following infection with both S . sonnei WT and S . sonnei Δg4c , most bacteria were associated with the lamina propria and the epithelium of the villi , particularly in areas of abscesses , rupture/destruction of the epithelial lining and of villi indentation ( Fig . 5A ) . In contrast , slices from S . sonnei Δg4c ( g4c ) -infected tissues showed a lower level of pathology ( Ameho score of 1 . 9 ) , with villi with a length to width ratio ( L/W of 7 . 0 ) similar to those following infection with S . sonnei -pSS or uninfected ( L/W of 7 . 7 and 8 . 4 , respectively ) , limited edema with few cells infiltrating the lamina propria , and few tissue lesions . The reduced severity of tissue damage in H&E-stained slices correlated with most of the S . sonnei Δg4c ( g4c ) bacteria being present in the lumen instead of in the epithelium . Villi of loops infected with the avirulent S . sonnei -pSS strain were long and narrow , similar to the normal rabbit epithelial architecture of uninfected tissue , with no evidence of mucosal alteration , a low inflammation score of 1 . 1 , and no invasive bacteria detected within the epithelium . Pro-inflammatory cytokines were measured to further analyze the inflammatory response of the tissue in the infected loops . In an initial experiment to establish the infectious dose of S . sonnei in the rabbit ligated loop model , 5x109 bacteria per loop were used . At this dose , the S . sonnei Δg4c capsule-deficient strain induced , on average , higher levels of Interleukin 8 ( IL-8 ) , IL-6 , and IL-1β compared to S . sonnei WT ( Fig . 5E ) but the number of replicate loops tested ( 2 each in 2 rabbits ) was too small to reach statistical significance . The high induction of pro-inflammatory cytokines was accompanied by severe tissue destruction . Thus , for the main experiment a slightly lower infectious dose ( 3x109/loop ) was chosen . At this dose , in general lower inflammation was observed . While the histology assessment showed a difference in the inflammatory potential of S . sonnei WT and Δg4c as described above , no difference was detected in the induction of pro-inflammatory cytokines , possibly since , at the lower dose , lesions are more dispersed and the analysis is performed on whole loops so that the results from the infected tissue are hidden by the results from the normal tissue . In the same model used for assessing the Shigella-induced histopathology , we then investigated how the G4C contributes to S . sonnei dissemination by evaluating the bacterial load in mesenteric lymph nodes , spleen , liver , or blood . Eight hours after infection of separate intestinal loops in the same animal with S . sonnei WT , S . sonnei Δg4c , S . sonnei Δg4c ( g4c ) , or S . sonnei-pSS the number of bacteria from these strains in the systemic organs was determined ( Fig . 6A ) . In each organ S . sonnei WT bacteria predominated ( on average 70% of the bacteria recovered in the specific organ ) . S . sonnei Δg4c bacteria accounted for 26% of bacteria per organ , and only a negligible number of S . sonnei -pSS and S . sonnei Δg4c ( g4c ) bacteria was found peripherally . Sensitivity to complement was assessed ( Fig . 6B ) as a measure for the ability of S . sonnei WT and uncapsulated S . sonnei Δg4c to survive in the systemic environment . S . sonnei WT was highly resistant to complement: even in 90% baby rabbit complement , the number of cells increased to 22-fold of the inoculum size during 3 h incubation ( approximately 4 . 5 generations ) . In contrast , in 50% baby rabbit complement the number of cells of the capsule-deficient strain increased to only 3-fold of inoculum size and 90% baby rabbit complement resulted in killing of the mutant to a viable cell count of 10% of the inoculum . S . sonnei -pSS and S . flexneri 2a served as controls . As previously reported [38 , 39] , S . sonnei -pSS was not able to survive in 50% complement ( Fig . 6B ) , and S . flexneri 2a was more sensitive to complement than S . sonnei WT ( S6 Fig ) . Thus , while the OAg is essential for resistance to serum lysis [38] the capsule strongly augments S . sonnei resistance to direct complement-mediated killing . Collectively , these data show that the lack of capsule increases S . sonnei local pathogenicity but reduces peripheral dissemination , likely due to reduced complement resistance . Conversely , overexpression of capsule ( S . sonnei Δg4c ( g4c ) ) decreased dissemination , presumably because these bacteria failed to invade epithelial cells effectively .
The presence of a group 4 capsule in S . sonnei was uncovered by the immunological characterization of the S . sonnei galU knockout deep rough mutant that does not possess an acceptor for the OAg polysaccharide on the LPS molecule . We demonstrated that this mutant still has the ability to assemble OAg units on the surface through an alternative pathway to form an OAg capsule , in accordance with the presence of an intact wbg OAg ( Wzy-dependent [29] ) synthesis cluster on the S . sonnei virulence plasmid , essential for both LPS OAg and OAg capsule . A low mobility material , not linked to the LPS , was extracted from S . sonnei WT and ΔgalU bacteria and was visualized in immunoblot analyses with a Phase I-specific antiserum . This observation was similar to the identification of a slowly migrating Phase I-reactive material in S . sonnei and in a Phase I-transformed Salmonella Typhi live vector vaccine candidate [18] . Although an OAg capsule-like material in S . sonnei has been suggested [40] , no further studies have clarified this hypothesis . We found that SDS-PAGE analyses could not differentiate between LPS alone and LPS plus capsule , with both running with a smearing component . Thus , in the present study the identity of the S . sonnei G4C was confirmed by genetic and structural analyses . Genes homologous to those identified in E . coli for the G4C were found in different Shigella genomes [30] . In agreement with this finding , S . sonnei strains 53G and 046 were confirmed to have an intact g4c operon and its functionality in S . sonnei 53G was demonstrated . In contrast , our genomic analysis revealed that S . flexneri 2a strains 2457T and 301 carry an inactivated operon due to a deletion in the etk locus . Consequently , in S . flexneri 2a the ΔgalU mutation resulted in OAg deficiency as previously reported [21] , as the strain lacks the capsule assembly system that would provide an alternative pathway for OAg surface expression to LPS . We studied the biochemical characteristics of the S . sonnei G4C extracted from GMMA since these outer membrane preparations contain very low amounts of DNA and cytoplasmic components . We used G4C positive and negative isogenic mutants from hyperblebbing S . sonnei strains with stabilized virulence plasmid-driven expression of the OAg to obtain highly purified exopolysaccharide and demonstrated that the g4c cluster encodes for the formation of a high molecular weight polysaccharide with about 10 times higher molecular weight than the main medium molecular weight LPS population . Phase I antigen comprising both , LPS OAg and OAg capsule , is likely to be a common feature in the S . sonnei serogroup . For instance , a similar exopolysaccharide profile to S . sonnei 53G was found in a different S . sonnei isolate ( 25931 , S2 Table ) . The LPS OAg and the capsular OAg share the same biosynthesis cluster . Accordingly , mutants in the wbg OAg operon ( S . sonnei -pSS and S . sonnei ΔOAg ) were both rough and uncapsulated . LPS core residues were found in the LPS-derived medium and low molecular weight polysaccharides , but not in the capsular polysaccharide , suggesting again that this population is not a high molecular weight LPS species , but is linked to the bacteria independently of the LPS . The S . sonnei Phase I polysaccharide has an uncommon zwitterionic structure [41] of sugars carrying a carboxylic group ( C-4 of the FucNAc4N ) and an amino group ( C-6 of the L-AltNAcA ) [9] . Given the presence of amino groups in the OAg units , fixation with formaldehyde is likely to cross-link the capsule and to bind it to LPS OAg and to surface proteins , making its detachment from the bacteria more difficult , as it was shown in EM and flow cytometry analyses . However , we found G4C bound to unfixed bacteria and on GMMA indicating that S . sonnei G4C is not completely released by bacteria . Therefore , the presence of a linker molecule could be envisaged . Structural analyses of G4C from Salmonella Enteritidis have shown that purified capsule has fatty acids at levels consistent with a lipid anchor [42] . Further studies are needed to test if the S . sonnei capsule is linked to the surface through such an anchor . The current model of Shigella pathogenesis is mainly derived from studies of S . flexneri [1] . S . flexneri has a bimodal LPS OAg distribution including the so-called very long OAg that contributes to virulence in S . flexneri 2a [23] and is absent in S . sonnei [18] . Instead , as S . sonnei but not S . flexneri 2a was found to be encapsulated , we investigated the role of G4C as a S . sonnei specific virulence factor . We demonstrated that S . sonnei G4C , together with the LPS OAg , constitutes a dense outer layer that impacts on S . sonnei surface accessibility and invasive potential . For example , the exopolysaccharide layer masked antibodies raised against OAg-lacking outer membrane GMMA particles from recognizing their target epitopes in vitro . Bacterial capsules have been characterized to modulate the functionality of other virulence factors , such as adhesins in E . coli [43] , fimbriae in Klebsiella [44] and pili in Neisseria [45] . G4C , in particular , have been shown to play critical roles in interactions between bacteria and their immediate environments [32] and hosts [31] . In pathogenic E . coli , G4C appeared to mask surface structures , inhibit the attachment of bacteria to tissue-cultured epithelial cells , diminish their capacity to induce the formation of actin pedestals , and attenuate T3SS-mediated protein translocation into host cells [31] . Similarly , in S . sonnei the presence of the capsule polysaccharide accounted for changes in cell invasion ability in vitro . Uncapsulated S . sonnei Δg4c were significantly more invasive than S . sonnei WT , while capsule overexpressing S . sonnei Δg4c ( g4c ) strain displayed reduced cell entry . Thus , the in vitro studies indicated a role of the G4C in negatively affecting the invasive abilities of S . sonnei . The virulence of Shigella is mainly mediated by the activity of an array of plasmid-encoded virulence factors among which the T3SS is one of the most important [1] . We evaluated the possibility of capsule-mediated shielding of T3SS in S . sonnei . We observed that the accessibility of the IpaB protein at the tip of the T3SS was increased in vitro in S . sonnei Δg4c compared to the WT strain . As similar amounts of IpaB were detected by Western blot this showed that the G4C at least partially covers the tip of the T3SS . The importance of a balance between surface LPS OAg length and T3SS exposure in Shigella has been demonstrated for S . flexneri 5a M90T , which has evolutionarily acquired a phage-encoded OAg glucosylation , reported to have optimized the length of the OAg for T3SS function without compromising the protective properties of the LPS [24] . Shielding of T3SS by a G4C has also been found in EPEC and EHEC [31] . Still , expression of G4C has been shown to be required for efficient host colonization by EHEC in vivo , and to be inversely regulated to T3SS expression during different stages of pathogenesis [31] . Therefore , we used the rabbit model of experimental shigellosis to investigate G4C contribution to pathogenicity of S . sonnei in vivo and to the induction of inflammatory host responses during the infection . Analysis of infected rabbit tissues indicated that the lack of the capsule in S . sonnei Δg4c caused a dramatic increase of pathogenicity and inflammatory potential in the gut , particularly characterized by the augmentation of mucus and blood production in the infected loop , rupture and destruction of the epithelial lining , and tissue inflammatory manifestations as determined by villi atrophy , submucosal edema , and Ameho grading . In addition , induction of cytokine production of the infected epithelium was higher in S . sonnei Δg4c-infected loops than in S . sonnei WT-infected loops at the infectious dose of 5x109 bacteria/loop . We hypothesize that in uncapsulated bacteria , virulence factors are not shielded by the capsule polysaccharide and this facilitates bacterial adhesion and entry in the host cells . Concomitantly , the recognition of invasive bacteria by the innate immune system of the host could be enhanced in the absence of the capsule , as demonstrated for Neisseria [46] , further augmenting the inflammatory response . In contrast , hyper-encapsulated S . sonnei Δg4c ( g4c ) strain showed an attenuated phenotype , with less bacteria invading the epithelium and less tissue inflammation compared to WT . We cannot exclude that in addition to the larger amount of capsule other factors contribute to the attenuated phenotype as we could not detect IpaB by Western blot; and it is not clear if this is a technical artefact due to interference of the capsule material with the immunodetection or a true lack of IpaB in S . sonnei Δg4c ( g4c ) . Moreover , the reduced pathogenicity of G4C-overexpressing S . sonnei Δg4c ( g4c ) could be related not only to reduced invasiveness but also to higher host tolerance to encapsulated bacteria , as was shown for Salmonella serovars expressing Vi capsule [47] . In contrast to the stronger pathogenicity in the gut , lack of G4C reduced the ability of S . sonnei to spread to systemic infection sites . This was in accordance with enhanced complement sensitivity of S . sonnei Δg4c compared to the WT . Similarly , the presence of the very long OAg was shown to be important for resistance to direct complement-mediated serum killing in S . flexneri 2a [48] . Complement resistance could play a role at the stage of inflammation in epithelial lesions involving complement recruitment [48] and at the stage of systemic disease which is increasing in children and immunocompromised patients [7 , 8] . Therefore , the capsule could be an important virulence factor for S . sonnei to survive host killing , both locally and systemically , and might play a similar role in infection as the very long OAg in S . flexneri 2a . Capsule deficiency could be beneficial in the early stages of the infection for proficiently invading the intestinal epithelium , but with a disadvantage in translocation . If G4C expression in S . sonnei is regulated during pathogenesis as the G4C in EHEC [31] , or in a growth-dependent manner as the very long OAg in S . flexneri [22] remains to be addressed . In conclusion , in S . sonnei expression of the capsule modulates virulence and results in a successful phenotype in vivo , with balanced capabilities to invade and persist in the host environment . Based on the substantial changes in pathogenesis observed upon deletion or overexpression of the S . sonnei capsule , we hypothesize that its level of expression may be under significant evolutionary pressure .
Shigella strains used in this work are listed and described in Table 1 . All S . sonnei strains are derivatives of S . sonnei 53G [49] ( S . sonnei WT ) , all S . flexneri 2a strains are derivatives of S . flexneri 2a 2457T [50] . The S . sonnei Phase II colony ( S . sonnei -pSS ) was isolated during in vitro cultivation of S . sonnei WT bacteria on TSB Congo red agar plates by screening for spontaneous loss of pSS [33] . The experimentally induced S . sonnei -pSS nalidixic acid resistant ( NAR ) strain was isolated by serial passages of S . sonnei -pSS on LB agar plates supplemented with increasing concentrations of nalidixic acid ( from 10 to 50 μg/mL ) . S . sonnei ΔgalU was obtained using the same 3-step PCR method and materials as for generation of S . sonnei ΔtolRΔgalU [33] . OAg deletion in S . flexneri 2a was performed as previously described [51] . For generating the other knockout mutants , the chloramphenicol resistance gene ( cat ) from pKOBEG [52] was used to replace S . sonnei virG [53] , galU [21] , and the wbg cluster , encoding the biosynthesis of the O repeating units , from gene wzz to wbgZ on pSS [20] . Erythromycin resistance gene ( erm ) from pAT110 [54] was used to replace the S . sonnei g4c operon coding sequence , from gene ymcD to gene etk [30] . Upstream and downstream flanking regions of the locus to be deleted and the antibiotic resistance gene chosen for replacement were amplified using the primers described in Table 2 and inserted into pBluescript ( Stratagene ) , so that the antibiotic resistance cassette interposed the flanking regions . A linear replacement construct ( upstream region—resistance cassette—downstream region ) was amplified by PCR and transformed into recombination-prone Shigella cells , produced by using the highly proficient homologous recombination system ( red operon ) [55] encoded on pAJD434 [56] . The deletion of the g4c operon in S . sonnei Δg4c was complemented in trans as follows . The g4c gene cluster with the 280 bp upstream region was amplified by PCR ( LongRange PCR Kit , QIAGEN ) and cloned in pACYC184 ( New England BioLabs ) yield pACYC ( g4c ) . g4c complementation in S . sonnei Δg4c ( g4c ) was checked by exopolysaccharide expression . E . coli and Shigella strains were routinely cultured in LB or in TSB medium . When needed , growth media were supplemented with 30 μg/mL kanamycin , 20 μg/mL chloramphenicol , 100 μg/mL erythromycin , 50 μg/mL nalidixic acid , 100 μg/mL trimethoprim , 100 μg/mL ampicillin . GMMA were prepared as previously described [33] or from flask cultures as follows . Bacteria were grown in TSB medium in 1 L flasks to an OD600 of 5 . Culture supernatants were collected by 10 min centrifugation at 4000 g and 0 . 22 μm filtered , concentrated using a 100 KDa regenerated cellulose membrane ( Millipore ) in a Stirred Ultrafiltration Cells ( Amicon ) , separated from soluble proteins by 2 h ultracentrifugation at 186000 g at 4°C ( Optima L‐series , 45Ti rotor , Beckman Instruments ) and resuspended in phosphate buffer saline ( PBS ) . All preparations were 0 . 22 μm filtered . GMMA protein concentration was quantified by Bradford Assay ( Bio-Rad Protein Assay ) using Bovine Serum Albumin ( BSA , Thermo Scientific Pierce ) as standard . Groups of 8 CD1 female mice of 4 to 6 weeks of age were immunized subcutaneously on days 0 , 21 and 35 with 2 μg of GMMA of S . sonnei ΔtolR , S . sonnei ΔtolRΔgalU , and S . sonnei ΔtolR -pSS strains in 100 μL PBS , or with PBS alone . Blood samples were collected before the first immunization ( preimmune sera ) and 14 days after the third injection . Overnight ( ON ) Shigella cultures were diluted in TSB medium to OD600 = 0 . 05 and grown to OD600 = 0 . 5 ( 2 . 5x108 CFU/mL ) . Cells were collected and diluted to 2x107 CFU/mL . For formalin fixation , bacteria were diluted in 0 . 5% formalin solution in PBS and fixed with agitation ON at room temperature . Live or fixed bacteria were washed and resuspended in TSB . Primary staining was performed with mouse sera , the IpaB mouse monoclonal antibody ( H16 ) [57] , the S . sonnei Phase I monovalent rabbit antiserum ( Denka Seiken , cat . # 295316 ) , or the S . flexneri type II monovalent rabbit antiserum ( Denka Seiken , cat . # 295019 ) at the desired dilution for 1 h at 4°C . For competitive staining experiments , 1:1000 diluted anti-S . sonnei ΔgalU GMMA sera were incubated with 11 . 25 μg of LPS ( as quantified by phenol-sulfuric assay ) [58] extracted from S . sonnei WT or S . sonnei -pSS bacteria , for 1 h at 4°C in PBS , prior to bacterial staining . After washing with 1% BSA in PBS , bacteria were incubated with Allophycocyanin-conjugated AffiniPure F ( ab' ) 2 fragment Goat Anti-Mouse IgG ( Jackson ImmunoResearch ) or with Alexa Fluor 488 F ( ab′ ) 2 fragment of Goat Anti-Rabbit IgG ( H+L ) ( Invitrogen ) antibodies for 1 h on ice . Bacteria were washed , fixed with 4% formalin in PBS for 20 min on ice and analyzed for cell-bound fluorescence using a FACSCanto II flow cytometer ( BD Biosciences ) . To evaluate IpaB exposure on the surface of S . sonnei WT , S . sonnei Δg4c , S . sonnei Δg4c ( g4c ) , S . sonnei ΔOAg and S . sonnei -pSS bacteria were stained with the 1:50 dilution of the IpaB monoclonal antibody . For quantitative analyses , the differential Mean Fluorescence Intensity ( ΔMFI ) was measured as the difference between the MFI of the immune staining and the MFI of the control staining using the secondary antibody alone . Flow cytometry data were processed using FlowJo software ( Tree Star ) . LPS and G4C crude extracts were prepared by the phenol-water method [59] , with modifications . For total cellular extraction , ON Shigella cultures were diluted in 50 mL LB supplemented with antibiotics , if needed , to OD600 = 0 . 1 and grown until OD600 = 1 . Bacteria were collected by centrifugation and resuspended in 500 μL PBS . After 5 min boiling , the mixture was treated with 0 . 5 μg/μL of Proteinase K ( Thermo Scientific Pierce ) at 60°C ON . An equal volume of saturated phenol solution ( pH 8 . 0 ) ( Sigma-Aldrich ) was added and incubated for 1 h at 70°C with occasional mixing . After 1 h centrifugation at 10000 g , the upper aqueous phase was mixed with 2 volumes of absolute ethanol and incubated for 1 h at -70°C . Samples were centrifuged at 12000 g for 30 min and the pellet containing LPS and capsule polysaccharides was dried in a rotary vacuum drier ( SpeedVac ) and dissolved in distilled water . To obtain LPS and capsule material with minimal protein and nucleic acids contaminants , GMMA were used as starting outer membrane preparations . After Proteinase K incubation step , samples were treated as described above . The lipid A moiety of LPS was subsequently removed from phenol-water extracts by mild acid hydrolysis treatment with 1% acetic acid for 1 . 5 h at 100°C and adjusted to pH 6 . 5 with ammonium hydroxide . Samples were centrifuged at 15000 g ON and the supernatant , containing exopolysaccharide , was collected and dialyzed against distilled water . Phenol-water extracts and total cell lysates were analyzed by 12% Bis‐Tris Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis ( SDS-PAGE ) and silver stained using the SilverQuest Silver Staining Kit ( Invitrogen ) . Samples were transferred to nitrocellulose membranes and incubated with a 1:1000 dilution of the primary antibody , washed and then incubated with a 1:5000 dilution of the appropriate secondary antibody ( Goat anti-Mouse IgG-Alkaline Phosphatase antibody , Sigma-Aldrich , or Goat anti-Rabbit IgG-Alkaline Phosphatase conjugate , Invitrogen ) . Immunoblots were developed using the SIGMAFAST BCIP/NBT tablet ( Sigma-Aldrich ) solution . At least two independent experiments were performed . Representative blots are shown . High Performance Liquid Chromatography-Size Exclusion Chromatography ( HPLC-SEC ) analysis was used to analyze the size distribution of exopolysaccharide populations and to isolate fractions of different molecular weight . Acid-cleaved exopolysaccharide samples were run on a TSK gel G3000 PWXL column ( 30 cm X 7 . 8 mm; particle size 7 μm; cat . # 808021 ) with a TSK gel PWXL guard column ( 4 . 0 cm X 6 . 0 mm; particle size 12 μm; cat . # 808033 ) ( Tosoh Bioscience ) . The mobile phase was 0 . 1 M NaCl , 0 . 1 M NaH2PO4 , 5% CH3CN , pH 7 . 2 at a flow rate of 0 . 5 mL/min ( isocratic method for 30 min ) . Void and bed volumes were calibrated with λ-DNA ( λ-DNA Molecular Weight Marker III 0 . 12–21 . 2 Kbp , Roche ) and sodium azide ( Merck ) , respectively . Column void volume ( T0 ) : 10 . 58 min; total volume ( Ttot ) : 23 . 03 min . Distribution coefficient , Kd = ( Tretention—T0 ) / ( Ttot—T0 ) . Polysaccharide peaks were detected by differential refractive index ( dRI ) or at 214 nm , when run with a dextran standard curve ( 270–12 kDa range ) to calculate the apparent average molecular weight . Exopolysaccharide was fractionated and collected as follow: HMW-PS , from 11 . 5 min to 13 min; MMW-PS , from 14 min to 16 min; LMW-PS , from 17 min to 18 min . 1D proton NMR analysis was performed to confirm the identity of the polysaccharide samples . NMR spectra of isolated Phase I exopolysaccharide fractions were collected using a standard one-pulse experiment after solubilization of polysaccharide samples in deuterated water . Experiments were recorded at 25°C on Varian VNMRS-500 spectrometer , equipped with a Pentaprobe . Chemical shifts were referenced to hydrogen deuterium oxide ( HDO ) at 4 . 79 ppm . For data acquisition and processing VNMRJ ver . 2 . 2 rev . C and Mestrenova 6 . 1 ( Mestrelab Research ) were used respectively . For alcian blue staining , single colonies were looped directly from the plate into fixative containing 2 . 5% glutaraldehyde , 2% paraformaldehyde , 0 . 075% alcian blue 8 GX and 50 mM L-lysine monohydrochloride in 0 . 025 M sodium acetate buffer at pH 5 . 7 for 2 hours at room temperature . Samples were then rinsed 3 times in buffer and post-fixed in 1% osmium tetroxide in 0 . 1 M sodium cacodylate buffer , rinsed again , dehydrated through an ethanol series , en bloc stained with 2% uranyl acetate at the 30% stage and embedded in TAAB 812 resin . Ultrathin 60 nm sections were cut on a Leica EM UC6 ultramicrotome , contrasted with uranyl acetate and lead citrate , examined on a 120 kV FEI Spirit Biotwin and imaged with a Tietz F4 . 15 CCD camera . For immunogold-labelling , single colonies were looped into 2% paraformaldehyde and 0 . 25% glutaraldehyde in 0 . 1 M phosphate buffer ( PB ) at pH 7 . 4 for 1 hour at room temperature , rinsed 3 times in buffer , infiltrated with 1% and then 10% gelatin before immersing in 2 . 3 M sucrose in PB ON at 4°C for cryoprotection . Frozen samples were prepared by mounting onto aluminum pins and rapidly immersing in liquid nitrogen in preparation for ultrathin 80 nm sectioning on a Leica EM FC6 ultramicrotome . Ultrathin sections were labelled as per Tokuyasu [60] , with the S . sonnei Phase I monovalent rabbit antiserum ( dil . 1:25 ) and detected with 10 nm protein A gold . Imaging was performed as above . Thickness of Phase I material extending beyond the outer membrane was evaluated on different micrographs fields , as average of 20 measurements along the bacterial surface . Invasiveness of the different S . sonnei strains was evaluated in vitro by using a gentamicin protection assay conducted on HeLa semiconfluent monolayers . Each condition was tested in triplicate in three independent experiments . Briefly , 5x105 HeLa cells/well ( HeLa ATCC , CCL-2 ) were seeded ON in 6 well cell culture plates ( Corning Costar ) in Dulbecco’s Modified Eagle Medium ( DMEM ) high Glucose ( Invitrogen ) , supplemented with 10% FBS ( New Zealand , Invitrogen ) . The following day , Shigella ON cultures were diluted in TSB medium ( supplemented with antibiotics , if needed ) and grown to OD600 = 0 . 5 . Bacteria were collected , diluted in DMEM and used to infect HeLa cells with a Multiplicity of Infection ( MOI ) of 10 bacteria/cell . After the addition of the bacteria , the cells were centrifuged at 1100 g for 18 min at 37°C and incubated for 1 h at 37°C . Subsequently , the monolayers were washed with PBS and the medium was replaced with DMEM containing 80 μg/mL gentamicin , to kill extracellular bacteria . After 2 h cells were washed and lysed by the addition of cold 0 . 5% sodium deoxycholate ( Sigma-Aldrich ) in water . Suitable dilutions were plated in triplicates on Congo red agar plates to determine the number of recovered viable intracellular bacteria . Shigella ON cultures were diluted in TSB medium and grown to OD600 = 0 . 5 , equivalent to 2 . 5x108 Shigella bacteria/mL . Bacteria were collected , diluted to 10 , 000/mL in PBS and used as 10x suspension for the assay . Lyophilized baby rabbit complement ( Cedarlane , CL3441 ) was reconstituted with 1 mL of sterile MilliQ water . The bacteria were mixed with reconstituted complement and PBS to yield 50% , 75% , 90% final complement concentration and an inoculum of 1000 bacteria/mL . Heat-inactivated complement was used as control . Bacterial counts were determined by plating on Congo Rod agar at time zero and after 3 h incubation at 37°C . To determine complement sensitivity of S . sonnei WT and S . sonnei Δg4c only red colonies possessing the OAg-encoding pSS virulence plasmid were counted at time zero as white colonies lacking pSS and thus the OAg are highly complement sensitive [38] . The results are expressed as x-fold increase/decrease of the cell count after incubation compared to the count at time zero ( inoculum ) . Virulence of the different S . sonnei strains was evaluated in vivo by testing their ability to induce a Shigella-dependent gut pathology and to spread to systemic sites in the rabbit model of ligated ileal loops in New Zealand White rabbits weighing 2 . 5–3 kg ( Charles River Breeding Laboratories , Wilmington , MA ) . In the main experiment , each loop was infected with 3x109 bacteria of a single strain in 500 μL of physiological saline buffer ( 0 . 9% NaCl ) . 12 loops per animal of 5 cm segments of ileum starting at the ileum-cecum transition were ligated , avoiding all Peyer’s patches , while maintaining the existing vasculature . Each strain was tested in 2 rabbits , in 3 replicate loops per rabbit , with a randomized order for each rabbit . Surgery was performed as described [61] . In a preliminary experiment , 5x109 bacteria/loop infectious dose was tested in 2 loops for each strain in 2 rabbits . After euthanasia , loops were dissected and processed for RNA extraction and histology . Uninfected tissues were collected as control . Mesenteric lymph nodes , spleen , liver and blood were collected and processed for bacterial counting . Appropriate dilutions of the different tissue samples were plated on selective and non-selective TSB-agar: S . sonnei Δg4c ( g4c ) was identified by growth on chloramphenicol; S . sonnei Δg4c was enumerated by growth on erythromycin minus the counts of S . sonnei Δg4c ( g4c ) determined on chloramphenicol as S . sonnei Δg4c ( g4c ) is resistant to both; S . sonnei -pSS was selected on nalidixic acid; S . sonnei WT does not carry any resistance markers and thus was counted by plating on non-selective medium and subtracting the number of the other strains determined on the selective media from the total number of CFU obtained on non-selective medium . For histopathological analysis of Shigella infected ileal loops and bacterial immune-localization , intestinal biopsies were fixed at 4°C in 4% paraformaldehyde in PBS , embedded in paraffin and sectioned into 7 μm slices using a microtome . Sections were deparaffinated , rehydrated and used for H&E staining , or for anti-Shigella immune staining . Immune staining was performed as follows . Sections were permeabilized for 15 min with antigen unmasking solution ( 10 mM Tris , 1 mM EDTA , 0 . 05% Tween20 , pH 9 ) , treated with 3 . 3% H2O2 for 10 min and washed . Samples were blocked for 15 min with Ultra V block ( Lab Vision Corp; Thermo Scientific , cat . # TA-125UB ) and incubated ON with the in-house mouse polyclonal anti-S . sonnei serum raised against S . sonnei ΔtolR GMMA . Samples were then incubated with a peroxidase labelled polymer conjugated to goat anti-mouse immunoglobulins ( DAKO , cat . # K4000 ) for 1 h , revealed with the 3-amino-9-ethylcarbzole AEC+ Substrate-Chromogen ( DAKO , cat . # K3461 ) , counterstained with hematoxylin and mounted with aqueous mounting medium ( Merck ) . Histology images were taken using light microscopy , at 4X magnification for H&E staining , and at 20X magnification for immune staining . Shigella-dependent intestinal inflammation and the degree of tissue alteration were assessed by evaluating the extent of villi atrophy and submucosal edema and by determining scores of inflammation according to the generally used criteria of the Ameho gradation scale [37] , in slices from rabbit loops after 8 hours infection with 3x109 bacteria/loop of a single strain . Pro-inflammatory cytokines ( IL-8 , IL-6 , and IL-1β ) gene expression was measured by RT-qPCR in rabbit loops after 8 hours infection with 5x109 bacteria/loop of a single strain . Primers and methods were previously described [61] . The expression levels of the cytokine genes were determined as fold induction over the housekeeping gene GAPDH in each loop . To compare expression levels elicited by S . sonnei Δg4c and S . sonnei WT the ratio of the fold induction ( over GAPDH ) in S . sonnei Δg4c infected loops and the fold induction in the adjacent S . sonnei WT infected loop was calculated . S . sonnei genes ( S . sonnei 046 ) : tolR: 3669577; virG: 3670887; galU: 3667724; wzz: 3670967; wzx: 3670970; wzy: 3670971; wbgZ: 3670977; ymcD: 3666464; ymcC: 3666463; ymcB: 3666462; ymcA: 3666461; yccZ: 3669961; etp ( yccY ) : 3669960; etk ( yccC ) : 3669959 . S . flexneri genes ( S . flexneri 2a 2457T ) : tolR: 1077009; galU: 1077674; rfbG: 1078499; rfbF: 1078523; rfc: 1078521 . The mouse immunization experiments performed at the Novartis Animal Facility in Siena , Italy , complied with the relevant guidelines of Italy ( Italian Legislative Decree n . 116/1992 ) and the institutional policies of Novartis . The animal protocol was approved by the Animal Welfare Body of Novartis Vaccines , Siena , Italy , and by the Italian Ministry of Health ( Approval number AEC 2009–05 ) . The experiments using the rabbit ileal loop model performed at Institut Pasteur , Paris , France , complied with the EU Directive 2010/63 and the French Decree 2013–118 . The respective protocol was approved by the Comité Regional d’Éthique pour l’Expérimentation Animale in Paris 1 ( protocol no . 20070004 ) and reviewed by the Global Animal Welfare Board of Novartis .
|
Shigellosis is a major global health concern . Recently , a shift in the dominance of types of Shigella that cause disease has been observed with S . sonnei increasing in prevalence under improved socio-economic conditions leading to a replacement of S . flexneri . Most of the knowledge of Shigella disease mechanisms has been obtained from studies of S . flexneri . We found that S . sonnei possesses a high molecular weight sugar capsule that is absent in S . flexneri 2a . Removal of the capsule made S . sonnei bacteria highly invasive in vitro and strongly inflammatory in vivo , but in contrast , there was reduced spreading of these mutant bacteria from the gut to peripheral organs in rabbits and higher sensitivity to complement-mediated lysis . Thus , the capsule plays a role in both , invasion and protection of the bacteria against the innate immune defense of the host during the infection . These findings indicate that the capsule is an important virulence factor for S . sonnei . Based on the substantial changes in pathogenesis observed upon removal and overexpression of the capsule , we hypothesize that its level of expression may be under significant evolutionary pressure .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
An O Antigen Capsule Modulates Bacterial Pathogenesis in Shigella sonnei
|
Many toxin-antitoxin operons are regulated by the toxin/antitoxin ratio by mechanisms collectively coined “conditional cooperativity” . Toxin and antitoxin form heteromers with different stoichiometric ratios , and the complex with the intermediate ratio works best as a transcription repressor . This allows transcription at low toxin level , strong repression at intermediate toxin level , and then again transcription at high toxin level . Such regulation has two interesting features; firstly , it provides a non-monotonous response to the concentration of one of the proteins , and secondly , it opens for ultra-sensitivity mediated by the sequestration of the functioning heteromers . We explore possible functions of conditional regulation in simple feedback motifs , and show that it can provide bistability for a wide range of parameters . We then demonstrate that the conditional cooperativity in toxin-antitoxin systems combined with the growth-inhibition activity of free toxin can mediate bistability between a growing state and a dormant state .
Many bacteria and archaea have multiple Toxin-Antitoxin ( TA ) loci [1] , where the toxin normally inhibits cell growth , while the antitoxin neutralizes the activity of the toxin by forming a tight TA complex . One of the known functions of TA loci is to respond to nutritional stress [2] , namely , toxins are activated upon nutritional starvation and slow down the rate of translation . When cells are under normal fast growth conditions , on the other hand , the majority of the cells will be in the antitoxin-dominated state , such that toxin activity is fully inhibited . It has been found that many bacterial TA loci are auto-regulated at the transcriptional level by a mechanism called “Conditional Cooperativity” ( CC ) [3] , where the transcription factor can bind cooperatively to the operator only if the concentrations of two different proteins satisfy a certain stoichiometric ratio . CC was quantitatively studied in one of the Escherichia coli TA loci , relBE [3]–[6] . Here the two proteins , the toxin ( mRNase ) RelE and the antitoxin RelB , are encoded by the same operon , which is negatively auto-regulated . The tight dimer is a weak transcriptional auto-repressor , but this repression is strongly enhanced by the presence of RelE and becomes strongest at ratio . Over-expression of RelE above twice of , though , will result in an abrupt de-repression of the promoter . This unique behavior is a consequence of formation of alternative hetero-complexes of RelB and RelE; and . Two bind to the promoter site cooperatively to repress the promoter strongly , while does not bind to the promoter . Interestingly , all plasmid and chromosome-encoded TA loci investigated are found to be regulated by CC so far , including relBE of E . coli [3] , [4] , vapBC of Salmonella enterica [7] , phd/doc of plasmid P1 [8] , [9] and ccdA/ccdB of plasmid F [10] . This suggest that CC is a common feature for TA loci . In our previous work , we have explored the function of CC in the starvation response of the RelBE system , and showed that CC prevents random toxin activation and promotes fast translational recovery when starvation conditions terminate . However , to reproduce the full dynamics of the starvation response , we took into account details of the RelBE system , which made the model rather specific to it . The primary purpose of this paper is to construct a simple mathematical model that demonstrates the functions of CC in a more general perspective . TA loci have been suggested to be involved in persister formation [11]–[16] . When an antibiotic is applied to a growing bacterial population , the majority of the bacteria are killed . However , a very small fraction of them survives and re-grows after the antibiotic is removed . If the progeny of the bacteria is again sensitive to the same antibiotic , they are called persisters , in contrast to the resistant bacteria that have acquired resistance to antibiotic by mutation . Persisters are genetically identical to the sensitive cells , but believed to be in a non- or slow-growing , dormant state . Since the majority of antibiotics interferes with the cell growth and division process , cells can survive if they grow slowly or not at all . The exact molecular mechanism underlying persistence is not fully understood . However , it has been found that mutations in hipAB genes severely increase the level of of persisters formation . Interestingly hipAB is one of the TA loci in E . coli [11] , [13] , [14] . In addition , recent experiments [15] showed that removal of 10 mRNase-encoding TA loci reduced the persister fraction significantly . These observations strongly suggest that TA loci are important factors for persister formation . One of the possible explanations is that stochastic activation of the toxin will slow down cell growth , resulting in a dormant state . This will be possible if the TA locus dynamics exhibits bistability , where a cell can be either in the antitoxin-dominated state that ensures the growth or in the toxin-dominated state that inhibits the growth . This viewpoint is also consistent with the observation that the persister state can be described as a metastable state with a constant stochastic switching rate to and from normal growing state [12] . This idea was theoretically pursued by Lou et al . [17] with a simple mathematical model that did not take CC into account . They concluded that , for bistability to be achieved , high cooperativity ( Hill-coefficients ) is necessary , both in transcriptional auto-regulation of the TA operon and in the free toxin activity . In this paper , we explore the basic features of CC as a regulation mechanism mediated by heteromer formation . We demonstrate that CC provides bistability in a simple feedback motif in a wide range of the parameters . We then construct a simplified model of TA system regulation and demonstrate that CC with growth rate-mediated feedback via toxin activity can provide the bistable alternatives between the antitoxin-dominated and the toxin-dominated states .
In this section , we construct a simple model of TA activity control with CR , a model aimed at capturing the central features of persister formation . We use the RelBE system as a reference because the molecular interactions and parameters are best known here . The reference parameters are listed in Materials and Methods . In RelBE [6] , the antitoxin RelB and the toxin RelE are encoded by the same operon , and transcriptionally auto-regulated by CC . RelE is metabolically stable , and its concentration decreases only by dilution due to cell division ( generation time ∼30 min in log phase growth in rich medium ) . On the other hand , RelB is actively degraded by protease Lon , resulting in its very short half-life of min . In spite of this , the RelB concentration in a normally growing cell is about 10 times of that of RelE [4] , suggesting that the RelB mRNA is translated about 100 times more often than RelE mRNA [6] . This situation is depicted in Fig . 3A1 . Since both toxin T and antitoxin A are regulated by the same promoter , the corresponding equations apply: ( 5 ) where and are the maximal production rate for and for , respectively . The dilution rate of is given by cell division , and is taken as a unit rate , while is the active degradation rate of . This motif , however , cannot exhibit bistability . Fig . 3A2 shows example null-clines , which have only one stable fixed point at the antitoxin dominated state . We performed parameter scan spanning from 1/8 to 8 fold relative to the values used for Fig . 3A2 , but did not find any combination of parameters that gives bistability , even if we allow cooperative binding of AT to DNA with Hill coefficient 2 ( data not shown ) . This absence of bistability is due to A being regulated identically to T . Accordingly , the de-repression of the promoter around increases not only the toxin production but also the antitoxin production , and the latter is so large that the system remains in the antitoxin-dominated state . When we include the activity of free toxin on cell growth , however , the model system can show bistability . This is because the toxin-induced arrest of cell growth prolong lifetime of T , while leaving A being degraded by Lon at a high rate . The mathematical formulation of this extended model is ( 6 ) ( 7 ) expressing that reduces all protein production , and accordingly also decreases the dilution by cell growth . represents the reduction of protein expression per free toxin ( ) molecule , and represents the growth inhibition per free toxin molecule . Notice that does not influence degradation of A , because it is anyway so unstable that cell division hardly affects its concentration . These terms correspond to the growth-rate dependent feedback [17] , [26] , [27] . The reduction of the protein production ( term ) can account for both direct activity of free toxin to TA locus and the global slowdown of the transcription rate due to change of physiological conditions [26] . Comparison of the present model with the steady state growth data in Ref . [26] is given in Text S1 . We expect because the slowing down of the growth rate is due to the global slowing down of the protein production . At the same time , there can be some quantitative difference because may include the effect specific to the TA locus . The growth-rate reduction mediated by T constitutes a positive feedback [17] , [26] , [27] on T accumulation , which is essential for bistability and persister formation . The term with reduces the production of both antitoxin and toxin , and thus overall weaken the ability to maintain the bistability . Note that primarily influences the transition state from A to T dominated state , because the reduction of production targets the short lived A protein first . Fig . 3B1 examines eqs . ( 6 ) – ( 7 ) with parameters extracted from the RelBE system [6] ( see the figure caption of Fig . 3 ) . The null-clines in Fig . 3B2 are from the case , exhibiting two stable fixed point , one at the antitoxin-dominated state ( the low- state , , ) and another at the toxin dominated state ( the high- state , , ) . Note that the antitoxin dominated state has almost the same concentrations as the stable fixed point in Fig . 3A2 with . The antitoxin dominated state scarcely depends on and , since there is almost no free toxin ( ) in the antitoxin dominated state . Figure 4A shows the ratio between the dilution rates at the low and high steady state , . The figure illustrates that our model predicts bistability for a wide range of parameters , and further that this bistability is indeed governed by the increase in cell generation parameterized by the term . For too large the bistability is counteracted because the toxin production is reduced too much by free toxin to accumulate enough for the stable high toxin state . Remarkably , for proportional reduction of protein production and increased cell generation , , the model predicts bistability for all . We also studied the robustness of the bistability against parameter change . One of the most crucial parameters for the bistability is the ratio , because this determines the difference of the concentration of and . We therefore varied with keeping constant , and searched for the bistable regime in space . The rest of the parameters are kept same as those used in Fig . 4A . Only is considered , because lower ratios prevent antitoxin domination due to its 10 times higher degradation rate . For rather small ( ) , too large makes the anti-toxin dominated state unstable , because very small amount of free toxin is enough to activate the positive feedback to toxin via the growth rate . With even larger , stronger feedback is needed to stabilize toxin-dominated state , reflected in larger values of and . We further performed scanning of other parameters . We fixed one parameter at a time and sampled the rest of the parameters randomly to test 1000 samples in logarithmic scale within the range between 1/8 to 8 fold of the reference values . We then systematically changed the fixed parameters between 1/8 to 8 fold and repeated the procedure , to see the effect of the parameter . We found that 20% to 80% of the samples showed bistability . The detailed results are given in Text S2 . We also explored the effect of the dissociation constant and more intensively , by changing from the reference value to 64 fold , since they describe the sharpness of the CR and this is expected to influence the bistability . We find that the number of bistability parameter sets decreases gradually with the fold change of and . Details are given in Fig . S4 .
Using known parameters for the RelBE system in E . coli , we constructed a minimal model for TA activity , combining conditional regulation with a feedback from free toxin to the cell growth . It was demonstrated that this model shows bistability for a wide range of parameters , with a stable state corresponding to the antitoxin-dominated , normal growing state , and another metastable state corresponding the toxin dominated state , potentially corresponding to the persister state . Noticeably , the model eqs . ( 6 ) – ( 7 ) did not rely on details of the molecular mechanisms of how the toxin works , and therefore the model is not limited to the RelBE system . The important assumptions are: ( i ) The TA system is conditionally regulated , ( ii ) toxins are stable and diluted mainly by cell division , while antitoxins are metabolically unstable , and ( iii ) free toxins reduce the productions of proteins and hence cell growth . All the conditions are satisfied in the TA loci that are confirmed to be regulated by conditional cooperativity [3] , . Our simple model pinpoints minimal ingredients for obtaining a persister state , but did not include stochastic production and/or degradation , and therefore cannot address the switching rates . In order to understand stochastic persister formation in E . coli , just performing stochastic simulation of the present motif is not enough , because the frequency of persisters depends on multiple parallel TA systems . In E . coli , 11 simultaneously interfering TA systems maintain a probability of persisters to be about 0 . 01% , while this probability is changed substantially first when about 50% of the TA systems is removed [15] . This clearly suggests that the interference of parallel systems has a strong influence to the switching behavior . Furthermore , comparing the stochastic simulations with the experimentally observed frequency of persisters requires a knowledge of the underlying distribution of the expression levels and corresponding growth rates in the cell population . It is not a simple task when the single cell growth rate depends on expression levels , because it feedbacks to the frequency of the cells as pointed out by Nevizhay et al . in [28] . In addition , it has been suggested that there is a strong link between the activation of the protease Lon and the TA-mediated persister formation , through the increase of the antitoxin degradation rate [15] , [16] . The fluctuation of the Lon activity may be particularly important in determining switching rates , because it can provide coherent noise that favours simultaneous switching of many TAs to the persister state . It should also be noted that the Lon activity is activated by polyphosphate , which is regulated by the stringent response signalling molecule ( p ) ppGpp [16] . We plan to extend the present model to include these features and study the switching behavior in near future . It is still interesting to think about possible implication of the observed switching rate to the present model . The fact that the persister formation is a rare event may indicate that the actual parameter value in the real system is located close to the boundary between the bistable region and the monostable region of the antitoxin-dominated state . Such parameter values can be chosen through selection process in a fluctuating environment , where slow growth of the persister pays off as a risk hedging strategy; the switching rate is expected to reflect the time scale of the temporal fluctuation of the environment [29] . Conditional regulation is an example of mixed feedback motifs [30] , where protein-protein interactions and transcriptional repression are combined . In natural systems , protein-protein interaction mediated bistable switch was previously found for example in the epigenetic switch of the TP901 phage [23] , [25] and in the sigma-factor/antisigma-factor system [24] . Conditional cooperativity in TA systems opens for a toolbox of regulatory units that can exhibit sufficient bistability to support also epigenetics . When removing the toxic ability of toxin , which has been done for RelE [3] , and separating antitoxin from the operon to allow independent control , the strong binding between RelE and RelB should provide extreme ultrasensitivity , and thus very well separated metastable states . This conditional cooprativity-mediated bistability is the base for the bistability in full TA systems , and thus for the type II persister formation [12] , [13] , where a cell can spontaneously switch between the dormant state and the growing state ( Fig . 5 ) . While simple protein-protein heteromers could produce ultrasensitivity , the non-monotonicity of the conditional cooperativity also secure that the antitoxin dominated state has a substantial amount of toxins present ( Fig . 5 ) . These toxins' activity is normally inhibited by short lived antitoxins , but the stored toxins can be used for faster switching to a dormant state if overall protein productions are externally inhibited , for example by starvation ( Fig . 5 ) . Therefore , the non-monotonicity may enhance the transition to type I persister formation [12] , [13] , where environmental stress triggers persister formation . The importance of the protein-protein interaction mediated ultrasensitivty [22]–[25] and the growth rate-mediated feedback [17] , [26]–[28] to bistable systems have been discussed as independent regulatory features in recent literature [31] . The uniqueness of the bistability in the TA system is that it combines both of these mechanisms . The need for combining these two mechanisms is closely associated with the fact that T and A are produced from the same operon , and thus are exposed to identical transcription regulation . Though it is difficult to get bistability with only one of the mechanisms [17] , the TA system realizes a persister state by regulating the products of one operon through a combination of growth modulation and hetero-complex formation .
All the numerical analyses are done using C++ codes developed by the authors . When necessary , was calculated by solving algebraic equations ( 2 ) and ( 3 ) with conservation of mass for a given amount of by Newton's method [32] . The bistable solutions in Fig . 2 B ( Fig . 4 ) were obtained by finding the fixed points for with eq . ( 4 ) ( and with eqs . 6 and 7 ) by Newton's method and then evaluating the stability based on the Jacobian . The trajectories that constitute the flux in Figs . 3A2 and 3B2 were calculated by the 4th-order Runge-Kutta method [32] . The values of the parameters used in the ODEs correspond to a conversion to dimensionless numbers of the parameters relative to the system we studied in [6] . In particular we used the lifetime of in exponential growth conditions ( ) as time-unit ( ) and the maximal amount of proteins produced in the unit time as concentration unit ( ) . In the system nM thus fixing we obtain nM , while min . The value of in the starved condition [6] was evaluated to be around in this units . However , it is expected to be smaller in the normal condition , since RelE cleaves mRNA at the ribosomal A-cite , which is expected to be more accessible at the starvation . Therefore , we mostly explore values smaller than 1000 . The reference parameters are shown in table 1 .
|
The effectiveness of antibiotics on many pathogenic bacteria is compromised by multidrug tolerance . This is caused by a small sub-population of bacteria that happen to be in a dormant , non-dividing state when antibiotics are applied and thus are protected from being killed . These bacteria are called persisters . Unraveling the basic mechanism underlying this phenomenon is a necessary first step to overcome persistent and recurring infections . Experiments have shown a connection between persister formation and the battle between a toxin and its antitoxin inside an E . coli cell . Toxin inhibits the cell growth but is neutralized by the antitoxin by forming a complex . The proteins also regulate their own production through this complex , thereby forming a feedback system that controls the growth of the bacterium . In this work we provide mathematical modeling of the feedback module and explore its abilities . We find that the auto-regulation with reduced growth associated with free toxins allows the cell to be bistable between two states: an antitoxin-dominated , normal growing one , or a dormant one caused by the activity of the toxin . The latter can be the simplest description of persister state . The toxin-antitoxin system presents a powerful example of mixed feedback design , which can support epigenetics .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Conditional Cooperativity of Toxin - Antitoxin Regulation Can Mediate Bistability between Growth and Dormancy
|
The ability of microbial species to consume compounds found in the environment to generate commercially-valuable products has long been exploited by humanity . The untapped , staggering diversity of microbial organisms offers a wealth of potential resources for tackling medical , environmental , and energy challenges . Understanding microbial metabolism will be crucial to many of these potential applications . Thermodynamically-feasible metabolic reconstructions can be used , under some conditions , to predict the growth rate of certain microbes using constraint-based methods . While these reconstructions are powerful , they are still cumbersome to build and , because of the complexity of metabolic networks , it is hard for researchers to gain from these reconstructions an understanding of why a certain nutrient yields a given growth rate for a given microbe . Here , we present a simple model of biomass production that accurately reproduces the predictions of thermodynamically-feasible metabolic reconstructions . Our model makes use of only: i ) a nutrient's structure and function , ii ) the presence of a small number of enzymes in the organism , and iii ) the carbon flow in pathways that catabolize nutrients . When applied to test organisms , our model allows us to predict whether a nutrient can be a carbon source with an accuracy of about 90% with respect to in silico experiments . In addition , our model provides excellent predictions of whether a medium will produce more or less growth than another ( ) and good predictions of the actual value of the in silico biomass production .
Our phenomenological model expresses the impact that different carbon sources ( or nutrients ) have on a microbe's ability to grow using only information on the chemical structure of these carbon sources and on the ability of a microbe to catabolize these nutrients . Our model is built to reproduce the predictions of flux balance analysis calculations ( FBA ) on metabolic reconstructions , thus it will suffer from the same limitations . Indeed , while FBA is a powerful tool to investigate microbial metabolism and microbial growth in particular , it has a number of limitations when estimating growth rates and the effect of media and environmental conditions on growth . It is well-known that microbes need a minimal medium and a carbon source in order to grow . Minimal media have been described for a number of species and contain essential chemical species without which the species would not be able to grow [20] , [29] . The growth rate of a microbe , however , depends on many other factors including the uptake rate of nutrients , temperature , regulation , the availability of oxygen , etc . FBA is a linear optimization method that predicts the maximal conversion of a set of carbon sources into biomass with a fixed minimal medium . In order for FBA conversion rates to reproduce empirical growth rates , one needs to consider an additional ATP maintenance flux which is obtained by fitting FBA results to empirical growth rates obtained for a certain temperature , minimal medium , and carbon source uptake . While ATP maintenance rates obtained for a specific minimal medium have been shown to give accurate predictions of growth rates in different minimal media for some organisms [8] , in principle one cannot assume that they are valid for predicting growth rates for arbitrary minimal media . Additionally , because metabolic reconstructions do not consider regulatory constraints , FBA will predict that a microbe is capable of uptaking two different sugars simultaneously , while it is well-known that if there is more than one sugar carbon source at high enough concentrations , the organism will exhaust the preferred one before consuming the others [30] . A microbe will , however , consume multiple sources of carbon other than sugars simultaneously and as a consequence grow faster [31] . It has been shown that steric constraints can already reproduce diauxic growth in some organisms in the presence of multiple sugars [32] , however , there is no general framework that is able to deal with this issue when using FBA . To develop a phenomenological model that is able to reproduce maximal biomass conversion rates per carbon source under aerobic conditions , we investigate the maximum amount of biomass that can be produced by an organism in the presence of a minimal medium , oxygen and one or more carbon sources including at most one sugar . To this end , we run FBA on metabolic reconstructions in which we remove ATP maintenance [27] ( see Methods for specific details ) . We concentrate on a training set of four species for which there are high-quality metabolic reconstructions available , and which cover a wide range of microbial phylogeny: a gram-negative bacterium ( E . coli [8] ) , a gram-positive bacterium ( B . subtilis [9] ) , an eukaryote ( Saccharomyces cerevisiae [21] ) , and an archaeon ( Methanosarcina barkeri [24] ) . We validate our model on a test set of three species for which we also have high-quality metabolic reconstructions—Helicobacter pylori , a gram-negative bacterium [20] , Staphylococcus aureus , a gram-positive bacterium [22] , and Mycobacterium tuberculosis , an acid-fast gram-positive bacterium [23] . The rationale for choosing a small number of species for model building and validation is that lower quality reconstructions are likely to have significant gaps that could incorrectly bias the determination of the model . Additionally , the aggregate set of nutrients available for these reconstructions is of 352 nutrients , which cover all nutrient types and 90 out of 97 pathways available in KEGG [33]–[35] , thus ensuring that our model is comprehensive . We believe that the development of a systems-level model of microbial metabolism is not only complementary to current approaches but offers some advantages with respect to them . Specifically , while FBA run on metabolic reconstructions already has the capability of predicting maximum biomass yield , our model has the advantage that since it does not consider microscopic details of the metabolism of a species it is directly applicable to any other organism growing under the conditions of our analysis . In fact , we show that solving over a thousand linear equations under constraints can be well approximated by a simpler model whose principles are easier to understand . As such , our model offers the possibility of uncovering universal features of the metabolism of organisms that other computational approaches are not capable of . The mathematical model we develop is thus a valuable tool from both fundamental and applied perspectives , since it can help understand the metabolism of organisms for which a metabolic reconstruction is not available , or guide the process of validating metabolic reconstructions .
The first step is to build a model that predicts whether an individual nutrient can be used as a source of carbon by species ( see Fig . 1 ) . If it can , we say that belongs to the group of nutrients that contribute to growth in ; otherwise we say that belongs to the group of nutrients that do not contribute to growth in species . We use flux balance analysis on the species in the training set to empirically determine which nutrients belong to and which ones to ( see Materials and Methods ) . We then use these data to build the model . In this section we describe the model ( which we summarize in Fig . 2 ) and validate it with the species in the test set . The next step is to determine for those nutrients that produce growth what is the maximal production of biomass when that nutrient is the only source of carbon . In fact , if a nutrient is the carbon-limiting source in a given medium , the biomass yield of the nutrient must be related to the number of carbons in the nutrient . In Fig . 6 , we display the in silico biomass production and the biomass yield of all nutrients as a function of the number of carbon atoms they contain , for the species in the training and test sets ( we do not consider M . barkeri in our model for biomass yield because M . barkeri can only grow anaerobically , and this has a significant effect on the energy used to polymerize the proteins and nucleic acids in the biomass ) . It is visually apparent from Fig . 6 that there is a strong correlation between the number of carbons in a nutrient and the biomass production . We thus model the biomass production induced by nutrient as ( 3 ) where is the ( nutrient-independent ) average biomass yield of the nutrients in The data suggests the existence of an upper bound for the biomass yield ( ) as a function of the number of carbons ( blue line in Figs . 6 , 7 , and 8 ) . The nutrients frequently found at or close to this upper bound are sugars , alcohols , and other compounds with hydroxyl groups . Many of the hydroxyl groups on these compounds are typically oxidized in order to reduce NAD to NADH , which is an important source of energy , and will in turn increase the biomass yield . For simplicity , we obtain as the average biomass yield for the sugars uptaken by an organism . The average nutrient has of the efficiency of these high-efficiency sugars , but nutrients vary wildly in their yield ( Fig . 7 ) . We consider one main factor that contributes to this variation in biomass yield: the number of carbons in a nutrient that are effectively catabolized . Finally , we want to model the maximal biomass production when there is more than a single source of carbon present in the medium , or in other words when organisms grow in a complex medium . We consider complex media in which nutrients are restricted to five classes: sugars , fatty acids , amino acids , purines , and pyrimidines , partly because they are commonly used in growth rate/biomass yield experiments , and partly because it simplifies the analysis of the results . In addition , complex purines and pyrimidines are a mixture of sugars and nucleobases , and as we are already considering sugars , we only use the simple nucleobases . The simplest plausible model for the contribution of nutrients to the biomass production is one in which each nutrient has an independent contribution to the biomass production . For each species , we calculate the biomass production on a complex medium containing nutrients using ( 5 ) We estimate for a nutrient from class using the average yield for that class in the organisms in the training set ( 6 ) where is the number of nutrients of class that are taken up by species , and represents the average over species in the training set . For tryptophan and for purines , we use the effective number of carbons when calculating yield , as described previously . To test this model , we randomly generate ensembles of complex media as described in the Data and Methods section . For each medium and for each species , we calculate ; we use FBA to find the actual in silico biomass production of the organism on the complex media . In Fig . 9 we show how compares to for each species , and for each of the complex media we generate . In these comparisons we also include predictions for the species in the test set ( with the exception of S . aureus , for whose reconstruction it is not clear what units were used for the biomass production ) . We find a very strong correlation ( using the Spearman rank correlation coefficient ) between the predictions of our model and in silico experimental results for training set species ( E . coli , B . subtilis , and S . cerevisiae ) and test set species ( H . pylori and M . tuberculosis ) . This indicates that the model accurately captures which media will result in faster/slower production of biomass . For each species , however , the model systematically under or over-predicts growth . A regression of the predicted biomass production versus experimental in silico production indicates that , with: for E . coli , for B . subtilis , for S . cerevisiae , for H . pylori , and for M . tuberculosis . The parameters used to make the predictions for the individual nutrients in the linear model were trained on three species . The linear model consistently under-predicted the in silico biomass production for these three species , and more so for media containing more nutrients . This is a strong indicator that the nutrients that are uptaken are used synergistically by the in silico organism to produce more biomass than expected , showcasing the effect of catabolic pathways being highly connected in the metabolic network . A more complete model for predicting biomass production in complex media will therefore need to take into account synergistic interactions among catabolic pathways . Our model sheds light on several questions related to the impact of nutrients on the biomass production of microbes . Our approach treats microbial metabolism as a “black box” that uses nutrients to reach optimal biomass production . Because the model does not take species-specific details into consideration , it is useful for generating predictions for any microbe , something that would be impossible with any existing modeling approach . To illustrate how one would proceed to extrapolate to “new” organisms and to show what kind of insights one could obtain , we generate predictions for four organisms that lack a metabolic reconstruction: R . palustris ( a gram-negative bacterium ) , L . monocytogenes ( a gram-positive bacterium ) , D . discoideum ( an eukaryote ) , and T . acidophilum ( an archaeon ) . We find that , overall , these species are predicted to take up fewer nutrients than the species in the test set ( Fig . 10A ) . This is a consequence of the limited annotation that the authors of TransportDB could use for the predicted protein transporters . One example of this limitation is that for all four species , there were many transporters predicted to take up amino acids , but there was no indication of which amino acids the transporters were specific for . Therefore , for the sake of prediction , we consider that all four species uptake all twenty natural amino acids . We then use the model of catabolic potential to predict whether each of these nutrients could be a source of carbon . In Fig . 10B we show the number of nutrients that belong to one of the nutrient classes previously described and whether these nutrients are or . For the fatty acids , none of which were predicted to be uptaken by any of the four species , we examined the enzymes available and found that only D . discoideum contains the enzymes for -oxidation , and could therefore catabolize fatty acids . Finally , we combine our models of biomass yield of nutrients , and of biomass production on complex media ( Fig . 11 ) . We choose four complex media which contain a different number of the same nutrients we used for the randomly-generated complex media , namely: sugars , fatty acids , bases , and amino acids . The four media contain: 1 ) glucose; 2 ) glucose and hexanoic acid; 3 ) glucose , hexanoic acid , adenine , guanosine , cytosine , and thymine; 4 ) glucose , hexanoic acid , adenine , guanosine , cytosine , and thymine and the twenty natural amino acids . We estimate the biomass production in the same manner that we described earlier for the randomly-generated complex media . We find that the biggest influence on the biomass production is the number of carbons available in the nutrients present in the medium; because we are now adding 20 amino acids to medium 4 , the biomass production increases almost -fold , on average , in that case . Note that these predictions for biomass production are based on the biomass yield per carbon available in the nutrients present in the medium . Biomass yield is a time independent quantity that cannot be directly associated to growth rate . However , because biomass yield gives an upper limit for growth rate [48] , our model can be used as a baseline for researchers to explore and model growth rates .
Microbes use nutrients found in their environments to grow . Understanding and developing quantitative models of this process is of fundamental importance in cell biology , physiology , medicine , evolution , synthetic biology and bioengineering , not to mention of practical importance to those that need to grow microbes in laboratories . Given the diversity of the microbial world and the number of combinations of nutrients available in their environments , this seems too difficult a problem . In this study , we focused on how microbes might catabolize nutrients to obtain carbon for biomass production . Our model comprises three levels , with each level building up on the results from the previous one . The first level concerns whether a nutrient will be catabolized . The second level concerns whether all of the carbon in a catabolized nutrient is available for biomass production . The final level incorporates the biomass yield for selected classes of nutrients and enables us to make a prediction on the biomass production of a microbe in a complex medium . To validate our approach , we compare the predictions of the complete model with in silico predictions of growth in complex media of species that are not part of the training set . Our results on these species are excellent predictions of which media will produce more/less growth . Finally , we looked at the biomass yield of sugars , amino acids , purines , pyrimidines , and fatty acids . We found little variation in the yield of these nutrients amongst different species , and were able to postulate a model for the biomass production of a microbe on a complex medium containing any number of these important nutrients . All of these nutrients have separate catabolic pathways , with the exception of some groups of amino acids which share a catabolic pathway . The fact that the in silico microbial biomass production was more than our model predicts indicates that each of these catabolic pathways can work together synergistically to improve the biomass production of the microbe . The ability of our model to predict biomass production on complex media must be balanced by an understanding of its limitations . First , we model microbial metabolism as a black box , and therefore we do not account for absent pathways that biosynthesize some biomass components . The absence of such pathways would require that the corresponding biomass components be made available in the medium used , but our model cannot be used to predict which of these are indeed required . This means we cannot use the model in its current form to predict the minimal medium that is needed for a microbe to grow . Second , microbial species are sometimes identified by the nutrients they take up and excrete . The ability of a microbe to transport such compounds in and out of the cell is largely dependent on the specific protein transporters available . A separate body of work exists , largely in the form of TransportDB [39] , and TCDB [49] , [50] , which will enable the researchers to predict the proteins that transport specific nutrients . We incorporate such predictions of nutrient transport into our model to make predictions for specific species that lack a metabolic reconstruction . Importantly , while knowledge of transporters can enable us to predict which nutrients in the medium can be taken up , we are currently unable to predict which nutrients are excreted as that would in principle require knowledge of the full metabolic network . Third , the stoichiometric method we describe for generating the biomass data cannot be used to predict the growth rate of a microbe because kinetic information is not included . However , our model provides a baseline to which one can add kinetic information . For example , the rate-limiting step in poor growth media is likely to be the rate at which a microbe takes up nutrients . Therefore one can use the kinetics of nutrient transport with our prediction of biomass yield to predict growth rate . All this notwithstanding , we believe that our approach and models open the door to significant advances in the quantitative modeling of microbial metabolism , and eventually of the metabolism of more complex organisms . In particular , our models could be extended to consider whether a nutrient acts , not only as a source of carbon , but also as a source of nitrogen or energy , or directly as a component of the biomass . Our models could also be extended to include more detailed information about pathways , or to consider functional metabolic modules [51]–[53] instead of pathways .
Stoichiometric methods have been widely used in metabolic engineering for over 20 years [59] , the most used of which is Flux Balance Analysis ( FBA ) [26] , [27] , [60] . FBA aims at determining the fluxes through each one of the metabolic reactions in an organism . Thus , FBA relies on the determination of a stoichiometric matrix that represents all the reactions and metabolites in an in silico organism , and a vector of uptake fluxes . In the matrix , each row corresponds to a reaction , and each column to a metabolite . is the stoichiometric coefficient of metabolite in reaction . The vector of uptake fluxes can have a non-zero entry for every transport reaction that moves a nutrient into the in silico organism . As a simple example , consider a metabolic network with three reactions: ( 7 ) where represents the transport reaction for uptaking Glucose from the environment . The stoichiometric matrix for metabolic network ( 7 ) is: ( 8 ) Assuming a steady-state concentration of every metabolite and requiring mass-conservation we must impose that ( 9 ) where is the set of fluxes through each metabolic reaction . In order to solve Eq . ( 9 ) , we need to provide the vector . The default flux for each uptake reaction is , meaning that the nutrient is not in the medium or that it cannot be uptaken by the organism . We set if nutrient is present in the medium and could be taken up by the organism . The system in Eq . ( 9 ) has many solutions . In our analysis , the biologically relevant metabolic state is the one that maximizes biomass production [59] . The problem of finding the fluxes through the reactions with a cost function that have to satisfy a number of constraints is a standard linear optimization problem that can be numerically solved using the subroutines provided in GLPK [61] . In order for us to model whether a nutrient can be a source of carbon and the biomass yield of the nutrient , we control the data in two ways . First , for some organisms the uptake of specific nutrients can lead to without being considered significant [8] . Secondly , ATP hydrolysis is integrated into the biomass of the in silico organisms , and is typically trained to better match empirical results based on growth rates . We are not exploring growth rates and we thus adjust the ATP hydrolysis in the biomass of the in silico organisms in order to better support the conclusions we reach . These controls are explained in detail in Text S1 . We tested our model on a large number of complex media generated using the nutrients available for uptake in the in silico organisms . Because the number of possible combinations of these nutrients was too large for us to test each one computationally , we considered an ensemble of 1000 randomly generated complex media . For each complex medium , every nutrient was made available for uptake with the probability , and was excluded with the probability . We generated ensembles for , , and ( for a total of 3000 random media ) . Sugars present an unusual case because a microbe such as E . coli has been known to exhibit diauxic growth [65]–[67] . This means that microbes regulate sugar uptake so that , despite having various sugars present in the medium , they will only take up one sugar at a time . Our model does not take gene regulation into account , and therefore we manually limited the number of sugars presents in each complex medium to one , specifically glucose .
|
The ability of microbial species to consume compounds found in the environment to generate commercially-valuable products has long been exploited by humanity . The vast untapped diversity of microbial species offers a wealth of potential resources . However , little is known about most microbial species . While the metabolic network of an organism can be studied to find its nutritional requirements , we lack a reliable metabolic reconstruction for most species . We use in silico organisms to systematically explore whether an arbitrary nutrient can stimulate growth as a single source of carbon , and how effectively it can be used by the organism . We find that we can predict whether a nutrient is a source of carbon and the biomass yield of that nutrient with a simple model that transcends the diversity of species and their environments . Our model for catabolic potential can therefore be used as a baseline model for any microbe for which we lack a metabolic reconstruction .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"biotechnology",
"bioengineering",
"systems",
"biology",
"biological",
"systems",
"engineering",
"biology",
"computational",
"biology",
"metabolic",
"networks",
"genetics",
"and",
"genomics",
"engineering"
] |
2012
|
Phenomenological Model for Predicting the Catabolic Potential of an Arbitrary Nutrient
|
Cervical cancer is one of the leading causes of cancer death in women worldwide . The causative agents of cervical cancers , high-risk human papillomaviruses ( HPVs ) , cause cancer through the action of two oncoproteins , E6 and E7 . The E6 oncoprotein cooperates with an E3 ubiquitin ligase ( UBE3A ) to target the p53 tumour suppressor and important polarity and junctional PDZ proteins for proteasomal degradation , activities that are believed to contribute towards malignancy . However , the causative link between degradation of PDZ proteins and E6-mediated malignancy is largely unknown . We have developed an in vivo model of HPV E6-mediated cellular transformation using the genetic model organism , Drosophila melanogaster . Co-expression of E6 and human UBE3A in wing and eye epithelia results in severe morphological abnormalities . Furthermore , E6 , via its PDZ-binding motif and in cooperation with UBE3A , targets a suite of PDZ proteins that are conserved in human and Drosophila , including Magi , Dlg and Scribble . Similar to human epithelia , Drosophila Magi is a major degradation target . Magi overexpression rescues the cellular abnormalities caused by E6+UBE3A coexpression and this activity of Magi is PDZ domain-dependent . Drosophila p53 was not targeted by E6+UBE3A , and E6+UBE3A activity alone is not sufficient to induce tumorigenesis , which only occurs when E6+UBE3A are expressed in conjunction with activated/oncogenic forms of Ras or Notch . Finally , through a genetic screen we have identified the insulin receptor signaling pathway as being required for E6+UBE3A induced hyperplasia . Our results suggest a highly conserved mechanism of HPV E6 mediated cellular transformation , and establish a powerful genetic model to identify and understand the cellular mechanisms that underlie HPV E6-induced malignancy .
Cervical cancer is the fourth leading cause of cancer death in women worldwide with ~500 , 000 new cases of cervical cancer annually and ~250 , 000 deaths worldwide . The main causative agents of cervical cancer are the high-risk human papillomaviruses ( HPVs ) . HPVs can induce hyperproliferative lesions in epithelia and are responsible for >90% of cervical and anal cancer , and more than 50% of vaginal , vulvar , penile and oropharyngeal cancers as well as a significant number of head and neck squamous cell carcinomas [1–4] . Although HPV vaccines are now available , it is still very important to understand the mechanism of HPV-induced tumorigenesis , given the 20 years or so lag between infection and cancer development and the low rates of vaccine uptake in many regions [5–7] The HPV oncogenes , E6 and E7 , are key to the cell transformation that underlies HPV-mediated cancer . Multiple studies have shown that E6 and E7 work cooperatively to induce carcinogenesis [8 , 9] . E7 is critical for early stages of tumor formation , causing benign tumors and targeting Rb , whereas E6 is thought to play an important role during the later stages of tumor progression [10–14] . E6 inactivates a range of targets including the tumor suppressor protein p53 and important polarity regulators including hDlg1 , Scribble/Vartul and MAGI-1 , all of which are directed for ubiquitin-mediated proteasomal degradation [14–19] . E6 directs the degradation of many of its substrates through recruitment of an E3 ubiquitin ligase , UBE3A/E6AP , with which it forms a stable complex and redirects its activity towards the different E6 target proteins [20 , 21] . Multiple functions of E6 , including interaction with UBE3A , p53 and PDZ domain-containing substrates , appear to be required for its ability to bring about cell transformation and to contribute towards malignancy in animal models . The PDZ binding motif ( PBM ) of E6 is almost exclusively found in the high-risk HPV types , and is essential for HPV-mediated malignancy and hyperplasia [22] . However , a major question remaining is how E6-mediated degradation of PDZ proteins leads to cellular transformation and malignancy within a living animal . Transgenic mice models of HPV 16 and 18 E6 have contributed to our understanding of HPV-mediated tumorigenesis but the underlying cellular mechanisms and the hierarchy in importance of the PDZ target proteins have not been elucidated . Drosophila melanogaster has been widely and successfully used as a powerful genetic model organism to study human diseases , including cancer , owing to the strong conservation of genes and signaling pathways between Drosophila and human . Many tumor suppressor genes , including Dlg , Scribble , and Lgl , as well as oncogenic signaling pathways such as Notch , were first identified in Drosophila [23–27] . Many epithelial derived tumors have been modeled in flies . For example , oncogenic Ras or Notch paired with loss of function mutations in Scribble result in the formation of metastatic tumors in Drosophila that share many characteristics with human tumors [28–31] . In this study we have developed an in vivo model of HPV E6-mediated cellular transformation using Drosophila . In this model the levels of E6 expression are high and reflect the higher levels of E6 expression seen during the later stages of malignant progression . We show that coexpression of HPV E6 and human UBE3A/E6AP in the wing and eye results in severe morphological defects , whereas E6 or UBE3A expression alone results in none . We find that E6 , in cooperation with UBE3A , targets PDZ proteins in a PBM-dependent manner , and these targets include Magi , Dlg and Scribble with Magi being a major degradation target . In contrast , Drosophila p53 was not degraded by E6+UBE3A . In addition to the loss of PDZ scaffolding proteins , E6+UBE3A expression in epithelia led to apoptosis paired with delamination . Importantly , Magi overexpression rescued the cellular abnormalities caused by E6+UBE3A and the Magi PDZ domains were necessary for this rescue . E6+UBE3A activity was not sufficient to induce tumorigenesis , but did result in malignancy when co-expressed with either an activated form of oncogenic Ras or Notch . These cells had the hallmarks of epithelial-mesenchymal transition ( EMT ) including morphological changes paired with elevated MMP1 and aPKC expression . To identify signaling pathways that modulate the E6+UBE3A effects , we conducted a genetic screen and found that E6+UBE3A interacted with the insulin receptor . Overall , our results establish the first Drosophila model to study the HPV E6-mediated cellular transformation and malignancy and suggest a high degree of conservation on the mechanism of HPV E6 mediated cellular transformation .
The following Drosophila strains were used , UAS-Magi::Cherry[32] , UAS-Dlg::GFP [33] , UAS-MagiΔPDZ and UAS-MagiΔWW from [34] , UAS-hUBE3A from [35] , Scrib::GFP [36] , UAS-Scrib::GFP [37] . UAS-p35 , UAS-mCD8::GFP , UAS-InR , UAS-InR DN , UAS-ArmS10 , EGFR DN , UAS-rlSem , UAS-Yki::GFP , UAS-Notch Intra , UAS-Bsk DN , UAS-Ras85D . V12 , P53::GFP , UAS-P53 H159N , UAS-Debcl , buffyH37 , DebclE26 , apterous-Gal4 , GMR-Gal4 , Gal80ts and all the transgenes and mutants used in the Table 1 were from the Bloomington Drosophila Stock Center . HPV18-E6WT and HPV18-E6V158A are described in Thomas et al ( 2005 ) [18] . Each cDNA was re-derived by PCR using oligos: 5’-accggtATGGCGCGCTTTGAGGATC-3’ for the 5 prime side of both clones and either 5’-ctcgagTTATACTTGTGTTTCTCTGC-3’ for the wild type 3 prime end , or 5’-ctcgagTTATAGTTGTGTTTCTCTGC-3’ for the Val-to-Ala mutant 3’ end . The underlined codons correspond to the mutated amino acid . All products were subcloned into pGEM-T ( Promega ) and verified by sequence analysis . Each product was subcloned into pBluescript ( KS+ ) ( Stratagene ) using Not I and Sac II restriction sites , then transferred into pUASTattB [38] modified to contain either Myc , HA , or Flag epitope tags 5’ to the multiple cloning site , using Age I and Xho I restriction sites . Injections and integrase-mediated insertion into w-; P ( CaryP ) attP2 was carried out by Rainbow Transgenic Flies , Inc ( Camarillo , CA , USA ) . For immunolabeling , wing discs from wandering third instar larvae and pupal eyes 42 hrs after puparium formation were dissected in PBS and fixed in 4% formaldehyde . Fixed tissues were washed three times in PBS solution containing 0 . 1% Triton-X-100 and blocked in 5% normal goat serum for 1 hour before incubation with primary antibodies . The primary antibodies used in this study were rabbit anti-Magi 1:200 [39] , rabbit anti-Baz 1:1000 [40] , mouse anti-Dlg 1:50 ( 4F3 , Developmental Studies Hybridoma Bank ) , rat anti-DE-cadherin DCAD2 1:50 ( Developmental Studies Hybridoma Bank ) , mouse anti-Flag 1:300 ( M2 , Sigma ) , rabbit anti-cleaved Cas3 1:300 ( Cell Signaling ) , rabbit anti-Myc 1:200 ( Abcam ) and rabbit anti-HA 1:200 ( Abcam ) [41] , rat anti-Magi 1:300 [42] , mouse anti-Arm 1:50 ( Developmental Studies Hybridoma Bank ) , mouse anti-MMP1 1:100 ( Developmental Studies Hybridoma Bank ) , rabbit anti-aPKC zeta C20 1:1000 ( Santa Cruz Biotechnology ) . The appropriate secondary antibodies were conjugated Alexa488 , Alexa594 , and Alexa 647 ( Invitrogen ) . Images were collected using a Leica SP8 Scanning multiphoton confocal microscope , processed in ImageJ . Figures were assembled using Adobe Photoshop . Low magnification images were collected on a Zeiss Axioskop with a 5X NA 0 . 50 air lens and AxioVision software . For quantification of protein levels , the fluorescence intensity at the plasma membrane was measured for a standardized ROI in both the overexpression and non-overexpression sides , using the “Find Edges” function of ImageJ followed by threshold adjustments and measurement of intensity . The data were then transferred to Excel for further analysis and plot creation . For each experiment five wing imaginal discs were analyzed and an unpaired t-test was used for statistical analysis . For quantification of the degree of rescue by Magi::cherry , MagiΔWW , MagiΔPDZ , Dlg::GFP , Scrib::GFP , flies that fulfilled all the following criteria were considered rescued: 1 . full wings , 2 . only small melanized blisters , 3 . fully differentiated wing veins and margins . Flies with rescued wing phenotypes were counted and divided by the total number of flies assayed and percentage was calculated . For SEM of fly eyes heads were fixed in 4% glutaraldehyde in PBS , washed three times to remove the fixative and subsequently dehydrated in an ethanol series . After dehydration samples were critical point dried and mounted on SEM stubs followed by sputter coating with a thin layer of AuPd . Samples were imaged using a Zeiss Neon high-resolution scanning electron microscope . These assays were performed as described previously [43] . Briefly , Drosophila Magi and mammalian MAGI-1 proteins were transcribed and translated in vitro , using the TnT kit ( Promega ) , and radiolabelled with [35S]-Methionine ( GE Healthcare ) . They were incubated at 30°C for the indicated times , with or without the addition of similarly translated HPV-16 and HPV-18 E6 proteins . The remaining proteins were detected by SDS-PAGE and autoradiography .
While transgenic mouse models of HPV E6 have significantly contributed to the understanding of HPV-mediated tumorigenesis , the role of the HPV-E6 PDZ targets in vivo has not been clearly established . As there are no other in vivo models of HPV E6 at present , and given the wide array of genetic tools and techniques available in fruit flies , we felt that developing a fly E6 model would provide the ability to further dissect the molecular pathways and identify novel partners involved in HPV E6-mediated cellular transformation . To do this , we expressed genes encoding the human HPV18 E6 tagged with Myc in two separate epithelia using the Gal4-UAS high-expression system [44] . When we expressed E6 in the wing or eye epithelia , using the apterous-GAL4 and GMR-GAL4 , respectively , we did not detect any abnormalities within these tissues ( Fig 1A , 1B , 1D , 1E , 1G and 1H ) . E6 requires the human E3 ubiquitin ligase ( UBE3A/E6AP ) for many of its functions . While Drosophila has a UBE3A homologue it was possible that HPV E6 was unable to activate or interact with the Drosophila UBE3A . This suggests that the human UBE3A has a specific interaction with E6 or a function that is not conserved in the Drosophila UBE3A protein . The former is likely as the LXXLL motif necessary for E6 binding to human UBE3A [45] is absent from Drosophila UBE3A ( this study ) . In support of this we found that expression of E6 together with a previously established transgene that expresses human UBE3A [35] resulted in severe abnormalities in wing and eye epithelia ( Fig 1C , 1F and 1I ) . These abnormalities included smaller and blistered wings that were full of melanized tissue and were held out . Similarly , co-expression of E6 and human UBE3A in the eye resulted in rough eyes with disorganized and fused ommatidia as well as increase in number of bristles ( Fig 1F and 1I ) . These abnormalities were specific to co-expression of E6 and UBE3A and were 100% penetrant , whilst neither E6 nor UBE3A expression alone resulted in any defects . These results show that expression of E6+UBE3A is deleterious in Drosophila epithelia , suggesting a conservation of cellular targets downstream of the E6+UBE3A complex . The wing and eye phenotypes prompted us to ask whether any of the known targets of E6 in human cells are also targeted in Drosophila epithelia . E6 targets PDZ domain proteins of the polarity and junctional network including MAGI-1 , hDlg1 and Scribble/Vartul for ubiquitin-mediated proteasomal degradation [14–19] , an activity that requires an intact PBM at the extreme C terminus of E6 [22] . We focused on the wing imaginal disc using the apterous-GAL4 driver to drive expression in the dorsal half of the wing disc to allow for direct comparison of protein levels in the presence and absence of E6+UBE3A . We focused our attention on the Drosophila homologues of Magi , Discs-large ( Dlg ) and Scribble ( Scrib ) and monitored the protein levels on the dorsal ( expressing E6+UBE3A ) versus ventral ( lacking E6+UBE3A ) side of the wing imaginal disc ( Fig 2 ) . We found that E6+UBE3A co-expression led to a significant loss of Magi from the adherens junction domain ( Fig 2A–2D ) , a modest reduction in the levels of Dlg ( Fig 2E–2H ) , and a weak reduction in the levels of Scrib ( Fig 2I–2L ) at their respective locations in the septate junction . In order to determine if all potential PDZ proteins were effectively targeted by E6 , we examined the levels and localization of two other PDZ proteins that are known to be in cell junctions , Bazooka ( Par-3 ) or Par-6 , and found no effect on the levels or the localization of these PDZ protein ( S1A–S1H Fig ) . These results indicate that E6 binding and degradation of a select group of PDZ proteins is specific . Similarly , there were no changes in the level or localization of E-cadherin at the adherens junction ( S1I–S1L Fig ) . Expression of E6 or UBE3A alone did not result in any changes in the levels or localization of Magi , Dlg or Scribble ( S1M–S1X Fig ) , confirming that alterations to the PDZ proteins are a result of the E6 and UBE3A complex . Our results are consistent with previous results from human cervical cancer cells in which MAGI-1 , hDlg-1 and hScrib are targeted for degradation by E6 . In contrast , Drosophila p53 was not targeted for degradation by E6+UBE3A ( Fig 2M–2P ) , which is consistent with evolutionary differences between human and fly p53 . Our results suggest that Drosophila Magi is particularly susceptible to E6 targeting . This is consistent with studies performed in mammalian cells , where MAGI-1 is one of the most strongly bound E6 PDZ targets and is also very efficiently degraded [15 , 43 , 46 , 47] . This suggests that the mechanism of E6 targeting these PDZ cell polarity-regulating proteins is conserved between humans and Drosophila . As Magi appears to be a major target of HPV E6+UBE3A in Drosophila we were interested in directly comparing the ability of E6 to degrade human MAGI-1 and Drosophila Magi in vitro ( Fig 2V ) . Drosophila Magi and human MAGI-1 were transcribed and translated in vitro , using the TnT rabbit reticulocyte lysate system , which includes functional UBE3A/E6AP [48] . These proteins were then incubated with similarly translated HPV-16 or HPV-18 E6 and the remaining protein analysed by SDS PAGE and autoradiography . As can be seen from Fig 2V , Drosophila Magi and mammalian MAGI-1 are almost equivalently susceptible to degradation induced by the high-risk cancer-causing HPV 16 and 18 E6 proteins . These findings highlight the high degree of evolutionary conservation in this particular HPV E6 target , and provide the molecular basis for the results obtained in vivo . HPV E6-mediated degradation of PDZ proteins in human epithelial cells requires an intact PBM at the C-terminus of the E6 protein [22] . We next tested whether the loss or reduction of PDZ proteins in the Drosophila epithelial cells was also dependent upon an intact E6 PBM . To do this , we generated a transgene expressing E6 with a point mutation that disrupts the E6 PBM ( E6V158A ) and prevents E6 binding to PDZ domain proteins [43] . When co-expressed with UBE3A , the E6V158A mutant did not trigger a loss of Magi ( Fig 2Q–2T ) , Dlg or Scrib ( S1Y-S1Z’ Fig ) . These results confirm that E6 targeting of Drosophila Magi requires an intact E6 PBM . The columnar epithelia of the Drosophila imaginal disc are ideal for studying cellular transformation and cellular signaling pathways , and have been used as in vivo models for detailed analysis of cellular and molecular mechanisms underlying cell polarity defects and cancers [29 , 30 , 49] . The eye disc in particular has been an excellent model to study the molecular and signaling mechanisms that underlie cell transformation and progression to cancer . When E6+UBE3A was co-expressed in the eye the earliest cellular phenotypes were observed during pupal stages of development . We observed a loss of Magi ( Fig 3C ) and a reduction in Dlg in the eye imaginal disc in the presence of E6+UB3EA ( Fig 3F ) : this mirrors what we observed in the wing imaginal disc where Magi levels were extensively reduced while Dlg was less so . E6+UBE3A co-expression in the eye resulted in severe tissue abnormalities . In the normal eye ommatidia are arranged in a hexagonal array to produce a stereotyped pattern . Each ommatidium consists of eight photoreceptor cells , four cone cells , three types of pigment cells ( primary , secondary and tertiary ) and bristle cells that form the bristles in the adult eye [50] ( Fig 3P ) . In eyes coexpressing E6 and UBE3A the overall structure of the eye and the organization of ommatidia was disrupted . Frequently , neighboring ommatidia were fused ( 70% of ommatidia; n = 10 discs ) and there was an increase in the number of cone cells , primary , secondary and tertiary pigment cells , as well as bristle cells ( Fig 3F and 3I ) . These cellular defects were evident in the adult eye with an increase in the number of bristles and the rough or glossy eye phenotype ( Fig 1G–1I ) . Immunolabeling for junctional and polarity markers , Ecad , beta-catenin ( Drosophila Armadillo , Arm ) and Par3 ( Drosophila Bazooka , Baz ) , revealed that the organization of photoreceptor cells within each ommatiduim was perturbed and that junctional as well as polarity proteins were mislocalized ( Fig 3G–3O ) . The displacement of these markers suggests that E6 disrupts the cell polarity and junctional integrity of photoreceptors in ommatidia . To determine whether the effects of E6+UBE3A were dosage dependent , we increased the expression of E6 and UBE3A by increasing the temperature to 29°C to increase the efficacy of Gal4 [51] . At 29°C , E6+UBE3A triggered extensive apoptosis on the apterous side compared with the control wildtype side , as detected by an antibody to cleaved-Caspase 3 ( Fig 4D–4F ) . Normally , expression of the baculovirus protein p35 blocks apoptosis , leading to compensatory proliferation or apoptosis-induced proliferation [52–54] . Blocking apoptosis with p35 [41] did not result in overgrowth when co-expressed with E6+UBE3A , but instead resulted in clusters of cells that expressed high levels of the polarity proteins , Baz and aPKC ( Fig 4J–4L ) in all discs observed ( n = 20 ) . Increased expression of vertebrate Par3 ( Baz ) and aPKC are associated with tumorigenesis [55] and progressive stages of cancer and EMT [56–58] . A common marker of transformation and EMT is the increased expression of matrix metalloproteinase 1 ( MMP1 ) [59 , 60] and we observed that all cell clusters expressed high levels of MMP1 ( Fig 4J–4R ) , suggesting that these cells were undergoing EMT . The cell clusters were not within the columnar epithelia , but were found under the epithelium at the basal side ( Fig 4J’–4R’; arrowheads ) . We also observed individual cells expressing high levels of MMP1 and polarity proteins away from the cell clusters within the basal side ( Fig 4P–4R , arrow ) . Conversely , when we examined the level and localization of the adherens junction protein Ecad , we did not detect any increase in the delaminated cell cluster . These results suggest that HPV E6 , in cooperation with UBE3A , triggers apoptosis and subsequent cell delamination when apoptosis is blocked . Consistent with previous studies from vertebrates , E6+UBE3A expression alone was insufficient to induce cellular transformation and only when it is paired with processes that block apoptosis is cell transformation observed . As c-Jun N-terminal kinase ( JNK ) is one of the main signaling pathways triggering apoptosis and MMP1 expression , we expressed a dominant negative form of Drosophila JNK ( bskDN ) to block JNK signaling in wing discs co-expressing E6+UBE3A . Blocking JNK signaling did not suppress the E6+UBE3A-mediated cell death ( Fig 4S–4U ) , suggesting that the E6+UBE3A mediated apoptosis did not involve the JNK signaling pathway and was driven through another cellular pathway . As Magi was strongly reduced by E6+UBE3A expression , we next asked whether Magi overexpression could suppress and rescue the defects caused by E6+UBE3A in the wing epithelia . We found that co-overexpression of a Cherry-tagged form of Drosophila Magi ( Magi::Cherry ) and E6+UBE3A partially rescued the adult wing phenotypes ( Fig 5B ) . Specifically , there was less melanized tissue , a reduced degree of blistering and the flies exhibited full wings with fully differentiated veins and wing margins . We quantified the degree of rescue of the wing phenotypes and found a significant reduction in aberrant wing phenotypes when Magi was co-expressed with E6+UBE3A ( Fig 5G ) . Consistent with this result , E6+UBE3A expressed in wing discs in a null mutant of Magi ( Magibst ) [32] resulted in pupal lethality , suggesting an enhancement of the phenotype . To determine whether Magi-mediated rescue of E6+UBE3A was specific , we examined the rescue capability of other E6 PDZ target proteins including Dlg and Scrib; neither Dlg nor Scrib could suppress the effects of E6+UBE3A ( Fig 5C , 5D and 5G ) . Examination of wing discs of third instar larvae revealed that overexpression of Magi completely blocked the apoptosis seen in E6+UBE3A co-expressing epithelia ( n = 20 discs ) , whereas overexpression of Scrib did not ( n = 10 discs ) ( Fig 5H–5M ) . These results suggest that Magi-mediated rescue of wing abnormalities is specific and that expression of Magi can block or reduce the defects caused by E6+UBE3A co-expression . Next , we asked which domains of Magi were responsible for suppressing the E6+UBE3A mediated defects . Drosophila Magi contains four PDZ domains and two WW domains . We expressed Magi transgenes lacking either the two WW domains ( MagiΔWW ) or the PDZ domains ( MagiΔPDZ ) [34] in the wing disc , along with E6+UBE3A . Expression of MagiΔWW significantly rescued the wing abnormalities caused by E6+UBE3A ( Fig 5F and 5G ) . Conversely expression of MagiΔPDZ failed to rescue the E6+UBE3A-mediated wing defects ( Fig 5E and 5G ) . These results suggest an essential role for the PDZ domains of Magi in blocking the deleterious effects of E6+UBE3A . While Magi is unlikely to be the sole PDZ protein degraded by E6+UBE3A , our data suggests that Magi is an important degradation target and that the highly conserved PDZ domains play a critical role in targeting by HPV E6 and human UBE3A . As neither Magi::Cherry nor MagiΔWW were able to fully rescue the defects caused by E6+UBE3A , we tested the degree to which each could be targeted for degradation by E6 . We examined the levels and localization of Magi::Cherry as well as the MagiΔWW and MagiΔPDZ mutants in wing imaginal discs co-expressing E6+UBE3A . Using apterous-GAL4 to drive expression in the dorsal half of the disc , both Magi::Cherry and MagiΔWW proteins were reduced at the plasma membrane although this reduction was not uniform ( Fig 6B and 6D ) . The Magi::Cherry expressed alone is uniformly distributed around the membrane and found in prominent intracellular puncta ( Fig 6A ) , whilst in the presence of E6+UBE3A Magi::Cherry levels are reduced in the puncta and at the membrane . Conversely the localization and expression level of MagiΔPDZ was not affected when co-expressed with E6+UBE3A ( Fig 6E and 6F ) . These results demonstrate that E6 was capable of degrading a proportion of the wild type Magi::Cherry and MagiΔWW proteins even when these were overexpressed , and this is likely the reason why a full rescue of the wing abnormalities was not obtained . In humans there is usually a period of 15–20 years from the time of HPV infection to the development of cancer . Transgenic mice also showed a latency of 16–20 months for cancer development in presence of HPV E6 [61] . These results suggest that cooperation between E6 and UBE3A is insufficient to cause uncontrolled growth and malignant transformation of epithelia , and that genetic events , such as genomic instability and spontaneous mutation , may also play a role in E6-induced epithelial transformation . Our results are consistent with this view , as co-expression of E6+UBE3A was not sufficient to cause cellular transformation and cancer . Mutations in oncogenes such as Ras have been implicated in cancer progression and HPV-mediated tumorigenesis [61–63] . To investigate this in our model system we expressed E6+UBE3A in epithelial cells that expressed an activated Ras , Ras85DV12 . As expression of Ras85DV12 driven by apterous-Gal4 is larval lethal , we carried out a temperature shift experiment where Gal4 was silenced during embryonic development using a temperature-sensitive Gal4 inhibitor , Gal80ts [64 , 65] . A shift to 29°C during the second instar larval stage activated Gal4 , and the wing imaginal discs of third instar larva after 24 hours at 29°C showed epithelial cells that expressed high levels of E6 and had altered morphology . These cells morphologically resembled mesenchymal cells ( Fig 7A and 7B ) , in that they acquired a fibroblast-like , flat morphology ( Fig 7B ) , they displayed filopodial-like processes ( Fig 7C; arrow ) , and were delaminated at the basal side of the columnar epithelia ( Fig 7D; arrow ) . However , only a subset of cells on the apterous side expressed high levels of E6::myc ( Fig 7B ) . As we found the effects of E6 to be dosage-dependent , we increased the timing of expression of E6+UBE3A to 48 hours by shifting to 29°C during the first instar larval stage . Wing imaginal discs of third instar larva after 48 hours of expression had clusters and individual cells that expressed high levels of MMP1 ( 100% penetrant; n = 20 discs ) ( Fig 7F , arrow ) . Cell clusters also had extending filopodial-like processes ( Fig 7G; arrow ) . Similar to when E6+UBE3A was co-expressed with p35 ( Fig 4J’ , 4M’ and 4P’ ) , these clusters were found under the epithelium at the basal side ( Fig 7H , arrow ) and were limited to the apterous side of the disc with no spread into the wildtype , ventral compartment of the wing disc . Overexpression of Ras85DV12 in the absence of E6+UBE3A caused disc overgrowth as expected , but no mesenchymal-like cells or clusters expressing MMP1 were detected ( Fig 7E ) . These results indicate that the combination of HPV E6 and UBE3A is insufficient to cause cellular transformation , which requires the additional cooperation of cellular oncogenes , such as activated Ras . It is also clear from this analysis that only a small proportion of E6-expressing cells underwent transformation , as initially we only saw single cells with altered morphologies , which , over time , expanded into clusters in a manner analogous to the clonal development of HPV-induced malignancies . As HPV E6 has been implicated in the later stages of tumorigenesis , we wanted to determine whether coexpression of E6+UBE3A and Ras85DV12 could cause malignancy and metastasis . In order to do this we expressed E6+UBE3A , Ras85DV12 and a membrane-bound GFP marker ( mCD8::GFP ) using GMR-Gal4 in the eye , so that we could readily detect the migration of malignant cells into other regions of the body . Expression of E6+UBE3A in the presence of Ras85DV12 resulted in cellular migration , with many GFP-positive cells detected in the abdomen of these animals ( Fig 7J , arrow ) , distant from the source of E6+UBE3A+Ras85DV12 expression in the eye imaginal disc ( Fig 7J , arrowhead ) ( 70% penetrant , n = 20 animals ) . GFP-positive cells were not detected in the abdomen when Ras85DV12 was expressed alone ( Fig 7I ) ( n = 10 animals ) . Expression of Ras 85DV12 alone or in combination with E6 and UBE3A caused pupal lethality with extensive tissue necrosis making the analysis of metastasis in the adult impossible . Similar results were obtained when E6+UBE3A were co-expressed with an activated form of Notch ( NotchACT ) where E6+UBE3A+NotchACT was lethal at the pharate adult stage ( prior to hatching ) . NotchACT plus E6+UBE3A cells labeled with GFP were observed in the abdomen ( Fig 7L , arrow ) distant from the eye imaginal disc ( Fig 7L , arrowhead ) ( 40% penetrant , n = 30 animals ) . GFP positive cells were never observed in the abdomen when NotchACT was expressed alone ( Fig 7K ) ( n = 20 animals ) . These results indicate that E6 + UBE3A , when expressed with activated Ras or Notch , leads to many of the cellular phenotypes associated with EMT and spread of transformed cells throughout the body . In order to test if these phenotypes were dependent on the function of the PDZ binding motif of E6 , we expressed the E6V158A mutant that is deficient in binding to PDZ proteins in conjunction with RaS85DV12 and UBE3A . In the absence of the PDZ protein interaction , we observed a reduced level of EMT and cell migration with a penetrance of 50% ( n = 30 animals ) ( Fig 7M ) . Expression of E6V158A in the presence of activated Notch and UBE3A did not result in cell migration away from the eye disc ( 100% penetrant , n = 20 ) ( Fig 7N ) . These results collectively suggest that E6 targeting of PDZ proteins plays a major role in EMT and cell migration induced by E6+UBE3A . Given the high degree of conservation of its genes and signaling pathways , the Drosophila eye appeared to be an ideal model system to identify the signaling pathways involved in E6+UBE3A-mediated cell abnormalities . Therefore , we carried out a small genetic screen to identify which signaling molecules and pathways functionally interact with E6 . We tested a range of signaling pathways ( Table 1 ) and observed no enhancement or suppression of the E6+UBE3A phenotypes , with the exception of the insulin receptor . Interestingly , we found no interactions with pathways known to influence apoptosis , such as the JNK pathway , p53 and mutants in the mitochondrial apoptosis pathway ( Buffy and Debcl ) ( S2 Fig ) , known to function downstream of the Drosophila retinoblastoma protein ( pRb ) [66] . In particular , blocking p53 function using a DN form [67] did not rescue the eye phenotypes ( S2 Fig ) . On the other hand , when we blocked insulin signaling by expressing a dominant negative form of the insulin receptor ( InRDN ) in conjunction with E6+UBE3A , the resulting eyes displayed large necrotic scars ( 100% of eyes , n = 70 ) , which is an indication of cell death ( Fig 8D ) . Co-expression of the insulin receptor ( InR ) in E6+UBE3A-expressing eyes resulted in hyperplasia ( Fig 8F; 100% of eyes , n = 100 ) compared with the expression of insulin receptor alone , which generated bigger eyes but no hyperplasia ( Fig 8E ) . In contrast , changes to the EGF receptor ( EGFR ) ( Fig 8J ) , or the MAP kinases ERK ( Drosophila Rolled ) ( Fig 8H ) and JNK ( Drosophila Basket ) ( Fig 8O ) , had no effect on the E6+UBE3A eye phenotypes . Similarly , disruption of Wnt signaling with activated beta-catenin ( Drosophila Armadillo ) ( Fig 8L ) or Hippo signaling ( Drosophila Yorkie ) ( Fig 8N ) had no effect . Thus , none of the other pathways tested showed any effect upon the E6+UBE3A phenotype , indicating that the effect of the insulin receptor is specific and suggesting that changes in insulin signaling may play a role in in the cell transformation and cancer progression induced by HPV .
In this study we have developed a new model of E6+UBE3A-mediated cell transformation in Drosophila melanogaster , an excellent system for further investigating the molecular mechanisms underlying E6-mediated cellular transformation . In our model we found that E6 expression alone is insufficient to disrupt the host cellular function but dysfunction also requires the human E3 ubiquitin ligase , UBE3A [20 , 21] . These results suggest that Drosophila UBE3A does not interact with the E6 protein , and this would be expected as Drosophila UBE3A lacks the critical interaction site which is bound by E6 on human UBE3A [68] . In support of this , co-expression of E6 and human UBE3A in the eye and wing resulted in severe defects . E6+UBE3A were insufficient alone to induce cellular transformation . When E6+UBE3A expression was combined with expression of activated Ras or Notch , it resulted in cellular transformation , EMT and cell migration throughout the body . This is consistent with previous findings in transgenic mouse models where activated Ras , in combination with E6 , results in the formation of malignant tumors [62] . Similarly , mutations in the Ras oncogene have been reported in HPV tumors [61 , 63] . Our results are consistent with these results , and together they support the notion that HPV alone is insufficient to generate malignant tumors and a second hit or mutation in a cellular oncogene is necessary . The expression of the HPV E7 protein in our model would be of interest to determine if the presence of E7 could increase the degree of cell transformation or allow cell transformation to occur at earlier stages in the life cycle . We further show that E6 targets a number of PDZ domain-containing proteins in Drosophila with apparently different degrees of degradation . Magi is a major target of E6 , showing a markedly greater susceptibility to E6-induced degradation compared with other polarity proteins such as Dlg and Scrib . Previous studies have also identified MAGI-1 as a major target of high-risk human papillomaviruses HPV 16 and 18 E6 oncoproteins in vertebrate epithelial cells [15] . We found that Drosophila Magi is also highly susceptible to E6-induced degradation in vitro , in a manner analogous to that of MAGI-1 . These findings suggest that there is a hierarchy of E6 targets in flies , and indicates that a number of cellular pathways targeted by E6 are conserved between insects and mammals . Interestingly , we did not observe any degradation of P53 in the presence of E6+UBE3A and an expression of a dominant negative form of P53 in E6+UBE3A expressing eyes did not have any effect on the cellular defects caused by E6+UBE3A co-expression . P53 is a key target of HPV E6-mediated polyubiquitination and degradation [69] . Our results collectively suggest that the cellular defects caused by E6+UBE3A co-expression in Drosophila are independent of P53 degradation , making our model system potentially useful for investigating the other activities of E6+UBE3A on cellular transformation . The degradation of Magi and the other PDZ domain-containing proteins requires an intact E6 PBM , demonstrating that the mechanism of interaction between E6 and its PDZ containing targets is conserved between insects and mammals . The cellular defects caused by E6+UBE3A co-expression could be partly due to loss or disruption of a suite of PDZ domain proteins targeted by E6 . Both Dlg and Scrib are polarity proteins and a reduction of these regulators , or a disruption in the balance between the levels of these proteins , could affect the polarity of cells as suggested by the mislocalization of polarity protein Baz ( Par3 ) in the eye . Magi plays a role in remodeling adherens junctions in interommatidial cells in the eye , and loss of Magi disrupts the organization of the interommatidial cells [70] . Consistent with this we saw disorganization of interommatidial cells and mislocalization of the adherens junction protein Armadillo ( beta-catenin ) , suggesting that loss of Magi may play a role in the development of E6+UBE3A-mediated cellular abnormalities . In support of this , overexpression of a full-length Magi , or a transgene lacking the two WW domains , partially rescued the E6+UBE3A-mediated defects , suggesting that disruption of Magi function was in part responsible for the observed phenotypes . Interestingly , a mutant Magi lacking the PDZ domains was unable to rescue the E6+UBE3A-induced defects , suggesting that a function of Magi involving PDZ interactions is essential for this activity , and that this is the function of Magi that is targeted by E6 and UBE3A . The incompleteness of the rescue obtained with wild type Magi could be for a number of reasons , including E6 degradation of the overexpressed Magi , or the possibility that other cellular targets that are degraded by E6 also play an important role . Indeed , many other PDZ proteins have been reported to be targets of HPV E6 including PSD95 [71] , PATJ [72 , 73] , MUPP1 [16] , TIP1 [74] , TIP2 [75] , PTPN3 [76 , 77] , PTPN13 [78] , and CAL [79] . A number of these proteins have homologs in Drosophila and are involved in important cellular processes including cell-cell junction , cell polarity and regulation of cell signaling . Hence disruption of these in combination could play a role in E6+UBE3A-mediated cellular defects . Using our model system we are in a position to determine the hierarchy and importance of targeting these PDZ proteins in HPV E6-mediated cellular transformation . Identifying Magi as a major target of HPV E6 in our model is similar to what is observed in human cells , and supports the finding that Magi has a significant biological relevance to HPV infection . Expression of a form of MAGI-1 resistant to E6-mediated degradation represses cancer cell proliferation and induces apoptosis [80] . Loss of tight junction integrity in an HPV-positive , tumor-derived cell line results from E6-mediated degradation of MAGI-1 [15] . These studies suggest that MAGI-1 is a key factor in cellular function and that its removal can contribute to cellular transformation . Strikingly , loss of the sole Magi gene in Drosophila has no deleterious effect , and animals deficient for Magi live to adulthood with no detectable abnormalities [39 , 70] . However , this does not rule out the importance of Magi as an important target of E6 , as Drosophila mutants of another major HPV target , p53 , are also adult viable . p53 mutants ( in both mouse and Drosophila ) live to adulthood with no detectable abnormalities [81 , 82] . However , these animals are more prone to developing cancer , and when exposed to stress-inducing conditions and genomic instability , such as irradiation , they fail to trigger apoptotic cell death to remove the damaged cells [82–86] . Hence , it is plausible that Magi could possess some tumor suppressor activity . Indeed , Magi , like p53 , is an extremely susceptible target for E6-mediated degradation , and therefore is unlikely to be subject to any selection pressure to mutate during the development of HPV-induced malignancy . Therefore , it is not surprising to find that , like p53 , it is wild-type in most cervical cancers , with MAGI-1 mutations detected in less than 1% of cervical cancers tested ( COSMIC , Catalogue of Somatic Mutations in Cancers: http://cancer . sanger . ac . uk/cosmic/gene/analysis ? ln=MAGI1_ENST00000402939 ) . Mutations , or aberrations in expression , of MAGI family members , have been found in global analyses of a number of different human cancers [87–92] . However it is not yet clear whether these mutations are drivers of carcinogenesis , or passengers . As the molecular mechanism and the steps that lead to tumor formation and malignancy in HPV-positive cells are largely unknown , establishing a model of E6-mediated cell transformation that is amenable to large scale , unbiased genetic analysis is essential . Our Drosophila model of HPV E6-mediated cell transformation is highly amenable to genetic manipulation , with a strong degree of conservation at the gene and cellular levels . As a first test of this model we have identified the insulin receptor as a potential partner in the cellular transformation mediated by expression of E6+UBE3A , using the classic approach of testing for genetic interaction in the Drosophila eye . A growing number of studies have implicated the insulin receptor pathway in cancer development and progression . Conditions associated with insulin resistance such as Type 2 diabetes mellitus ( T2DM ) , obesity , and metabolic syndrome are now recognized as major risk factors for development and progression of cancer [93–98] . Increased levels of insulin receptor have been reported in cancer cells [99 , 100] and insulin-like growth factor ( IGF ) signaling was shown to play a role in HPV-infected lesions and tumorigenesis [101] . Additionally , E6 can cause hyper-activation of the insulin receptor and activation of downstream signaling , including the PI3K and MAPK pathways [102] . For instance , the ability of HPV E6 plus E7 to transform keratinocytes is increased after down-regulation of the insulin-like growth factor binding protein 2 ( IGFBP2 ) [101] . IGFBP2 in this system suppresses Insulin Growth Factor ( IGFI/II ) , such that when IGFBP2 is down regulated IGFI/II stimulates the IGF receptor 1 ( IGF1R ) and AKT signaling . In high-grade pre-malignant cervical lesions infected with HPV16 the IGFBP2 levels are reduced , suggesting that changes in insulin signaling may play a key role cancer progression [101] . Earlier studies also revealed that E6 binds and degrades TSC2 , a component of insulin signaling necessary for blocking the mTOR complex and growth , thus leading to activation of mTOR signaling and induction of growth [103] . These findings , together with our results , indicate that insulin signaling plays a critical role in HPV-mediated carcinogenesis and hence it is of significant interest to further explore the underlying mechanism of hyperplasia caused by interaction between HPV E6 and the Insulin receptor identified in this study . Altogether we believe that our Drosophila E6+UBE3A model is an important new tool to study the molecular mechanism underlying HPV-mediated cancer and malignancy . As adult flies expressing E6+UBE3A exhibit severe wing and eye abnormalities this represents a uniquely valuable model with which to conduct modifier genetic screens to identify molecules that can inhibit E6-induced cellular defects . As a proof of principle , we have confirmed Magi as an essential element in the ability of E6 to induce cell transformation , demonstrated cooperation between E6 and different cellular oncogenic signaling pathways , and identified the insulin receptor as a downstream effector of E6 induced malignancy .
|
Human papillomaviruses ( HPV ) are the causative agents of cervical cancer , one of the leading causes of cancer death in women worldwide . The E6 oncoprotein encoded by HPV has been implicated in the progression of primary tumors to metastatic disease and we have developed a new model in the fruit fly ( Drosophila melanogaster ) to study the cellular effects of E6 . The E6 protein recruits an E3 ubiquitin ligase ( UBE3A ) to induce the degradation of a number of cellular proteins , including members of the MAGUK family of scaffolding proteins that control the structure and polarity of epithelial cells: Dlg , Scribble and Magi . Expression of E6 and human UBE3A in the wing and eye of Drosophila disrupted these tissues . Similar to human cells we found that Drosophila Magi was a major E6 degradation target and that overexpression of Magi rescued the tissue disruption . However , Drosophila p53 was not degraded by E6/UBE3A , making our fly model potentially useful for studying the p53-independent activities of the E6+UBE3A complex . When we paired E6 expression with oncogenic proteins , including activated Ras , we observed that epithelia were transformed into mesechymal-like cells that left the epithelium and spread through the body . As a test of the potential of our system , we carried out a pilot genetic screen and identified the insulin receptor as a strong modulator of the E6-mediated disruption of Drosophila tissues . Therefore , we have developed a new system and approach to help us better understand the mechanisms that underlie how HPV infection leads to cell transformation and cancer .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"diabetic",
"endocrinology",
"cell",
"processes",
"animals",
"hormones",
"endocrine",
"physiology",
"animal",
"models",
"oncology",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"eyes",
"morphogenesis",
"drosophila",
"research",
"and",
"analysis",
"methods",
"insulin",
"endocrinology",
"head",
"imaginal",
"discs",
"insects",
"arthropoda",
"biochemistry",
"carcinogenesis",
"anatomy",
"cell",
"biology",
"phenotypes",
"apoptosis",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"ocular",
"system",
"insulin",
"signaling",
"organisms"
] |
2016
|
A Drosophila Model of HPV E6-Induced Malignancy Reveals Essential Roles for Magi and the Insulin Receptor
|
Endosymbiosis has driven major molecular and cellular innovations . Plasmodium spp . parasites that cause malaria contain an essential , non-photosynthetic plastid—the apicoplast—which originated from a secondary ( eukaryote–eukaryote ) endosymbiosis . To discover organellar pathways with evolutionary and biomedical significance , we performed a mutagenesis screen for essential genes required for apicoplast biogenesis in Plasmodium falciparum . Apicoplast ( − ) mutants were isolated using a chemical rescue that permits conditional disruption of the apicoplast and a new fluorescent reporter for organelle loss . Five candidate genes were validated ( out of 12 identified ) , including a triosephosphate isomerase ( TIM ) -barrel protein that likely derived from a core metabolic enzyme but evolved a new activity . Our results demonstrate , to our knowledge , the first forward genetic screen to assign essential cellular functions to unannotated P . falciparum genes . A putative TIM-barrel enzyme and other newly identified apicoplast biogenesis proteins open opportunities to discover new mechanisms of organelle biogenesis , molecular evolution underlying eukaryotic diversity , and drug targets against multiple parasitic diseases .
Plasmodium spp . , which cause malaria , and related apicomplexan parasites are important human and veterinary pathogens . These disease-causing protozoa are highly divergent from well-studied model organisms that are the textbook examples of eukaryotic biology , such that parasite biology often reveals striking eukaryotic innovations . The apicoplast , a nonphotosynthetic plastid found in apicomplexa , is one such “invention” [1 , 2] . These intracellular parasites evolved from photosynthetic algae that acquired their plastids through secondary endosymbiosis [3] . During secondary endosymbiosis , a chloroplast-containing alga , itself the product of primary endosymbiosis , was taken up by another eukaryote to form a secondary plastid [4] . Despite the loss of photosynthesis in apicomplexans , the apicoplast contains several prokaryotic metabolic pathways , is essential for parasite replication during human infection , and is a target of antiparasitic drugs [5–8] . There are many traces of the apicoplast’s quirky evolution in its present-day cell biology , in particular the pathways for its biogenesis . Like other endosymbiotic organelles , the single apicoplast cannot be formed de novo and must be inherited by its growth , division , and segregation into daughter cells . The few molecular details we have about apicoplast biogenesis hint at the major innovations that have occurred in the process of adopting and retaining this secondary plastid . The apicoplast is bound by four membranes acquired through successive endosymbioses , such that apicoplast proteins transit through the endoplasmic reticulum ( ER ) and use a symbiont-derived ER-associated degradation ( ERAD ) -like machinery ( SELMA ) to cross the two new outer membranes [9] . Curiously , autophagy-related protein 8 ( Atg8 ) , a highly conserved eukaryotic protein and key marker of autophagosomes , localizes to the apicoplast and is required for its inheritance in Plasmodium and the related apicomplexan parasite Toxoplasma gondii [10–12] . While SELMA and PfAtg8 are clear examples of molecular evolution in action , other novel or repurposed proteins required for apicoplast biogenesis remain undiscovered . So far , new apicoplast biogenesis proteins have primarily been discovered through candidate approaches . SELMA was first identified as homologs of the ERAD machinery encoded in the nucleomorph , the remnant nucleus of the eukaryotic symbiont found in some algal secondary plastids [13] . Nuclear-encoded versions of SELMA containing apicoplast-targeting sequences were then detected in the genomes of apicomplexan parasites ( which lack a nucleomorph ) [14] . Because the apicoplast proteome is enriched in proteins likely to perform biogenesis functions such as protein import or genome replication , several apicoplast-targeted proteins of unknown function have also been shown to be required for its biogenesis [15] . ATG8’s novel apicoplast function was discovered serendipitously by its unexpected localization on the apicoplast instead of autophagosomes [10] . Though candidate approaches have yielded new molecular insight [16–18] , in general they are indirect and may bias against novel pathways . In blood-stage Plasmodium falciparum , a method to chemically rescue parasites that have lost the apicoplast has paved the way for functional screens [19–21] . Addition of isopentenyl diphosphate ( IPP ) to the growth media is sufficient to reverse growth inhibition caused by apicoplast loss because it is the only essential metabolic product of the apicoplast in the blood stage . Taking advantage of the apicoplast chemical rescue , we recently took the first unbiased approach to discover a new apicoplast biogenesis protein [22] . We first screened for small-molecule inhibitors that specifically disrupt apicoplast biogenesis in P . falciparum . Subsequent target identification led us to a membrane metalloprotease , the P . falciparum ATP-dependent zinc metalloprotease 1 ( PfFtsH1 ) , with an unexpected role in apicoplast biogenesis . This chemical genetic screen has the advantage of unbiased sampling of druggable targets in apicoplast biogenesis pathways . Unfortunately , it lacks throughput given the painstaking process of mapping inhibitors to their molecular targets . Forward genetic screens are widely performed to uncover novel cellular pathways , such as those required for apicoplast biogenesis . Recently , genome-scale deletion screens performed in several apicomplexan parasites have uncovered a plethora of essential genes of unknown function [23–25] . Several challenges impede large-scale functional analysis of these essential genes . First , targeted gene modifications are still slow and labor intensive in P . falciparum , the most deadly of the human malaria species , despite an available in vitro blood culture system [26 , 27] . Second , efficient methods for generating conditional mutants , such as RNA interference ( RNAi ) or Clustered Regularly Interspaced Palindromic Repeats interference ( CRISPRi ) systems , are lacking in all apicomplexan organisms [28] . Finally , high-throughput , single-cell phenotyping for important functions need to be developed [29] . Overcoming these significant limitations , we designed a forward genetic screen using chemical mutagenesis , apicoplast chemical rescue , and a fluorescent reporter for apicoplast loss to identify essential apicoplast biogenesis genes in blood-stage P . falciparum . The screen identified known and novel genes required for apicoplast biogenesis and is , to our knowledge , the first forward genetic screen to assign essential cellular functions in P . falciparum .
The apicoplast must be propagated during parasite replication , such that biogenesis defects result in newly replicated daughter parasites that do not contain an apicoplast . Because no clearance mechanism is known to eliminate the apicoplast in asexual blood-stage parasites , defective organelle biogenesis resulting in loss of inheritance is likely the major cause of apicoplast loss . To isolate rare P . falciparum mutants that have lost their apicoplast due to biogenesis defects , we set out to design a live-cell reporter for apicoplast loss . The apicoplast contains a prokaryotic caseinolytic protease ( Clp ) system composed of a Clp chaperone ( ClpC ) that recognizes and unfolds substrates and a Clp protease ( ClpP ) that degrades the recognized substrates [30–32] . We hypothesized that ( 1 ) in the presence of a functional apicoplast , ClpCP could be co-opted to degrade and turn “off” a fluorescent reporter , whereas ( 2 ) loss of the apicoplast would result in loss of ClpCP activity and therefore turn “on” the reporter ( Fig 1A ) . Clp substrates are typically recognized by unstructured degron sequences , the best studied of which is a transfer-messenger RNA ( ssrA ) that appends a short peptide to the C-terminus of translationally stalled proteins [33 , 34] . However , the substrate specificity of the apicoplast ClpCP system has yet to be defined . Reasoning that the PfClpCP homolog might recognize similar degrons as bacterial or algal Clp systems , we tested two degrons recognized by Escherichia coli ClpXP—the E . coli ssrA peptide ( EcssrA ) and X7—and the predicted ssrA peptide in the red alga Cyanidium caldarium , CcssrA ( S1 Table ) [35–37] . To assess their functionality in P . falciparum , the C-terminus of an apicoplast-targeted green fluorescent protein ( acyl-carrier protein leader peptide [ACPL-GFP] ) was modified with each of the degrons ( Fig 1B ) . A cytosolic mCherry marker was also expressed on the same promoter as ACPL-GFP via a T2A “skip” peptide to normalize apicoplast GFP levels to stage-specific promoter expression [38] . Each construct was then integrated into an ectopic attB locus in Dd2attB parasites to generate the reporter strains [39] . For each reporter strain , the ratio of GFP:mCherry fluorescence ( as detected by flow cytometry ) was assessed in untreated parasites containing an intact apicoplast , designated apicoplast ( + ) , and in parasites treated with actinonin ( which causes apicoplast loss via inhibition of FtsH1 ) and rescued with IPP rendering them apicoplast ( − ) . In the absence of a degron , the GFP:mCherry ratio decreased modestly in apicoplast ( − ) parasites compared to apicoplast ( + ) ( Fig 1C , S1 Fig and S1 Data ) . Though GFP was present in both populations , live fluorescence microscopy confirmed its localization to a branched structure characteristic of the apicoplast in apicoplast ( + ) parasites and to dispersed punctae as previously reported in apicoplast ( − ) parasites ( S1 Fig ) [19] . Addition of each of the three degrons to the C-terminus of ACPL-GFP caused a 60%–84% decrease in the GFP:mCherry ratio in apicoplast ( + ) parasites , consistent with specific degradation of GFP ( Fig 1C ) . In ACPL-GFP-EcssrA and ACPL-GFP-CcssrA populations , the GFP:mCherry ratio recovered to 88% and 44% relative to the ACPL-GFP population in apicoplast ( − ) parasites , respectively ( Fig 1C and 1E , S1 Fig and S1 Data ) . Unexpectedly , ACPL-GFP-X7 populations showed further reduction of GFP:mCherry in apicoplast ( − ) parasites ( Fig 1C , S1 Fig and S1 Data ) . Notably , cytoplasmic mCherry levels were similar between apicoplast ( + ) and apicoplast ( − ) parasites in all reporter strains , suggesting that the differences in GFP levels were not due to altered expression levels ( S1 Fig and S1 Data ) . Of the 3 degrons tested , both the EcssrA and CcssrA peptide caused apicoplast-specific GFP degradation as designed . We further characterized ACPL-GFP-EcssrA because the greatest recovery of GFP fluorescence following apicoplast loss was observed with this reporter ( Fig 1C ) . Consistent with degradation of GFP in the apicoplast being dependent on EcssrA peptide , ACPL-GFP-EcssrA protein was detected at a significant level only in apicoplast ( − ) parasites , while unmodified ACPL-GFP was detected in both apicoplast ( + ) and ( − ) parasites ( Fig 1D ) . Of note , cleavage of the apicoplast-targeting ACPL sequence does not occur in apicoplast ( − ) parasites , resulting in an ACPL-GFP protein of higher molecular weight compared to apicoplast ( + ) parasites [19] . Flow cytometry and live fluorescence microscopy confirmed that apicoplast ( + ) parasites displayed only cytosolic mCherry fluorescence , whereas apicoplast ( − ) parasites displayed both cytosolic mCherry and dispersed , punctate GFP fluorescence ( Fig 1E and 1F ) . As expected , addition of IPP alone did not result in significant formation of apicoplast ( − ) parasites or recovery of ACPL-GFP-EcssrA ( S1 Fig and S1 Data ) . Taken together , these results demonstrate that ACPL-GFP-EcssrA serves as a specific reporter for apicoplast loss in P . falciparum . Next , we used the ACPL-GFP-EcssrA reporter strain to perform a phenotypic screen for apicoplast ( − ) mutants ( Fig 2A ) . Ring-stage parasites were mutagenized with the alkylating agents ethyl methanesulfonate ( EMS ) or N-ethyl-N-nitrosourea ( ENU ) with the expectation that this would generate a more diverse population of parasites , some of which harbor mutations in apicoplast biogenesis genes rendering them apicoplast ( − ) [21 , 22] . To rescue the lethality of apicoplast loss , mutagenized parasites were supplemented with 200 μM IPP in the growth media . A control group of non-mutagenized parasites was also cultured with IPP to assess whether apicoplast ( − ) mutants naturally occurred over time in the absence of selective pressure to maintain the organelle . After two replication cycles to allow for initial apicoplast loss , mCherry-expressing parasites displaying GFP fluorescence greater than about 70% percentile were isolated by fluorescence-activated cell sorting ( FACS ) . Selected mCherry ( + ) and GFP ( + ) parasites were then allowed to propagate to a detectable parasitemia before being subjected to another round of FACS . In one exemplary ENU-mutagenized population , a distinct population of GFP-positive parasites began to emerge after just two rounds of FACS ( Fig 2B ) . Other populations showed enrichment beginning after three rounds of FACS . No significant enrichment of GFP signal was observed when parasites were ( 1 ) grown with IPP over time without FACS or ( 2 ) grown without IPP and subjected to FACS , suggesting that we specifically enriched for apicoplast ( − ) mutants . Parasites from the final enriched apicoplast ( − ) populations were individually sorted to generate apicoplast ( − ) clones derived from single parasites . Each clonal population was then re-checked for IPP-dependent growth and GFP fluorescence . To determine the genetic basis of apicoplast loss , we sequenced the genomes of 51 apicoplast ( − ) clones ( mutagenized = 40; non-mutagenized = 11 ) and three parent populations . Sequenced reads were mapped to the genome of P . falciparum 3D7 ( version 35 ) with an average read depth of 26 for all samples . Notably , while the apicoplast genome was sequenced at an average depth of 16 reads in the parent populations , it was only detected with an average of 0 . 02 reads in apicoplast ( − ) clones ( Fig 2C and S2 Data ) . Because the organellar genome is a marker for the apicoplast , the absence of the apicoplast genome confirmed the loss of the apicoplast in the sequenced clones [19 , 22] . We next performed single nucleotide variant ( SNV ) analysis . A raw variant list was generated for each sample by comparison to the reference 3D7 genome and included SNVs found in parent populations at ≥5% allele frequency ( minimum one read ) , and in apicoplast ( − ) clones at ≥90% ( minimum five reads ) . Any SNV identified in apicoplast ( − ) clones that was also identified in any of the three parent populations was removed . We also filtered out SNVs detected in noncoding regions or resulting in synonymous amino acid changes in coding regions . Finally , SNVs identified in hypervariable regions of the genome ( including the rifin , stevor , and EMP gene families ) and/or previously annotated in the PlasmoDB single nucleotide polymorphism ( SNP ) database were excluded . After these filtering steps , 23 apicoplast ( − ) clones had at least one but no more than three SNVs that differed from the parent populations ( Fig 2D; S2 Table ) . Because genes required for apicoplast biogenesis ought to be essential , we used essentiality data from literature or whole-genome deletion screens performed in blood-stage P . falciparum and P . berghei to prioritize gene candidates [24 , 25] . Of 18 unique SNVs identified , 12 were in genes categorized as “essential” in blood-stage P . falciparum and/or P . berghei ( Table 1 and S2 Table ) . Although PfFtsH1 ( Pf3D7_1239700 ) is categorized as “dispensable” in the P . falciparum deletion screen , it has been shown experimentally to be essential [25] . Overall , a mutation in one of these 12 essential gene candidates was identified in each of the 23 apicoplast ( − ) clones , consistent with the mutation causing apicoplast loss . Potentially disruptive mutations included a I437S variant in the known apicoplast biogenesis protein ( PfFtsH1 ) , truncation of Atg7 ( PfAtg7 ) likely required for a cytoplasmic pathway for apicoplast biogenesis , and truncations of three proteins of unknown function ( Pf3D7_0518100 , Pf3D7_1305100 , and Pf3D7_1363700 ) ( Table 1 ) . The remaining candidates contained point mutations and had no known prior function in apicoplast biogenesis or localization to the organelle . PfFtsH1 is an apicoplast membrane metalloprotease that was previously identified in a chemical genetic screen as the target of actinonin , an inhibitor that disrupts apicoplast biogenesis . Subsequent knockdown of PfFtsH1demonstrated that it is required for apicoplast biogenesis [22] . We hypothesized that the I437S variant identified in our screen disrupted PfFtsH1 function , leading to apicoplast loss . PfFtsH1 contains both an ATPase and peptidase domain . To test the effect of I437S on enzyme activity , we compared the activity of I437S with that of wild-type ( WT ) enzyme , an ATPase-inactive E249Q variant , and a peptidase-inactive D493A variant ( Fig 3A ) . All enzymes were purified without the transmembrane domain as previously described ( S3 Fig ) [22] . We first measured peptidase activity on a fluorescent casein substrate . Neither the I437S nor the D493A variant had detectable peptidase activity , whereas WT enzyme turned over substrate at 0 . 4 min−1 ( Fig 3B , S3 Table and S3 Data ) . Similarly , the I437S and E249Q variants showed no detectable ATP hydrolysis activity , in contrast to both WT and the D493A variants ( Fig 3C , S3 Table and S3 Data ) . Taken together , these results validate the identified missense mutation leading to expression of inactive PfFtsH1 I437S variant as causative for apicoplast loss . Though non-apicoplast proteins are expected to play a role in apicoplast biogenesis , all apicoplast biogenesis proteins validated so far have localized to the apicoplast because this criterion is most often used to select candidates . A significant advantage of our forward genetic screen is that it can uncover non-apicoplast pathways required for apicoplast biogenesis , which are biased against by other approaches . A cytoplasmic protein strongly identified in our screen was PfAtg7 ( Pf3D7_1126100 ) , which contained a nonsense mutation causing a protein truncation at position Q719 . The premature stop codon was upstream of the E1-like activating domain , consistent with PfAtg7 loss of function ( Fig 4A ) . In yeast and mammalian cells , Atg7 is required for conjugation of Atg8 to the autophagosome membrane [41] . PfAtg7 has not specifically been implicated in apicoplast biogenesis; however , PfAtg8 has been shown to localize to the apicoplast membrane and be required for apicoplast biogenesis . In analogy with its role in autophagy , PfAtg7 is likely required for conjugating PfAtg8 to the apicoplast membrane . Therefore , we suspected the loss-of-function mutant we identified caused apicoplast loss via loss of membrane-conjugated PfAtg8 [12] . To confirm PfAtg7’s role in apicoplast biogenesis , we generated a P . falciparum strain in which it is conditionally expressed . The endogenous PfAtg7 locus in a NF54 strain harboring a CRISPR cassette was modified with a C-terminal triple hemagglutinin ( HA ) tag and a 3′ untranslated region ( UTR ) RNA aptamer sequence that binds a tetracycline repressor ( TetR ) and development of zygote inhibited ( DOZI ) fusion protein to generate PfAtg7-TetR/DOZI [42 , 43] . In the presence of anhydrotetracycline ( ATc ) , the 3′UTR aptamer is unbound , and PfAtg7-3×HA protein was detectable , albeit at low levels consistent with its low expression in published transcriptome data ( PlasmoDB . org ) ( S2 Fig ) . Removal of ATc causes binding of the TetR/DOZI repressor , which resulted in undetectable PfAtg7 protein levels by western blot within one replication cycle ( S2 Fig ) . Immunofluorescence of HA-tagged PfAtg7 showed diffuse cytoplasmic localization in the presence of ATc and overall reduction of fluorescence signal upon removal of ATc ( S4 Fig ) . Parasitemia also decreased over the course of several replication cycles , consistent with a previous study showing that PfAtg7 is essential ( Fig 4B and S4 Data ) [40] . We tested whether the requirement for PfAtg7 was due to its role in apicoplast biogenesis . Growth inhibition caused by PfAtg7 knockdown was partially rescued by addition of IPP ( Fig 4B and S4 Data ) . Furthermore , in −ATc/+IPP parasites , the apicoplast marker acyl-carrier protein ( ACP ) mislocalized from a discrete organellar localization to multiple cytoplasmic puncta , a hallmark of apicoplast loss ( Fig 4C ) [19] . Loss of transit peptide cleavage of the apicoplast protein ClpP also confirmed apicoplast loss , because mislocalized apicoplast proteins do not undergo removal of their targeting sequences ( Fig 4D ) [15 , 44] . IPP rescue of PfAtg7 knockdown parasites was incomplete , raising the possibility that PfAtg7 is also required for a non-apicoplast function . Alternatively , PfAtg7 knockdown may cause stalling of apicoplast morphologic development leading to general cellular toxicity that cannot be fully rescued with IPP until apicoplast loss is complete . To test these models , instead of PfAtg7 knockdown followed by apicoplast loss , we first induced apicoplast loss with actinonin and then assessed the effects of PfAtg7 knockdown . PfAtg7 knockdown in apicoplast ( − ) parasites did not cause any additional growth defect and was fully rescued by IPP , suggesting that the partial rescue observed in apicoplast ( + ) parasites was due to the order of disruption ( S5 Fig and S4 Data ) . These results confirmed that PfAtg7 is required for apicoplast biogenesis and likely is its only essential function . Finally , to determine whether PfAtg7’s role in apicoplast biogenesis was via PfAtg8 membrane conjugation , we transfected PfAtg7-TetR/DOZI with a transgene encoding GFP-PfAtg8 . In this strain , GFP-PfAtg8 primarily localized to a branched structure in schizonts , consistent with its previously described apicoplast localization ( Fig 4E ) [10 , 45] . Upon PfAtg7 knockdown , GFP-PfAtg8 localization to this discrete structure was lost , and accumulation in the cytoplasm was observed within a single replication cycle prior to apicoplast loss ( Fig 4E ) . This result suggests that , like yeast and mammalian Atg7 homologs , PfAtg7 has a conserved function in conjugating Atg8 to lipids . Altogether , PfAtg7 stood out as a cytoplasmic protein required for apicoplast biogenesis identified in our screen . The real power of forward genetics is the ability to uncover novel pathways without any a priori knowledge . Therefore , we next turned our attention to the nearly 50% of the Plasmodium genome annotated as “conserved Plasmodium protein , unknown function . ” Three candidate genes ( Pf3D7_0518100 , Pf3D7_1305100 , and Pf3D7_1363700 ) encoding proteins of unknown function were identified by nonsense mutations that caused protein truncation . The position of the premature stop codon near the 5′ end ( Pf3D7_0518100 , Pf3D7_1305100 ) or in the middle ( Pf3D7_1363700 ) of the genes suggested that these were loss-of-function mutations ( Fig 5A ) . Incidentally , all were also identified in a recently published proteomic dataset of apicoplast proteins , and immunofluorescence colocalization with the apicoplast marker ACP confirmed that Pf3D7_0518100 and Pf3D7_1305100 are localized to the apicoplast ( S4 Fig ) [15] . Therefore , we assessed whether knockdown of these genes disrupted apicoplast biogenesis . Similar to the experiments performed to validate PfAtg7 , ATc-regulated knockdown strains for each of the candidate genes were generated . Upon ATc removal , protein levels for each gene decreased within 24 hours as detected by western blot ( S2 Fig ) . Significant growth inhibition was also observed for all candidate genes tested , with varying kinetics of growth inhibition observed for each candidate ( Fig 5B , 5E and 5H and S4 Data ) . Of note , the gene essentiality demonstrated here for Pf3D7_1305100 and Pf3D7_1363700 confirmed whole-genome essentiality data reported in P . berghei and/or P . falciparum . However , the essentiality of Pf3D7_0518100 did not agree with its “dispensable” annotation in the P . falciparum dataset . IPP supplementation reversed growth inhibition for all the candidates , demonstrating that their essentiality was due to an apicoplast-specific function ( Fig 5B , 5E and 5H and S4 Data ) . Finally , mislocalization of ACP and loss of transit peptide cleavage of ClpP confirmed apicoplast loss for all candidates ( Fig 5C , 5D , 5F , 5G , 5I and 5J ) . Because these genes have so far lacked any functional annotation and given their shared knockdown phenotype , we designated them “apicoplast-minus , IPP-rescued” ( amr ) genes: amr1 ( Pf3D7_1363700 ) , amr2 ( Pf3D7_0518100 ) , and amr3 ( Pf3D7_1305100 ) . Taken together , we successfully identified three novel proteins of unknown function required for apicoplast biogenesis , prioritizing these amr genes for functional studies . To set up future functional studies , we noted that PfAMR1 contained a TIM-barrel domain with closest homology to indole-3-glycerol phosphate synthase ( IGPS ) , a highly conserved enzyme in the tryptophan ( trp ) biosynthesis pathway [46–48] . This was surprising because Plasmodium and the related apicomplexan parasite Toxoplasma are trp auxotrophs [49–51] . Indeed , analysis of >30 apicomplexan genomes did not detect any of the other six trp biosynthetic enzymes , except the terminal enzyme tryptophan synthase ( TS ) -β , which was horizontally transferred into Cryptosporidium spp . [52] . Therefore , we suspected that PfAMR1 may have a function unrelated to trp biosynthesis . To test the conservation of active-site residues , we aligned the sequences of several known IGPSs with IGPS homologs identified from P . falciparum , T . gondii , and Vitrella brassicaformis ( S6 Fig ) [53–55] . V . brassicaformis , a member of the Chromerids , is the closest free-living , photosynthetic relative to apicomplexan parasites . It contains a secondary plastid with the same origin as the apicoplast but , as a free-living alga [56] , is also expected to have intact trp biosynthesis . Known catalytic and substrate binding residues based on enzyme structure-function studies performed in bacteria were first identified [57] . For known IGPSs and one of the V . brassicaformis IGPS homologs ( Vbra_4894 ) , the key catalytic and substrate binding residues were all conserved , despite the vast evolutionary distance between bacteria , metazoans , and Chromerids/apicomplexans ( Fig 6A ) . However , in PfAMR1 and the other two V . brassicaformis homologs , key functional residues were not conserved . Based on the conservation of functional residues , we separated these sequences into two groups: “canonical IGPS proteins” ( which have been shown , or are likely , to encode for IGPS activity ) and “IGPS-like proteins” ( e . g . , PfAMR1 ) , which we suggest have functionally diverged . We next looked at the pattern of canonical IGPS versus IGPS-like proteins through two biological transitions: loss of trp biosynthesis ( Vitrella versus Plasmodium spp . ) and loss of the apicoplast ( Plasmodium versus Cryptosporidium spp . ) ( Fig 6B ) . As expected for a role in trp biosynthesis , canonical IGPS proteins were retained until the emergence of parasitism . In addition , genes encoding the remaining set of enzymes for trp biosynthesis were identified in the V . brassicaformis genome [58] . Unlike canonical IGPSs , however , IGPS-like proteins were retained in parasites that have lost trp biosynthesis . Instead , loss of IGPS-like proteins is associated only with loss of the apicoplast in Cryptosporidium spp . This pattern of acquisition and loss of IGPS-like proteins suggests an apicoplast-specific function separate from trp biosynthesis . Finally , we performed functional complementation to test the biochemical activity of canonical IGPS and IGPS-like genes from V . brassicaformis and P . falciparum . An E . coli strain ( trpC9800 ) containing an inactivating mutation in trpC , the E . coli IGPS homolog , was grown on minimal agar ( M9 ) [59] . As expected , trpC9800 was dependent on trp supplementation for growth ( Fig 6C ) . Complementation with the Vbra_4894 homolog restored trpC9800 growth on M9 , comparable to that of an E . coli strain with intact trp biosynthesis , suggesting that Vbra_4894 is an IGPS protein ( Fig 6D ) . In contrast , neither PfAMR1 nor any of the IGPS-like genes were able to functionally replace trpC , supporting an alternative biochemical function . Because glutathione S-transferase ( GST ) -tagged complemented protein could not be detected in any strain , we cannot rule out that the lack of functional complementation was due to differential protein expression or solubility below the detection limit of western blot . However , isopropyl β-D-1-thiogalactopyranoside ( IPTG ) induction of protein expression was toxic for all complemented strains , suggesting all constructs supported protein expression . Overall , we propose that AMR1 has evolved a new biochemical function required for apicoplast biogenesis .
Apicoplast biogenesis provides a fascinating window into molecular evolution , including examples of proteins that have been reused ( e . g . , translocon on the inner chloroplast membrane/translocon on the outer chloroplast membrane [TIC/TOC] complexes ) [17 , 60 , 61] , repurposed ( e . g . , Atg8 , symbiont-derived ERAD-like machinery [SELMA] ) [12 , 14] , or newly invented in the process of serial endosymbioses . Overcoming significant technical challenges in the Plasmodium experimental system , we designed a forward genetic screen to identify essential apicoplast biogenesis pathways . This singular screen opens up opportunities to discover evolutionary innovations obscured by candidate-based approaches , including cytoplasmic pathways and genes lacking any functional annotations . In addition to confirming the role of PfFtsH1 in apicoplast biogenesis and identification of PfAtg8 conjugation machinery , we identified several proteins of unknown function required for apicoplast biogenesis that have so far gone undetected . Because our reporter specifically looked for apicoplast loss , we cannot rule out the possibility that some identified genes may be involved in maintenance of already formed apicoplasts . However , because no clearance mechanism is known for defective apicoplasts , we are not aware of any pathway by which defective apicoplasts would lead to organelle loss independent of parasite replication . One surprising gene we identified was PfAMR1 , which encodes a TIM-barrel domain found in diverse enzymes catalyzing small-molecule reactions [46] . PfAMR1 may have evolved from gene duplication of IGPS , an enzyme in the trp biosynthetic pathway [47] . However , the evolutionary pattern of retention in apicomplexan parasites lacking tryptophan biosynthesis and loss in Cryptosporidium spp . , concomitant with plastid loss , supports a critical function of PfAMR1 in the apicoplast independent of tryptophan biosynthesis . We hypothesize that PfAMR1 may be involved in biosynthesis of a specialized lipid or signaling molecule required specifically for building new plastids in this lineage . Multiple new amr genes identified in this study provide striking examples of the unexpected findings enabled by unbiased screens . Uncovering novel apicoplast biogenesis pathways also has important biomedical applications . While targeting the metabolic function of the apicoplast has been a major strategy for antimalarial drug discovery [62] , it has become apparent that apicoplast biogenesis is equally as , or likely more , valuable as a therapeutic target [22] . These distinct pathogen pathways are nonetheless required in every proliferative stage of the Plasmodium life cycle and conserved among apicomplexan parasites . Targeting apicoplast biogenesis has the benefit of efficacy against multiple Plasmodium life stages and multiple pathogens . Consistent with this broad utility , antibiotics that inhibit translation in the apicoplast and disrupt its biogenesis are used clinically for malaria prophylaxis , acute malaria treatment , and treatment of babesiosis and toxoplasmosis [7 , 8 , 63] . Until now , a forward genetic screen for essential pathways in blood-stage Plasmodium has not been achieved . Previous screens in murine P . berghei and the human malaria parasite P . falciparum identified nonessential genes required for gametocyte formation [64 , 65] , the developmental stage required for mosquito transmission . Functional screens for essential pathways have been impeded by several technical challenges , including the low transfection efficiency of P . falciparum , in vivo growth requirement of P . berghei , and absence of efficient methods for generating conditional mutants in both organisms [28] . Nonetheless , genome-scale deletion screens in P . berghei and P . falciparum using a homologous recombination-targeted deletion library or saturating transposon-based mutagenesis , respectively , have revealed a plethora of essential genes [24 , 25] . Functional assignment of these essential genes is a priority . In this context , the apicoplast biogenesis screen presented here is a major milestone towards unbiased functional identification of novel , essential genes . A top priority for “version 2 . 0” is to expand the screen to genome scale , maximizing our ability to uncover novel pathways . Apicoplast biogenesis is a complex process encompassing a multitude of functions and is expected to require hundreds of gene products . The identification of 12 candidate genes and our sparse sampling of known genes suggest that the current screen is far from saturating . The most significant bottleneck is the dependence of this screen on chemical mutagenesis . Mapping mutations by whole-genome sequencing limits the number of mutants that can be analyzed . Even for sequenced clones , less than half had a detectable point mutation in a coding region . The remaining apicoplast ( − ) clones may have contained mutations in noncoding regions or other types of mutations that are more difficult to detect ( insertion , deletions , or copy number variations ) . Particularly for apicoplast ( − ) clones selected from non-mutagenized conditions , we considered the possibility that some parasites might spontaneously fail to inherit the apicoplast due to mechanical defects during parasite replication; these daughter cells resulting from erroneous apicoplast division and segregation usually would not survive but are rescued by IPP . Finally , specific mutations identified in candidate genes also need to be validated one by one . In this study , four nonsense mutations were validated by conditional knockdown , while a missense mutation in PfFtsH1 was validated using an available activity assay . Although we were able to demonstrate loss of function as result of the PfFtsH1 I437S variant , other missense mutations identified in genes of unknown function will be challenging to follow up with available genetic tools . Given these limitations , an alternative mutagenesis method will increase the screen throughput . Options include adaptation of the piggyBac transposon developed for the P . falciparum deletion screen [25] or development of large-scale targeted mutagenesis . Switching to more genetically tractable apicomplexan organisms , such as P . berghei or Toxoplasma , would provide ready options for large-scale targeted gene disruptions [23 , 24] . However , these would have to be performed under conditional regulation because chemical rescue of the apicoplast is not feasible in these organisms . We anticipate that continued advances in genetic methods in apicomplexan organisms will open up opportunities to expand this screen in the future .
Human erythrocytes were purchased from the Stanford Blood Center ( Stanford , California ) to support in vitro P . falciparum cultures . Because erythrocytes were collected from anonymized donors with no access to identifying information , IRB approval was not required . All consent to participate in research was collected by the Stanford Blood Center . P . falciparum Dd2attB ( MRA-843 ) were obtained from MR4 . NF54Cas9+T7 Polymerase parasites were kindly provided by Jacquin Niles . Parasites were grown in human erythrocytes at 2% hematocrit ( Stanford Blood Center ) in RPMI 1640 media ( Gibco ) and supplemented with 0 . 25% Albumax II ( Gibco ) , 2 g/L sodium bicarbonate , 0 . 1 mM hypoxanthine ( Sigma ) , 25 mM HEPES ( pH 7 . 4; Sigma ) , and 50 μg/L gentamicin ( Gold Biotechnology ) at 37°C , 5% O2 , and 5% CO2 . For transfections , 50 μg of plasmid DNA was added to 200 μL packed red blood cells ( RBCs ) , adjusted to 50% hematocrit in RPMI 1640 , and electroporated as previously described [66] . On day 4 post transfection , parasites were selected for with 2 . 5 . mg/L blasticidin S ( RPI Research Products International ) . TetR/DOZI strains were cultured with 500 nM ATc for the entire duration of transfection . For TetR/DOZI strains expressing ACPL-GFP or GFP-PfAtg8 , parasites were additionally selected for with 500 μg/mL G418 sulfate ( Corning ) during transfection . Oligonucleotides were purchased from the Stanford Protein and Nucleic Acid facility or IDT . gBlocks were ordered from IDT . Molecular cloning was performed using In-Fusion Cloning ( Clontech ) or Gibson Assembly ( NEB ) . Primer and gBlock sequences for all cloning are available in S3 Table . To generate the plasmid pRL2-mCherry-T2A-ACPL-GFP , T2A-ACPL-GFP was first amplified from the pRL2-ACPL-GFP vector . mCherry was amplified from pTKO2-mCherry vector ( kind gift from J . Boothroyd ) and inserted in front of T2A-ACPL-GFP in the pRL2 backbone using the In-Fusion Cloning kit ( Takara ) . To generate the pL2-mCherry-T2A-ACPL-GFP-degron plasmids , T2A-ACPL-GFP-degron was amplified from pRL2-mCherry-T2A-ACPL-GFP . For CRISPR-Cas9–based editing of endogenous Pf3D7_0518100 , Pf3D7_1126100 , Pf3D7_1305100 , and Pf3D7_1363700 loci , sgRNAs were designed using the eukaryotic CRISPR guide RNA/DNA design tool ( http://chopchop . cbu . uib . no/ ) . To generate a linear plasmid for CRISPR-Cas9–based editing , left and right homology regions were first amplified for each gene . For each gene , a gBlock containing the recoded coding sequence C-terminal of the CRISPR cut site and a triple HA tag was synthesized with appropriate overhangs for Gibson Assembly . This fragment along with the left homology region was simultaneously cloned into the FseI/ApaI sites of the linear plasmid pSN054-V5 . Next , the appropriate right homology region and a gBlock containing the sgRNA expression cassette were simultaneously cloned into the AscI/I-SceI sites of the resultant vector to generate the plasmids . To generate plasmid for expression of GFP-PfAtg8 , GFP with a GlyAlaGlyAla linker was amplified from pRL2-ACPL-GFP . PfAtg8 was amplified from P . falciparum gDNA . Both fragments were inserted into pfYC110 vector [38] using the In-Fusion Cloning kit . V . brassicaformis RNA from strain CCMP3346 was purchased from the National Center for Marine Algae and Microbiota and was subsequently used to generate cDNA using Superscript III cDNA Kit ( Life Technologies ) . For Plasmodium PF3D7_1363700 cloning , codon optimized gBlocks were used to construct the Plasmodium construct . Constructs were cloned into the pGEXT vector using the In-Fusion Cloning kit . Ring-stage mCherry-T2A-ACPL-GFP-degron parasites were treated with 10 μM actinonin ( Sigma ) and 200 μM IPP ( Isoprenoids , LLC ) to disrupt the apicoplast . Both treated and nontreated parasites were analyzed two cycles post treatment at the schizont stage on a BD Accuri C6 flow cytometer . For each condition , 100 , 000 to 500 , 000 events were recorded . Uninfected RBCs were first removed from the population by setting a gate for mCherry fluorescence . For each strain , the average GFP and mCherry fluorescence intensities were then calculated for the infected cell population using FlowJo . For example , if 10 , 000 infected cells were counted , then the GFP and mCherry fluorescence for each cell was measured by the flow cytometer , and the average fluorescence was determined for the whole population . To calculate the GFP:mCherry ratios for comparative analysis of degron efficiency , the GFP to mCherry fluorescence ratio for each individual infected cell was first obtained . The fluorescence ratios of all infected cells were then averaged to determine the overall population GFP:mCherry ratio . Ring-stage mCherry-T2A-ACPL-GFP-EcssrA ( EcssrA ) parasites were seeded onto a 96-well plate at a volume of 200 μL , 2% hematocrit , and 0 . 5%–1% parasitemia . Parasites were either untreated or treated with 1 mM EMS or 100 μM ENU for 2 hours , and then washed three times afterwards to remove the mutagen from the growth media . Parasites were cultured in growth media + 200 μM IPP for the duration of the screen . At 120 hours post treatment , mutants were isolated on a Sony SH800S Cell Sorter . Uninfected RBCs were first analyzed to set the gate for overall fluorescence . Untreated EcssrA parasites were analyzed to gate for positive and negative mCherry and GFP expression , respectively . Actinonin-treated EcssrA parasites were analyzed to gate for positive GFP expression . Non-mutagenized and mutagenized parasites displaying both positive mCherry and GFP expression were FACS’d into a new 96-well plate . Enriched parasites were allowed to propagate to a detectable parasitemia before being subjected to subsequent FACS rounds . Mutants were enriched until mCherry and GFP fluorescence approached actinonin-treated levels . In the final enrichment , up to 52 mutants were single-cell cloned . Mutants that survived single-cell sorting were split into growth media containing either 200 μM IPP or no IPP . Mutants displaying growth only in IPP were expanded to a 10 mL culture at approximately 10% parasitemia and ring-stage synchronized; 5 mL of culture was harvested for DNA extraction , and 5 mL culture was frozen at −80°C . Ring-stage parasites were isolated from RBC in 0 . 1% saponin and washed three times in PBS . gDNA from mutants and the parental strain was isolated using the Quick-DNA Universal Kit ( Zymo Research ) . Paired-end gDNA libraries were generated and barcoded for each mutant and the parental using the Nextera Library Prep Kit , modified for 8 PCR cycles ( Stanford Functional Genomics Facility ) . Up to 26 pooled libraries were analyzed per lane of an Illumina HiSeq 4000 using 2 × 75 bp , paired-end sequencing ( Stanford Functional Genomics Facility ) . Fastq files were checked for overall quality using FastQC . Ten and 15 bp were trimmed from the 5′ and 3′ ends of all 75 bp sequence reads , respectively , to remove low-quality reads; 20 and 30 bp were trimmed 5′ and 3′ ends of 150 bp sequence reads , respectively , from an additional parental Dd2 strain sequenced by the DeRisi lab ( https://www . ncbi . nlm . nih . gov/sra/SRX326518 ) . The resulting paired-end sequencing reads were mapped using Bowtie2 against the P . falciparum 3D7 ( version 35 ) reference genome . One mismatch per read was allowed , and only unique reads were aligned ( reads were removed if they aligned to more than one region of the genome ) . PCR duplicates were removed using Samtools rmdup , and raw SNVs were called using Samtools mpileup . Indels were not analyzed . Bcftools was used to generate the raw variant list for parental ( allele frequency ≥ 0 . 05 , depth ≥ 1 ) and mutant ( allele frequency ≥ 0 . 9 , depth ≥ 5 ) strains . Variants found in the parental strain were excluded from the mutant variant list . Variants were filtered to only include protein-coding mutations ( missense and nonsense ) . Mutations found in hypervariable gene families were excluded . Remaining variants that were previously annotated in PlasmoDB were excluded to generate the final variant list . The reported variants were confirmed to meet the filtering requirements using Samtools tview . Mutants containing nonsense mutations were Sanger sequenced to confirm the presence of the mutations prior to genetic validation . The custom script and parameters used for analysis are available at https://github . com/yehlabstanford/biogenesis_screen . Parasites were separated from RBCs by lysis in 0 . 1% saponin and were washed in PBS . Parasite pellets were resuspended in PBS containing 2× NuPAGE LDS sample buffer and boiled at 95°C for 10 minutes before separation on NuPAGE gels . Gels were transferred onto nitrocellulose membranes using the Trans-Blot Turbo Transfer System ( Bio-Rad ) . Membranes were blocked in 0 . 1% Hammarsten casein ( Affymetrix ) in 0 . 2× PBS with 0 . 01% sodium azide . Antibody incubations were performed in a 1:1 mixture of blocking buffer and Tris-buffered saline with Tween ( TBST ) -20 ( 10 mM Tris [pH 8 . 0] , 150 mM NaCl , 0 . 25 mM EDTA , 0 . 05% Tween 20 ) . Blots were incubated with primary antibody overnight at 4°C at the following dilutions: 1:20 , 000 mouse-α-GFP JL-8 ( Clontech 632381 ) , 1:20 , 000 rabbit-α-Plasmodium aldolase ( Abcam ab207494 ) , 1:1 , 000 mouse-α-HA 2–2 . 2 . 14 ( Thermo Fisher 26183 ) , 1:1 , 000 guinea pig-α-ATG8 ( Josman LLC ) , and 1:1 , 000 rabbit-α-ClpP ( kind gift from W . Houry ) . Blots were washed three times in TBST and were incubated for 1 hour at room temperature in a 1:10 , 000 dilution of the appropriate fluorescent secondary antibodies ( LI-COR Biosciences ) . Blots were washed twice in TBST and once in PBS before imaging on a LI-COR Odyssey imager . Parasites expressing GFP-PfAtg8 were grown in the presence or absence of ATc for 24 hours; 10 ml cultures were lysed with 0 . 1% saponin and washed 3 times with PBS . Parasite pellets were resuspended in ice-cold lysis buffer ( 1× PBS , 1% Triton X-114 [Thermo Scientific 28332] , 2 mM EDTA , 1× protease inhibitors [Pierce A32955] ) and incubated on ice for 30 minutes . Cell debris were removed by 10-minute centrifugation at 16 , 000 × g , 4°C . Supernatant was transferred to a fresh Eppendorf tube , incubated 2 minutes at 37°C to allow phase separation , and centrifuged 5 minutes at 16 , 000 × g at room temperature . The top ( aqueous ) layer was transferred to another tube . The interphase was removed to avoid cross-contamination between the layers . The bottom ( detergent ) layer was resuspended in 1× PBS , 0 . 2 mM EDTA to equalize the volumes of the two fractions . Both fractions were subjected to methanol-chloroform precipitation , resuspended in PBS containing 2× NuPAGE LDS sample buffer , boiled for 5 minutes at 95°C , and analyzed by western blot as described above . For live imaging , parasites were settled onto glass-bottom microwell dishes Lab-Tek II chambered coverglass ( Thermo Fisher 155409 ) in PBS containing 0 . 4% glucose and 2 μg/mL Hoechst 33342 stain ( Thermo Fisher H3570 ) . Cells were imaged with a 100× , 1 . 4 NA objective on an Olympus IX70 microscope with a DeltaVision system ( Applied Precision ) controlled with SoftWorx version 4 . 1 . 0 and equipped with a CoolSnap-HQ CCD camera ( Photometrics ) . Images were captured as a series of z-stacks separated by 0 . 2 μm intervals , deconvolved ( except for mCherry images ) , and displayed as maximum intensity projections . Brightness and contrast were adjusted equally in SoftWorx or Fiji ( ImageJ ) for display purposes . For immunofluorescence , fixed-cell imaging , parasites were first fixed with 4% paraformaldehyde ( Electron Microscopy Science 15710 ) and 0 . 0075% glutaraldehyde in PBS ( Electron Microscopy Sciences 16019 ) for 20 minutes . Cells were washed once in PBS and allowed to settle onto poly-L-lysine-coated coverslips ( Corning ) for 60 minutes . Coverslips were then washed once with PBS , permeabilized in 0 . 1% Triton X-100/PBS for 10 minutes , and washed twice more in PBS . Cells were treated with 0 . 1 mg/mL NaBH4/PBS for 10 minutes , washed once in PBS , and blocked in 5% BSA/PBS . Primary antibodies were diluted in 5% BSA/PBS at the following concentrations: 1:500 rabbit-α-PfACP ( kind gift from S . Prigge ) and 1:100 rat-α-HA 3F10 ( Sigma 11867423001 ) . Coverslips were washed three times in PBS , incubated with secondary antibodies goat-α-rat 488 ( Thermo Fisher A-11006 ) and donkey-α-rabbit ( Thermo Fisher A10042 ) at 1:3 , 000 dilution , and washed three times in PBS prior to mounting in ProLong Gold antifade reagent with DAPI ( Thermo Fisher ) . Ring-stage TetR/DOZI strain parasites were washed two times in growth media to remove ATc . Parasites were divided into three cultures supplemented with 500 nM ATc , no ATc , or no ATc + 200 μM IPP . Samples were collected at the schizont stage in each growth cycle for flow cytometry analysis and western blot . Parasites in each condition were diluted equally every growth cycle for up to six growth cycles . For parasitemia measurements , parasite-infected or uninfected RBCs were incubated with the live-cell DNA stain dihydroethidium ( Thermo Fisher D23107 ) for 30 minutes at a dilution of 1:300 ( 5 mM stock solution ) . Parasites were analyzed on a BD Accuri C6 flow cytometer , and up to 100 , 000 events were recorded . The parent His6-SUMO-PfFtsH191-612-GST plasmid as well as E249Q and D493A mutants were obtained from laboratory stocks . An I437S mutant was constructed by site-directed mutagenesis . Recombinant proteins were expressed from these plasmids and purified as described [22] . Rates of ATP hydrolysis by PfFtsH1 were measured using a coupled spectrophotometric assay [67] in protein degradation ( PD ) buffer ( 25 mM HEPES [pH 7 . 5] , 200 mM NaCl , 5 mM MgSO4 , 10 μM ZnSO4 , 10% glycerol ) with 3% dimethyl sulfoxide ( DMSO ) or 50 μM actinonin in 3% DMSO at 37°C . PD rates were measured by incubating PfFtsH1 ( 1 μM ) with FITC-labeled ( 2 μM , Sigma C0528 ) and unlabeled casein ( 8 μM ) in PD buffer plus 3% DMSO or 50 μM actinonin in 3% DMSO . Reactions were started by adding ATP ( 4 mM ) or buffer with a regeneration system ( 16 mM creatine phosphate and 75 μg/mL creatine kinase ) , and degradation was followed by measuring the fluorescence intensity ( excitation 485 nm; emission 528 nm ) at 37°C . IGPS and IGPS-like proteins from V . brassicaformis were identified by BLAST through using CryptoDB . First , secondary structure was predicted using PSI-PRED in the XtalPred suite [53 , 55] . Only the sequences containing the TIM barrels of each sequence were used for alignment because there are large N- and C-terminal extensions in the noncanonical proteins . PROMALS3D was subsequently used to perform a multiple sequence alignment based on secondary structure and homology to proteins with determined 3D structures [54] . W3110trpC9800 E . coli strain was purchased from the Yale University Coli Genetic Stock Center and were made chemically competent using calcium chloride . BL21 Star ( DE3 ) competent cells ( Thermo Fisher ) were used for the WT condition . The competent W3110trpC9800 cells were transformed with the pGEXT vectors containing the different Vitrella or Plasmodium IGPS and IGPS-like genes and were plated on LB agar plates containing carbenicillin . For each construct , a colony was picked and washed in M9 minimal media ( M9 ) ( 22 mM potassium phosphate monobasic , 22 mM sodium phosphate dibasic , 85 mM sodium chloride , 18 . 7 mM ammonium chloride , 2 mM magnesium sulfate , 0 . 1 mM calcium chloride , and 0 . 4% glycerol ) , resuspended in M9 , and streaked onto M9/agar plates containing either carbenicillin ( 100 μg/mL ) or carbenicillin and 1 mM L-tryptophan ( Sigma ) . Plates were incubated at 37°C and allowed to grow for two days , after which images of the plates were taken .
|
Plasmodium parasites , which cause malaria , and related apicomplexan parasites evolved from photosynthetic algae that acquired their chloroplast through two successive endosymbioses . Although no longer photosynthetic , the apicomplexan plastid—or apicoplast—was retained in these pathogens and provides critical metabolites during host cell infection . The apicoplast is of major interest for its unique biology and potential to yield new antimalarial drug targets . Here , we focused on the critical genes required to grow , divide , and inherit new apicoplasts during parasite replication . Given the apicoplast’s divergent evolution , most of these cannot be recognized by their homology to genes with known functions . Instead , we overcame significant technical challenges in the Plasmodium experimental system to perform an unbiased screen to search for these critical genes . Our screen has uncovered new genes with intriguing evolution and function that open up opportunities to understand and ultimately exploit apicoplast biology . Finally , assigning new , essential gene functions in Plasmodium parasites remains a daunting task . The successful identification of essential gene functions using an unbiased approach in this study provides a viable route for expansion of this screen or developing screens for other novel Plasmodium pathways in the future .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"parasite",
"groups",
"parasite",
"replication",
"plasmodium",
"green",
"fluorescent",
"protein",
"cloning",
"parasitic",
"protozoans",
"parasitology",
"apicomplexa",
"luminescent",
"proteins",
"protozoans",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"malarial",
"parasites",
"proteins",
"molecular",
"biology",
"biochemistry",
"methods",
"&",
"resources",
"eukaryota",
"genetic",
"screens",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"organisms"
] |
2019
|
A mutagenesis screen for essential plastid biogenesis genes in human malaria parasites
|
Optimal Bayesian models have been highly successful in describing human performance on perceptual decision-making tasks , such as cue combination and visual search . However , recent studies have argued that these models are often overly flexible and therefore lack explanatory power . Moreover , there are indications that neural computation is inherently imprecise , which makes it implausible that humans would perform optimally on any non-trivial task . Here , we reconsider human performance on a visual-search task by using an approach that constrains model flexibility and tests for computational imperfections . Subjects performed a target detection task in which targets and distractors were tilted ellipses with orientations drawn from Gaussian distributions with different means . We varied the amount of overlap between these distributions to create multiple levels of external uncertainty . We also varied the level of sensory noise , by testing subjects under both short and unlimited display times . On average , empirical performance—measured as d’—fell 18 . 1% short of optimal performance . We found no evidence that the magnitude of this suboptimality was affected by the level of internal or external uncertainty . The data were well accounted for by a Bayesian model with imperfections in its computations . This “imperfect Bayesian” model convincingly outperformed the “flawless Bayesian” model as well as all ten heuristic models that we tested . These results suggest that perception is founded on Bayesian principles , but with suboptimalities in the implementation of these principles . The view of perception as imperfect Bayesian inference can provide a middle ground between traditional Bayesian and anti-Bayesian views .
An important function of the visual system is to make inferences about the environment from noisy sensory input . It is often claimed that human performance on perceptual inference tasks is optimal or “Bayesian” [1–5] , meaning that subjects supposedly perform as well as theoretically possible given the amount of sensory noise in their observations . Evidence for this claim has mainly come from tasks in which subjects integrate two sensory cues to estimate a common source . The optimal strategy in these tasks is to compute a weighted average of the two cues , where each weight depends on the cue’s reliability: the more reliable the cue , the more strongly it weighs in on the decision [6] . Reliability-based weighting is a hallmark of Bayesian observers and predicts that a subject’s estimates are biased towards the more reliable cue . This prediction has been confirmed in a wide range of experiments in which two sensory cues need to be combined to estimate a common source . Examples include integration of a visual and haptic cue to estimate the height of an object [7] , a visual and proprioceptive [8] or auditory [9] cue to estimate object location , and two visual cues to estimate object depth [10 , 11] or object slant [12] . More recent work has reported that optimality in perception extends to tasks with as many as eight cues and with highly non-linear optimal decision rules , including visual search [13–17] , categorization [18] , change detection [19] , change localization [20] , and sameness discrimination [21] tasks . While these studies have provided valuable insights into basic mechanisms of perception , they have also been criticized . One criticism is that the emphasis on optimality has led to an underreporting and underemphasizing of studies that have found violations of optimality [22 , 23] . Another , more fundamental criticism is that optimal models often lack explanatory power due to being overly flexible [23–26] . The risk of too much flexibility is that it may allow an optimal model to account for data from suboptimal observers . For example , when sensory noise levels are fitted as free parameters—as in most studies—an optimal model may account for suboptimalities in inference by overestimating these noise levels . Similarly , a freely fitted lapse rate may help an optimal model to explain away errors that were in reality caused by poor decision making . In addition to this methodological concern , several recent studies have suggested that neural computation is inherently imprecise [27–31] , which makes it a priori implausible that humans perform optimally on any non-trivial task . Here , we revisit optimality in perception by using a method that takes note of the concerns described above in three ways . First , we constrain flexibility of the optimal model by imposing prior distributions on its parameters; this reduces the risk that the optimal model explains away decision suboptimalities as sensory noise or attentional lapses . Second , for each model that we test , we also include a variant with computational imperfections . Such imperfections may produce suboptimal behavior , even when subjects use a decision strategy that is based on Bayesian principles . By including these models in our analyses , we can distinguish performance loss caused by using a wrong decision rule from performance loss due to imperfect execution of a rule . Third , besides only testing which kind of model accounts best for behavior , we will also quantify performance loss and partition this loss into different sources ( see [27] for a similar approach ) . We choose visual search as our experimental task . Despite the complexity of the optimal decision rule for this task , several previous studies have reported that humans perform near-optimally [13–15] . We include experimental conditions in which stimuli are corrupted by external noise , which makes the task more consistent with naturalistic conditions , where inference often involves dealing with both internal and external sources of uncertainty [32 , 33] . We fit several Bayesian model variants as well as ten heuristic models to the experimental data . To preview our main result , we find no evidence for perfect optimality , nor for any of the heuristic-based strategies . Instead , the data are best explained by an “imperfect Bayesian” , in which decisions are based on Bayesian principles , but subject to imperfections in the implementation of these principles .
The study was approved by the Regional Ethical Review Board in Uppsala and conducted according to the Declaration of Helsinki Principles . Study subjects gave written informed consent prior to their enrollment in the experiment . The experimental data and Matlab code to reproduce the main figures and to fit the models are available at https://osf . io/dkavj/ . Thirty subjects were recruited via advertisements at the psychology department of Uppsala University in Sweden and received payment in the form of cinema tickets or gift vouchers . All subjects had self-reported normal or corrected-to-normal vision and gave informed consent before the start of the experiment . No subjects were excluded from any of the analyses . Stimuli were black ellipses ( 0 . 35 cd/m2 ) with an area of 0 . 60 deg2 presented on a gray background ( 71 cd/m2; Fig 1A ) . The task-relevant feature in all experiments was ellipse orientation , with 0° defined as vertical . The eccentricity of the ellipses differed across stimuli and conditions . Ellipse eccentricity is formally defined as 1−b2a2 , where a and b specify the ellipse’s semi-major axis and semi-minor axis , respectively . To avoid confusion with visual field eccentricity , we will refer to this eccentricity as “elongation” . Differences in elongation were used to create differences in the level of sensory noise across stimuli ( Fig 1B ) . Stimuli were generated using the Psychophysics Toolbox [34] for Matlab and presented at fixed locations along an invisible circle at the center of the screen and with a radius of 7 degrees of visual angle . Each subject completed multiple experimental sessions that lasted about one hour each . At the start of the first session , they received general information about the experiment . Thereafter , they performed a discrimination task ( Fig 1A ) followed by one condition of the visual search task . In the remaining sessions , they only performed the visual search task ( Fig 1C ) . We created eight conditions for the visual search task by using a 2×4 factorial design ( Table 1 ) . The factors specify the stimulus presentation time ( short vs . unlimited ) and the level of external uncertainty ( none , 5% , 10% , and 15%; explained below ) . Different groups of subjects performed different subsets of these conditions . On each trial , the subject was presented with a single ellipse ( 67 ms ) and reported whether it was tilted clockwise or counterclockwise with respect to vertical ( Fig 1A ) . Trial-to-trial feedback was provided by briefly turning the fixation cross in the inter-trial screen green ( correct ) or red ( incorrect ) . The elongation of the stimulus was 0 . 80 on half of the trials ( “low reliability” ) and 0 . 94 on the other half ( “high reliability” ) , randomly intermixed . The stimulus location was randomly drawn on each trial from the set of four cardinal locations ( “north” , “east” , “south” , and “west” ) . On the first 20 trials , the orientation of the stimulus was drawn from a uniform distribution on the range −5° to +5° . In the remaining trials , a cumulative Gaussian was fitted to the data collected thus far and the orientation for the next trial was then randomly drawn from the domain corresponding to the 55–95% correct range . This adaptive procedure increased the information obtained from each trial by reducing the number of extremely easy and difficult trials . Subjects completed 500 trials of this task . In this condition , subjects were on each trial presented with four oriented ellipses . On half of the trials , all ellipses were distractors . On the other half , three ellipses were distractors and one was a target . The task was to report whether a target was present . Targets were tilted μtarget degrees in clockwise direction from vertical and distractors were tilted μtarget degrees in counterclockwise direction . The value of μtarget was customized for each subject ( Table 2 ) such that an optimal observer with sensory-noise levels equal to the ones estimated from the subject’s discrimination-task data had a predicted accuracy of 85% correct ( averaged over trials with different combinations of low and high reliability stimuli ) . Stimulus display time was 67 ms and each stimulus was presented with an ellipse elongation of either 0 . 80 ( “low reliability” ) or 0 . 94 ( “high reliability” ) . On each trial , the number of high-reliability stimuli was drawn from a uniform distribution on integers 0 to 4 and reliability values were then randomly distributed across the four stimuli . The four stimuli always appeared at the four cardinal locations ( “north” , “east” , “south” , and “west” ) . Feedback was provided in the same way as in the discrimination task . The task consisted of 1500 trials divided equally over 12 blocks with short forced breaks between blocks . The three visual search conditions with external uncertainty and short display time were identical to condition A , except that the orientations of the target and distractors were not fixed , but instead drawn from partly overlapping Gaussian distributions ( Fig 1D ) . These distributions had means μtarget and −μtarget ( see above ) , respectively , and a standard deviation σexternal . The value of σexternal was customized for each subject ( Table 2 ) such that the accuracy of an optimal observer would drop by 5 , 10 , or 15% compared to the same condition with σexternal = 0 ( no external uncertainty ) . We refer to these percentages as levels of external uncertainty . Subjects completed 1500 trials divided equally over 12 blocks with short forced breaks between blocks . These three conditions were identical to conditions B-D , except for the following two differences . First , stimuli were presented with an ellipse elongation of 0 . 97 and stayed on the screen until a response was provided , such that the sensory noise levels were reduced to a presumably negligible level . Second , this condition contained 500 instead of 1500 trials . Each subject completed this condition before the equivalent condition with short display times . All statistical tests were performed using the JASP software package [35] . Besides p values we also report Bayes factors , which specify the ratio between how likely the data are under one hypothesis ( e . g . , the null hypothesis ) compared to how likely they are under an alternative hypothesis . An advantage of Bayes factors is that they can be used to both reject and support a hypothesis , whereas p values can only reject . All reported Bayes factors were computed using the default settings for the effect size priors ( Cauchy scale parameter = 0 . 707; r scale for fixed effects = 0 . 5 ) .
Before introducing the models , we derive the Bayesian decision variable for our visual search task . We denote target presence by a binary variable T ( 0 = absent , 1 = present ) , set size by N , the stimulus values by s = {s1 , s2 , … , sN} , and the observer’s noisy observations of the stimulus values by x = {x1 , x2 , … , xN} . We make the common assumption that each stimulus observation , xi , is corrupted by zero-mean Gaussian noise , i . e . , xi = si+ε , where ε is a Gaussian random variable with a mean of zero . The standard deviation of this noise distribution , denoted σi , is assumed to depend on the reliability of the stimulus , which in our experiment differed across locations ( low vs . high reliability ) . The Bayesian observer reports “target present” if the posterior probability of target presence exceeds that of target absence , p ( T = 1|x ) >p ( T = 0|x ) . This strategy is equivalent to reporting “target present” if the log posterior ratio exceeds 0 , d ( x ) ≡logp ( T=1|x ) p ( T=0|x ) >0 , where d ( x ) is referred to as the global decision variable . Under the generative model for our task ( S1 Fig ) this evaluates to d ( x ) =log ( 1N∑i=1Ndlocal ( xi ) ) , ( 1 ) where dlocal ( xi ) =exp[ ( xi+μT ) 2− ( xi−μT ) 22 ( σi2+σexternal2 ) ] ( 2 ) is referred to as the local decision variable ( see S1 Appendix for a derivation ) . Hence , the optimal decision variable is the log of an average of local decision variables , each of which represents the evidence ( posterior ratio ) for target presence: dlocal ( xi ) <1 is evidence for a distractor at location i and dlocal ( xi ) >1 is evidence for a target; a value of exactly 1 represents equal evidence for both options . We mentioned earlier that optimal observers weight each cue by its reliability . In ( Eq 2 ) , this weighting occurs through sensory noise levels σi: the larger σi , the closer the local evidence associated to stimulus xi is to 1 . The first model that we consider is the Bayesian observer without any imperfections beyond sensory noise . This observer—which we refer to as the “flawless Bayesian”—is assumed to have perfect knowledge of the statistical structure of the task and to use ( Eq 1 ) to compute its decision variable . Moreover , the flawless Bayesian is assumed to compute without error . The model’s only free parameters are the sensory noise levels σi . In conditions with unlimited display time , we fix σi either to 0 ( no noise ) or to a value obtained from a control experiment ( explained in Results ) . In conditions with short display time , we fit σi separately for stimuli with low reliability ( σlow ) and stimuli with high reliability ( σhigh ) . Heeding the concern that an excess of flexibility in optimal models can make suboptimal behavior look optimal [23] , we constrain these parameters by imposing prior distributions on their values ( see S1 Appendix ) . Moreover , we refrain from adding a bias parameter to this model , for two reasons . First , while many previous studies—including some of our own ( e . g . [19 , 21] ) –have not considered it problematic to allow for bias when testing for optimality , being biased is strictly speaking a violation of optimality . Second , and more importantly , a response bias can be confounded with biases caused by other , less obvious kinds of suboptimalities , as we will explain in our presentation of Model 2 . Our second model is a Bayesian observer with imperfections in the computation of the decision variable . Such imperfections may produce suboptimalities in performance and could be caused by many different factors , such as noise in the neural mechanisms that compute the decision variable , incomplete knowledge of the statistical structure of the task , uncertainty about the experimental parameters , and suboptimal cue weighting . To get an idea of how computational imperfections affect a Bayesian observer’s decisions , we perform simulations with imperfect variants of Model 1 . The imperfections in these variants create errors in the model’s decision variable , as compared to the decision variable of the flawless Bayesian observer . We simulate a large number of trials and find that for all tested imperfections , the distribution of this error is reasonably well approximated by a Gaussian distribution ( Fig 2 ) . Importantly , the mean of this Gaussian is not always zero , which indicates that computational imperfections may produce a systematic error in the decision variable , i . e . , a bias . Since this computational bias is indistinguishable from a simple response bias , the two can easily be confounded , which is the main reason why we did not include a response bias in the flawless Bayesian model . The finding that different kinds of suboptimality produce similar errors in the decision variable implies that it will be difficult to distinguish between them in model comparison . However , the upside of this similarity is that it allows us to test for computational imperfections in a rather general way: instead of implementing a separate model for each possible computational imperfection , we can test for a range of different imperfections by using a single model with Gaussian noise on the optimal decision variable . We implement this “imperfect Bayesian” model by adding a noise term η to ( Eq 1 ) , d ( x ) =log ( 1N∑i=1Ndlocal ( xi ) ) +η . ( 3 ) We denote the mean ( bias ) and standard deviation of this “late” noise by μlate and σlate , respectively , which are fitted as free parameters . The first two models weight each stimulus by its reliability , which is a hallmark of Bayesian observers . Model 3 is a variant that ignores differences in cue reliabilities and instead weighs them equally . In this model , ( Eq 2 ) is replaced with dlocal ( xi ) =exp[ ( xi+μT ) 2− ( xi−μT ) 22 ( σsingle2+σexternal2 ) ] , ( 4 ) where σsingle is a free parameter that determines the weight assigned to every stimulus . For lack of a better term , we refer to this model as the “ignorant Bayesian” . Model 4 is a variant of this model in which we add computational imperfections in the same way as in Model 2 , i . e . , by adding biased Gaussian noise to the global decision variable . In Models 1–4 , decisions were made based on the optimal decision variable or an impoverished variant of it . We next introduce five models with heuristic decision strategies . Just as in the Bayesian models , the decision rule in these models consists of comparing a decision variable d ( x ) with some criterion c . However , d ( x ) is now computed using simple heuristics rather than being derived from Bayesian decision theory . Moreover , criterion c is fitted as a free parameter in the heuristic models , while in the Bayesian models the optimal criterion was 0 by construction . The first heuristic model that we consider uses the maximum-of-output or “max” decision rule , which has its origin in signal detection theory [36] and is a commonly used heuristic in models of visual search ( e . g . , [16 , 37–39] ) . In the present task , the rationale is that since target orientations are on average larger than distractor orientations , one might perform well by reporting “target present” whenever the maximum observation , xi , exceeds some threshold , c . The decision variable of the Max model is thus simply the maximum stimulus observation , d ( x ) =maxxi . The next two heuristic models make decisions based on how much the stimulus observations deviate from the expected target value , which on average is smaller when a target is present . Model 6 uses the minimum absolute deviation as its decision variable , d ( x ) =min|xi−μtarget| , which again is compared with a decision criterion c . Similarly , Model 7 uses the Minkowski distance between any stimulus observation and the expected target value as its decision variable , d ( x ) = ( ∑i=1N|xi−μtarget|β ) 1/β , where β is a free parameter . Since Models 6 and 7 are both based on absolute deviations from the target , the sign of the deviation does not matter for the amount of evidence that an observation gives for target presence . This differs from the decision strategy in the Bayesian models , where a deviation in the direction of the distractor always constitutes less evidence for a target than a deviation in the direction away from the distractor . The next and final two heuristics are inspired by previous findings that the visual system represents summary statistics of the stimuli that it observes , including their mean and variance [40–42] . These statistics could be used to solve detection tasks of the kind used in our experiment , where both the mean and variance of the stimulus observations are expected to be larger on trials with a target compared to trials without a target . Therefore , Model 8 uses the mean of observations as the decision variable , d ( x ) =1N∑i=1Nxi , and Model 9 uses the variance , d ( x ) =1N∑i=1N ( xi−x¯ ) 2 , where x¯ is the average of the stimulus observations . The final five models that we consider are imperfect variants of the heuristic models . In these models , the decision variable is corrupted in the same way as in the imperfect Bayesian models . However , since bias in heuristic models is already captured in the criterion value , c , we fix μlate to 0 and only fit σlate as a free parameter . Models of perceptual decision-making tasks often include a lapse rate to account for random guesses caused by attentional lapses . In such models , it is assumed that responses on some of the trials were the result of guessing rather than a decision strategy . The lapse rate parameter specifies the estimated proportion of guessing trials . If we do not include a lapse rate in our models , then we run the risk of underestimating how good the subjects’ decision strategies were , because guessing behavior can then only be accounted for as suboptimalities in their decision strategies . On the other hand , if we do include a lapse rate , then we give models a possibility to explain away decision suboptimalities as lapses , which brings along the opposite risk: we might overestimate how good subjects’ decision strategies were . In an attempt to minimize both risks , we include a lapse rate in all models , but in the Bayesian models we constrain this parameter by imposing a prior distribution on its values ( see S1 Appendix ) . We use an adaptive Bayesian optimization method [43] to find maximum-likelihood estimates of model parameters , at the level of individual subjects . Model evidence is measured as the Akaike Information Criterion [44] and interpreted using the rules of thumb provided by Burnham & Anderson [45] . We performed a model recovery analysis [46] to verify that the models make sufficiently diverging predictions to distinguish them in a model comparison ( see S2 Fig ) .
Under the assumption that stimulus observations are corrupted by Gaussian noise , the predicted proportion of “clockwise” responses in the discrimination task is a cumulative Gaussian function of stimulus orientation . We refer to the standard deviation of this Gaussian as the sensory noise level . To verify that differences in stimulus elongation caused differences in sensory noise levels , we fitted two cumulative Gaussian models to the data . In the first model , the noise level is independent of ellipse elongation and fitted as a single free parameter . In the second model , the sensory noise levels are fitted as separate parameters for the low- and high-reliability stimuli , which we denote by σ˜low and σ˜high , respectively . The second model accounts well for the data ( Fig 1B ) and model comparison favors this model for every subject ( ΔAIC range: 0 . 50 to 22 . 3; mean±sem: 8 . 6±1 . 3 ) . Moreover , for every subject the estimated noise level is higher for the low-reliability stimulus than for the high-reliability stimulus ( Table 2 ) . Hence , the stimulus-reliability manipulation works as intended . We use noise estimates σ˜low and σ˜high to customize the target and distractor distributions in the visual search experiment ( Table 2 ) and to constrain the Bayesian models fitted to the data from that experiment ( S1 Appendix ) . While previous studies ( e . g . , [47] ) have reported that performance on discrimination tasks is sometimes better for stimuli at the vertical meridian ( “north”/“south” locations ) than for stimuli at the horizontal meridian ( “east”/“west” locations ) , we do not find evidence for such an effect in the present experiment . Performance differed little across locations , ranging from 74 . 3±1 . 1% correct at the “east” location to 75 . 0±1 . 0% at the “north” location . A Bayesian one-way ANOVA provides strong evidence for the null hypothesis of there being no effect ( BF01 = 20 . 5 , p = . 97 ) . We assume for the moment that sensory noise in the visual search conditions with unlimited display time was negligible , i . e . , σi = 0 . Under this assumption , the stimulus observations are identical to the true stimulus values , x = s , which allows us to write the optimal decision variable , ( Eq 1 ) , directly as a function of s , d ( s ) =log ( 1N∑i=1Nexp[ ( si+μtarget ) 2− ( si−μtarget ) 22σexternal2] ) . ( 5 ) Since there are no unknowns in this equation , we can compute the optimal decision variable for each trial that was presented to a subject . The flawless Bayesian responds “target present” on each trial with d ( s ) >0 and “target absent” otherwise . Hence , if subjects are optimal , then their proportion of “target present” responses should be a step function of d ( s ) , transitioning from 0 to 1 at d ( s ) = 0 . In all three conditions , subjects clearly deviate from this prediction ( Fig 3B , circles ) . Next , we fit the models to the data from the conditions with short display times . Model comparison ( Fig 4A ) selects the Imperfect Bayesian as the preferred model and rejects all other models with large margins ( ΔAIC≥19 . 6±4 . 0 ) . This result is consistent with the results above , except that both Max models are now convincingly rejected . The main conclusion that we draw from this model-comparison result is that subjects neither seem to behave optimally , nor do they seem to use a heuristic decision strategy . Instead , their decisions seem to be based on Bayesian principles that are implemented or executed imperfectly . Model comparison using cross-validation gives near-identical results ( S3 Fig ) . The reason why the Max and Bayesian model were tied in the conditions with unlimited display time is that they make near-identical predictions when all stimuli have the same reliability , as we have observed in earlier work [48] . Therefore , it is important to use mixed-reliability designs when testing these two models against each other: while reliability-based weighting is an inherent property of Bayesian decision making , there is no natural way to incorporate such weighting in a Max model . Our finding that only a Bayesian model accounts well for data from mixed-reliability conditions strongly suggests that humans take stimulus reliability into account during perceptual decision making . It is worth noting that it is unlikely that the superiority of the Imperfect Bayesian was due to it being overly flexible . First , it does not have more parameters than most of the heuristic models ( Table 3 ) . Second , while parameters in the heuristic models were entirely unconstrained , we imposed prior distribution on parameters of the Bayesian models . Third , a model recovery analysis ( S2 Fig ) showed that the Imperfect Bayesian is never selected when data are generated from one of the other 13 models . The optimality indices reported above estimate how much performance was lost due to computational imperfections . For comparison , we also estimate performance loss caused by sensory noise . To this end , we compute a variant of the earlier introduced optimality index , ( Eq 6 ) . In this variant , we “turn off” the sensory noise when computing doptimal′ , by fixing σi to 0 . This new optimality index expresses empirical performance relative to an optimal observer without sensory noise . We refer to our original index as the “relative optimality” index and to this new index as the “absolute optimality” index [3] . The difference between these two indices gives an estimate of the amount of optimality loss due to sensory noise . To illustrate this , consider an example in which a subject has a relative optimality index Irelative = 0 . 80 and an absolute optimality index Iabsolute = 0 . 70 . In this example , the subject has an optimality loss of 0 . 20 when sensory noise is not considered to be a form of suboptimality and a loss of 0 . 30 when it is . We would in this case conclude that sensory noise accounted for 33 . 3% of the optimality loss ( 0 . 10 out of a total loss of 0 . 30 ) and computational imperfections for 66 . 7% ( 0 . 20 out of 0 . 30 ) . When applying this method to the data from conditions with unlimited display time , we find that computational imperfections account for an estimated 92 . 6±3 . 8% of the performance loss and sensory noise for the remaining 7 . 4±3 . 8% . In conditions with short display time , we find that computational imperfections account for 27 . 0±5 . 1% of the performance loss and sensory noise for 73 . 0±5 . 1% . As expected , when sensory noise levels are low , performance loss is almost entirely attributed to computational imperfections . Nevertheless , even in conditions with considerable levels of sensory noise , we estimate that almost a third of the performance loss was due to computational imperfections . Next , we have a look at the best-fitting parameter estimates in the Imperfect Bayesian model . One-way ANOVAs suggest that there is an effect of the level of external uncertainty on both σlow ( BF10 = 9 . 06; p = . 005 ) and σhigh ( BF10 = 1 . 78; p = . 042 ) . Visual inspection of the parameter estimates ( Fig 5 ) reveals that this is mainly due to the condition with the highest level of external uncertainty , in which the sensory noise estimates are visibly higher than in the other conditions . However , the stimuli were extremely similar between the different conditions , which makes it implausible that there were large differences in sensory noise levels . Hence , despite our efforts to constrain these parameters , they may still have been overestimated in the condition with the highest level of external uncertainty . This means that we might have underestimated the magnitude of the deviation from optimality in that condition . For the two parameters that control the late noise distribution , we find neither an effect of the level of internal uncertainty ( BFinclusion = 0 . 25 for both μlate and σlate ) nor of the level of external uncertainty ( μlate: BFinclusion = 0 . 52; σlate: BFinclusion = 0 . 10 ) . Finally , for the lapse rate parameter we find evidence against an effect of the level of internal uncertainty ( BFinclusion = 0 . 56 ) and in favor of an effect of the level of external uncertainty ( BFinclusion = 1 . 22 ) . However , the evidence for this effect is very weak and the estimated lapse rates are very small in all conditions , so we do not consider this finding to be of any significance . While the use of mixed reliabilities is a powerful way to test predictions that are unique to Bayesian models , it has the side effect that a stimulus may “pop out” when its reliability differs from that of all other stimuli . Stimuli that pop out may inadvertently draw attention and be given more weight , which would cause a suboptimality in performance , because the optimal weight is entirely determined by the reliability of a stimulus . We find that accuracy was slightly higher on trials in which the target popped out ( 72 . 5% correct ) than on trials in which it did not ( 69 . 4% correct ) , which suggests that pop-out items indeed drew subjects’ attention . A t-test supports that there is a difference in accuracy between these two groups of trials ( BF10 = 4 . 92; p = . 008 ) . To verify that the deviation from optimality in the conditions with short display time were not entirely caused by this pop-out effect , we fit the models again after filtering out pop-out trials . In this analysis , we thus only consider trials with 0 , 2 , or 4 high-reliability stimuli ( 60% of the data ) . Note that only a third of the trials in this modified dataset has mixed reliability . As before , we find that model comparison selects the Imperfect Bayesian as the preferred model . However , the difference with the Max models is smaller now ( Flawless Max: ΔAIC = 7 . 0±2 . 2; Imperfect Max: ΔAIC = 9 . 3±2 . 3; the difference with all other heuristic models is still large , ΔAIC≥48 . 8±7 . 9 ) . This was to be expected , because we filtered out most of the mixed-reliability trials and we already established that the Max and Bayesian decision rules are indistinguishable on single-reliability data . When we constrain the parameters in the Max model in the same way as in the Bayesian models—which makes a fairer comparison—the Imperfect Bayesian outperforms both Max models with decent margins ( Flawless Max: ΔAIC = 10 . 3±2 . 5; Imperfect Max: ΔAIC = 15 . 6±2 . 6 ) . The optimality index in this analysis is 0 . 797±0 . 026 , which is nearly identical to the value we obtained in the analysis that included all trials ( 0 . 808±0 . 037 ) . Indeed , a t-test provides moderate evidence for the null hypothesis that there is no difference ( BF01 = 3 . 58 , p = . 52 ) . Altogether , our conclusions are largely the same under inclusion and exclusion of pop-out trials , which suggests that pop-out effects play a relatively minor role in explaining the identified suboptimalities . So far , we have included a lapse rate in all our models . To assess whether our conclusions would have been different if we had not included a lapse rate , we rerun all analyses with lapse rates fixed to 0 . The model comparison results are very similar to the results reported above: in conditions with unlimited display time , the imperfect Max and Bayesian models are indistinguishable ( ΔAIC = 1 . 29±0 . 95 in favor of Bayes ) and all other models are strongly rejected ( ΔAIC≥96±12 ) ; in conditions with short display time , the imperfect Bayesian is selected as the preferred model and all other models are again strongly rejected ( ΔAIC≥32 . 2±4 . 4 ) . However , the optimality indices are now slightly lower: I = 0 . 876±0 . 018 ( 13 . 3±1 . 8% deviation from optimality ) in conditions with unlimited display time and I = 0 . 796±0 . 037 ( 20 . 4±3 . 7% deviation from optimality ) in conditions with brief display time . This was to be expected , because errors that were explained as lapses in our original analysis can now only be explained by suboptimalities in the decision strategy . As before , a two-way Bayesian ANOVA suggests that there is no effect of the level of external uncertainty ( BFinclusion = 0 . 167 ) nor of the level of internal uncertainty ( BFinclusion = 0 . 842 ) on the optimality index . Altogether , we conclude that removing the lapse rate from the models does not significantly change our conclusions . Finally , we check what happens to the results when we remove the constraints on the parameters of the Bayesian models , by refitting Models 1–4 without prior distributions on parameters σlow , σhigh , and λ . The model comparison result is again very similar to our previous results: in conditions with unlimited display time , the imperfect Max and Bayesian models are indistinguishable ( ΔAIC = 1 . 4±1 . 1 in favor of Bayes ) and all other models are strongly rejected ( ΔAIC≥49 . 4±3 . 7 ) ; in conditions with short display time , the imperfect Bayesian model convincingly outperforms all other models , including the Max models ( ΔAIC≥21 . 9±3 . 9 ) . Also , we again find no evidence for an effect of the level of internal or external uncertainty on the optimality index ( BFinclusion = 0 . 56 and 0 . 19 , respectively ) . However , the estimated deviation from optimality is now 11 . 7±2 . 1% , which is substantially lower than the 18 . 1±2 . 2% that we found with constrained parameter fits . This was to be expected , because without parameter constraints , models may explain away some of the computational suboptimalities by overestimating the lapse rate and/or sensory noise levels . Indeed , the average estimated lapse rate is now 13 . 7±2 . 2% , compared to 3 . 7±1 . 1% in the constrained fits ( BF+0 = 582; p < . 001 ) . For some subjects the estimated lapse rate is now even over 50% , which seems unrealistically high . Hence , it appears that lapse rates are overestimated in the unconstrained fit . The estimated sensory noise levels , on the other hand , are very similar to the estimates obtained with the constrained fitting method ( σlow = 6 . 58±0 . 55 vs . 6 . 75±0 . 54; σhigh = 2 . 67±0 . 38 vs . 3 . 35±0 . 37 ) . Indeed , a t-test supports the hypothesis that there is no difference ( BF01 = 4 . 43; p = 0 . 44 ) . We speculate that the richness of data from mixed-reliability experiments is itself a sufficient constraint on the parameter values , in particular when subjects use a decision strategy that is sensitive to reliability differences between stimuli . A summary of our results so far is presented in Table 4 . Taken together , these results strongly suggest that our experimental subjects used a strategy that resembles the Bayesian one , but with imperfections in its execution . While model identifiability problems ( Fig 2 ) discouraged us from testing specific theories about the origin of such imperfections , there is one proposal that we believe is worth testing explicitly here , because it has some precedence in the literature . It has been argued that instead of performing exact Bayesian inference , humans may be drawing samples from the posterior distribution , which often is computationally cheaper and more tractable [49–54] . In the limit of an unlimited number of samples , Bayesian sampling is equivalent to exact Bayesian inference . However , for finite numbers of samples , Bayesian sampling leads to imperfections and biases in the observer’s decisions . To test whether Bayesian sampling may explain the decision imperfections that we observed in our data , we implemented a variant of the Flawless Bayesian model in which d ( x ) is transformed into the posterior probabilities for “target presence” and “target absent” through p ( T=1|x ) =ed ( x ) 1+ed ( x ) and p ( T=0|x ) =1−p ( T=1|x ) . While the Flawless Bayesian reports “target present” whenever p ( T = 1|x ) >p ( T = 0|x ) , the Bayesian sampling model draws n samples from a Bernoulli distribution with a success rate equal to p ( T = 1|x ) and reports “target present” when the number of success samples exceeds the number of failures , where n is a free integer parameter . We find that the Bayesian sampling model convincingly outperforms the Flawless Bayesian ( Model 1 ) with an AIC difference of 20 . 6±3 . 2 . However , it does not account for the data as well as the Imperfect Bayesian ( Model 2 ) does ( ΔAIC = 20 . 3±8 . 4 in favor of the Imperfect Bayesian ) . Therefore , we conclude that while Bayesian sampling may explain some of the decision imperfections , it cannot explain all of it .
Deviations from optimality are often taken as evidence for heuristic decision making . However , this is not necessarily true: Bayesian observers can also be suboptimal . In particular , it has been argued that imprecisions in neural systems and the need to use deterministic approximations in complex computations may be the main reason why humans are unable to perform optimally on many tasks [27–31] . Such imperfections are orthogonal to the underlying decision strategy , because they may apply to both Bayesian and heuristic decision strategies . However , most previous work has only compared models with the optimal decision strategy against models with heuristic strategies , without testing for computational imprecisions . Claims of optimality made in those works are probably too strong , because evidence for a Bayesian decision strategy does not imply optimality . To avoid such overly strong claims , we advocate using a factorial modeling approach by crossing the decision strategy ( Bayesian vs . heuristic-based strategies ) with the absence or presence of computational imprecisions . Such an approach can decompose suboptimality into two different sources: using a fundamentally wrong decision strategy and having imperfections in the execution of this strategy . Only evidence for Bayesian decision-making without imprecisions should be considered as evidence for optimal behavior . It has recently been argued that instead of focusing on the binary question whether or not a particular behavior is optimal , it would be more fruitful to start building process models that precisely characterize the sources that make humans prone to errors [23] . The approach that we took here can be seen as a step in this direction , as it aims at distinguishing between different kinds of suboptimality and quantifying the amount of performance loss caused by each of them . A similar approach was recently developed by Drugowitsch and colleagues [27] , who examined sources of suboptimality in a visual categorization task . They estimated that about 90% of the performance loss was caused by imprecisions in mental inference and the remaining 10% by stochasticity in sensory input and response selection . In our visual search task , we found a numerically similar contribution of computational imprecisions in the conditions with unlimited display time ( 92 . 6% ) . However , in the conditions with brief display time , we found that only about a third of the optimality loss was due to computational imprecisions . This can be understood by considering that sensory noise levels were probably higher in our experiment , due to a difference in stimulus presentation time ( 67 ms to encode four stimuli in our study vs . 333 ms per stimulus in the study by Drugowitsch et al . ) . We tried to further decompose suboptimalities into more specific sources , such as “noise in the computation of local decision variables” , “incorrect knowledge of the experimental parameters” , and “suboptimal cue weighting” . However , as demonstrated by the simulation results presented in Fig 2 , different types of suboptimalities have near-identical effects on the response data , due to which we were unable to reliably distinguish between them using model comparison . Future studies may try to solve this model-identifiability problem by using experimental paradigms that provide a richer kind of behavioral data to further constrain the models ( e . g . , by collecting confidence ratings [55 , 56] ) . Moreover , we believe that it may be fruitful to further investigate the Bayesian sampling hypothesis as a possible source of suboptimality in our task . Although we found that Bayesian sampling alone cannot explain the observed decision imperfections , we did not test any models that combine sampling with other sources of suboptimality . While reliability-based cue weighting is an inherent property of Bayesian observers , heuristic models do not have a natural way of taking reliability into account . Therefore , within-display manipulation of stimulus reliability provides a strong tool to distinguish between the Bayesian model and heuristic-based models in model comparison . Indeed , we found that we were unable to distinguish between the Bayesian and Max models in conditions with fixed stimulus reliability , while the Max model was convincingly rejected in conditions with mixed reliabilities . These results strongly suggest that humans—just like Bayesians—take into account stimulus reliability during perceptual decision making . This finding is consistent with previous studies that have drawn a similar conclusion in the context of not only visual search [13] , but also categorization [18] , change detection [19] , and same/different discrimination [21] tasks . However , unlike those previous studies , we do not interpret this finding as evidence for near-optimality , because we also found evidence for substantial suboptimalities that are seemingly caused by computational imperfections . Although reports of optimality have dominated perceptual decision-making literature , we are certainly not the first to report evidence for suboptimalities . For example , numerous sensory cue combination studies have reported overweighting of one of the sensory cues [57–64]; Bhardwaj et al . [65] found that visual search performance is suboptimal when stimuli are correlated; Ackermann and Landy [66] reported that subjects fail to maximize reward in a visual search task with unequal rewards across target locations; and Qamar et al . [67] found that both humans and monkeys performed suboptimally in a relatively simple visual categorization task . However , none of those studies used the factorial modeling design that we proposed and , therefore , could not distinguish between suboptimalities due to a fundamentally wrong decision strategy and suboptimalities due to computational precisions . An important aspect of our analysis is that we included models with “late noise” on the decision variable . We are not the first to do so . An example of our own previous work—in which we referred to it as “decision noise”—is the change detection study by Keshvari et al . [19] , where we found that inclusion of late noise did not substantially improve the model fits . However , sensory noise levels in that study were fitted in an entirely unconstrained way , while it is conceivable that there was a trade-off between effects of noise on the decision variable and effects of sensory noise on model predictions . Moreover , in that study we assumed random variability in encoding precision , which a later study showed may be confounded with decision noise [68] . Therefore , it is possible that computational imperfections in the study by Keshvari et al . went unnoticed due to confounding them with sensory noise or variability in precision . Another body of work that has considered noise on the decision variable are the studies by Summerfield and colleagues ( e . g . , [69 , 70] ) . They have shown that in the presence of late noise , subjects can—and often do—obtain performance benefits by using “robust averaging” , i . e . , down-weighting outlier cues when computing the global decision variable . From an optimal-observer perspective , our task can also be conceived of as an averaging task , even though the averaging is over local posterior evidence values , ( Eq 2 ) , rather than directly over stimulus values . We performed simulations to examine whether robust averaging also gives performance benefits in our task , but we did not find any evidence for this . While late noise seems to be an important factor in explaining behavior on our visual search task , it seems to play no role in explaining behavior on classical cue combination tasks [12 , 63] . There are two differences between these tasks that may explain the difference in findings . First , subjects in our task had to combine four cues instead of two . Second , and perhaps more importantly , the optimal decision rule in our task is substantially more complex: while optimality on cue combination tasks can be achieved using only linear operations , our visual search task required non-linear computations , ( Eq 2 ) . Previous work has suggested that information processing in the human brain proceeds mostly by linear additive integration ( e . g . , [71 , 72] ) , which would lead to suboptimalities if the optimal strategy requires non-linear computations . It would be interesting to investigate in future work whether subjects are perhaps using linear approximations to optimal decision rules in complex tasks such as visual search . A difference between most laboratory stimuli and naturalistic stimuli is that the former are typically deterministic , while the latter are often probabilistic [32] . In the present study , we mimicked the probabilistic character of naturalistic stimuli by adding external uncertainty . While we are not the first to do so in a perceptual decision-making task ( e . g . , [27 , 67 , 73 , 74] ) , we are unaware of any previous work that has systematically varied this level of uncertainty to them . Moreover , previous work did not examine the relation between the magnitude of the external uncertainty and the magnitude of deviations from optimality . None of our analyses provided evidence that external uncertainty affects how much performance deviates from optimality . This is somewhat surprising , because the stimulus distributions in our experiment were arbitrary and entirely novel to our subjects . It is worth noting , however , that this robustness of the degree of suboptimality under different stimulus conditions is similar to findings in an earlier study by Acerbi , Vijayakumar , and Wolpert [54] . A possible explanation could be that the brain may be familiar with Gaussian-like stimulus ambiguity and can therefore quickly incorporate novel kinds of external uncertainty , as long as it follows a Gaussian distribution . An interesting direction for future work would be to further investigate the relation between different types of external uncertainty and optimality in human decision making .
|
The main task of perceptual systems is to make truthful inferences about the environment . The sensory input to these systems is often astonishingly imprecise , which makes human perception prone to error . Nevertheless , numerous studies have reported that humans often perform as accurately as is possible given these sensory imprecisions . This suggests that the brain makes optimal use of the sensory input and computes without error . The validity of this claim has recently been questioned for two reasons . First , it has been argued that a lot of the evidence for optimality comes from studies that used overly flexible models . Second , optimality in human perception is implausible due to limitations inherent to neural systems . In this study , we reconsider optimality in a standard visual perception task by devising a research method that addresses both concerns . In contrast to previous studies , we find clear indications of suboptimalities . Our data are best explained by a model that is based on the optimal decision strategy , but with imperfections in its execution .
|
[
"Abstract",
"Introduction",
"Methods",
"Models",
"Results",
"Discussion"
] |
[] |
2019
|
Imperfect Bayesian inference in visual perception
|
Active sensing involves the fusion of internally generated motor events with external sensation . For rodents , active somatosensation includes scanning the immediate environment with the mystacial vibrissae . In doing so , the vibrissae may touch an object at any angle in the whisk cycle . The representation of touch and vibrissa self-motion may in principle be encoded along separate pathways , or share a single pathway , from the periphery to cortex . Past studies established that the spike rates in neurons along the lemniscal pathway from receptors to cortex , which includes the principal trigeminal and ventral-posterior-medial thalamic nuclei , are substantially modulated by touch . In contrast , spike rates along the paralemniscal pathway , which includes the rostral spinal trigeminal interpolaris , posteromedial thalamic , and ventral zona incerta nuclei , are only weakly modulated by touch . Here we find that neurons along the lemniscal pathway robustly encode rhythmic whisking on a cycle-by-cycle basis , while encoding along the paralemniscal pathway is relatively poor . Thus , the representations of both touch and self-motion share one pathway . In fact , some individual neurons carry both signals , so that upstream neurons with a supralinear gain function could , in principle , demodulate these signals to recover the known decoding of touch as a function of vibrissa position in the whisk cycle .
Animals navigate the world around them with actively moving sensory organs [1] . This process results in a blend of sensory input from the presence of two underlying sensory signals . One input is from the environment or object under study , while the second is from self-generated movement of the sensor [2] . The detection of an external stimulus with confidence , as well as the ability to confirm the position and trajectory of the sensor , depends on the ability of the animal to distinguish among internally versus externally generated sensations . Ambiguity among these sources leads to unpleasant outcomes , such as vertigo [3] and motion sickness [4] for the case of vestibular control . To resolve this ambiguity , nervous systems use three complementary signaling mechanisms to reference input from a sensory organ relative to the position of the sensors [5] . One is to encode self-generated sensor movement by the exo-receptors that also encode changes in the external environment; this is denoted peripheral re-afference . A second mechanism is to use muscular endo-receptors to encode elongation and contraction force , as performed by spindle fibers and Golgi tendons , respectively; this is denoted proprioception . A third mechanism is to generate a central copy of the motor commands for the intended sensor position; this is denoted corollary discharge . These three mechanisms report complementary , but not necessarily complete [6] , information on sensor position . While movement of a limb involves proprioceptive and corollary discharge reference signals , current evidence suggests that facial muscles , which bridge attachment points across soft tissue as opposed to bone , contain neither spindle fibers nor Golgi tendons [7–11] . Additional evidence demonstrates that despite the presumed lack of proprioceptors in the vibrissa musculature , neuronal signals related to rhythmic self-generated vibrissa motion , i . e . whisking , are encoded predominantly through peripheral sensory mechanisms [12–14] . Together , these observations lead to the hypothesis that self-generated vibrissa motion is encoded through re-afferent activation of mechanoreceptors . Specifically , activation of lanceolate- and/or Merkel-ending trigeminal neurons could presumably encode both re-afferent and ex-afferent input . These primary sensory neurons have identical , broad axonal arborizations across nuclei in the trigeminal brainstem [15 , 16] . Vibrissa self-motion signals are thought to inform the rodent about the position of its vibrissae upon tactile contact with an object [17–20] , though an alternative possibility based on contact forces has been proposed [21] and critiqued [22] . How might the animal determine the location of objects that it contacts with its moving vibrissae ? Past work shows that the strength of vibrissal ex-afferent touch responses , as measured in cortex , are strongly modulated by the phase in the whisk cycle at the moment of contact [20] . The responses of these units , therefore , contain the information necessary to determine object location through self-motion , but the underlying neuronal architecture required to achieve this cortical representation of object location remains unknown . Elements of signal detection theory [23] suggest two scenarios to demodulate touch relative to phase in the whisk cycle . One scenario is that the whisking and touch signals are encoded by different populations of peripheral receptors and are maintained as separate whisking and touch pathways to somatosensory cortex . A plausible scheme for demodulation involves gating of the touch signal by the separate whisking signal [20] . A second scenario is that both whisking and touch signals are encoded by the same sensory receptors and central neurons to cortex . In this case , a gain function with an accelerating nonlinearity [24] can serve to demodulate the touch signal . As a means to gain insight into the particular scenario used by rodents to merge touch and self-motion of the vibrissae , we examine the response of neurons along the two dominant ascending somatosensory pathways [25 , 26] . Our investigation is motivated by the pioneering work of Ahissar and colleagues [27] , who addressed the issue of pathways at the level of thalamus . These investigators made use of anesthetized animals , in which whisking was induced by electrical stimulation of the buccal motor branch of the facial nerve [28] . Under these conditions , the neuronal spikes rates are much reduced by the effects of anesthesia and the concurrent loss of neuromodulation . Furthermore , the process of electrical stimulation leads to the preferential activation of motoneurons with large caliper axons , as opposed to physiological recruitment , which begins with fibers of small caliper and progresses to those of larger caliper [29] . Thus there is a need for a thorough reexamination of the signaling of vibrissa input along ascending somatosensory pathways . The more familiar of the two pathways , the lemniscal somatosensory pathway , includes trigeminal nucleus principalis ( PrV ) and the upstream dorso-medial division of ventral-posterior-medial ( VPMdm ) thalamic nucleus ( Fig 1 ) . Neurons along this pathway spike vigorously in response to stimulus-induced deflection of one or multiple vibrissa [30–33] . Yet there is limited information on the nature of the response to vibrissa self-motion [34] . A second pathway , the paralemniscal pathway , encompasses the rostral aspect of spinal trigeminal nucleus interpolaris ( SpVIr ) , the upstream posterior medial ( PO ) thalamic nucleus , and includes collaterals to the ventral aspect of zona incerta ( ZIv ) , a region that further provides feedforward inhibition to PO thalamus ( Fig 1 ) [35 , 36] . Neurons along this pathway in PO thalamus spike , albeit less prominently , in response to deflection of the vibrissae [31 , 37] , yet there is apparently contradictory data on the nature of the self-motion response [27 , 38] . Lastly , we consider an alternate origin for whisking-related re-afference and ask if whisking is encoded by mechanoreceptors in the mystacial pad , which moves in phase with the vibrissae during whisking [39] . Encoding of self-motion in these receptors would represent re-afferent signals that are , in principle , independent of vibrissa touch . The result of these measurements defines the utilization of different pathways for sensorimotor signaling and constrains computational models of vibrissa-based object location [19 , 40] .
Although current evidence suggests a lack of proprioceptive innervation of most facial muscles in a number of species , data specific to the innervation of the rodent vibrissa musculature are more limited [8] . We therefore used three complementary anatomical techniques to determine whether vibrissa muscles contain endo-receptors ( Fig 2 ) . First , a classic measure to observe endo-receptors is via the labeling of spindle-like proprioceptive afferent endings [42] . Spindles appear as helical-shaped fine processes that surround intrafusal muscle fibers . Spindles are well known to be prominent in the masseter muscle [43] , as confirmed by immunostaining of neurofilament proteins from tangential sections of the muscle ( Fig 2a ) . We thus searched for spindle-like endings in the mystacial pad , in both intrinsic and extrinsic muscles , as compared to sections of masseter muscle from the same animals . The number of motoneuron endplate claws in the same sections serves to normalize our counts . We observed spindles in the vibrissa musculature in only one of three animals ( 2 , 480 endplates across 23 sections ) ( Fig 2c ) , which correspond to 0 . 0012 ± 0 . 0007 ( mean ± SD ) spindles/plate compared to 0 . 0279 ± 0 . 0054 for the masseter muscle ( 970 endplates across 36 sections ) ( Fig 2b ) . Thus the vibrissa muscles contain over 20-fold fewer spindles than a muscle with known proprioceptive control ( Fig 2d ) . As a second measure , we prepared transverse sections of both the mystacial pad and the masseter muscles and directly stained both the intrafusal and extrafusal fibers . The intrafusal fibers are identified by their small size and bundling of multiple fibers within a capsule ( arrows in Fig 2e and 2f ) . Here the total number of extrafusal fibers in a section serves as the normalization . We observed intrafusal fibers in the vibrissa musculature in two of three animals ( 53 , 800 extrafusal fibers across 13 sections ) ( Fig 2f ) , which corresponds 0 . 00011 ± 0 . 00005 intrafusal to extrafusal fibers compared to 0 . 00160 ± 0 . 00017 for the masseter muscle ( 56 , 960 extrafusal fibers across 13 sections ) ( Fig 2e ) . Thus the vibrissa muscles contain 15-fold fewer intrafusal fibers than a muscle with known proprioceptive control ( Fig 2d ) . As a final measure , we asked if γ-motoneurons , which innervate intrafusal fibers , are present in the lateral facial nucleus . This nucleus contains the motoneurons for the vibrissa musculature [44 , 45] . As a positive control , we compared staining in the lateral facial nucleus to the trigeminal motor nucleus , which innervates the masseter and other jaw muscles and , consistent with the presence of spindles in the masseter muscle ( Fig 2a ) , is known to contain γ-motoneuron efferents [46 , 47] . Recently , it has been demonstrated that γ-motoneurons can be distinguished from α-motoneurons based on their size and the relative intensity of anti-ChAT and anti-NeuN staining . Specifically , both α- and γ-motoneurons are labeled intensely with anti-ChAT , but α-motoneurons have larger somata and are labeled by anti-NeuN , whereas γ-motoneurons are smaller and are not labeled by anti-NeuN [48] . We analyzed immunohistochemical labeling on rat brainstem sections for ChAT and NeuN ( Fig 3 ) and considered only neurons whose nucleus was contained in the section as indicated by a DAPI counterstain ( Fig 3a and 3b ) . Qualitatively , the trigeminal motor nucleus contained two populations of motoneurons . Larger motoneurons were labeled both by anti-ChAT and anti-NeuN , whereas smaller motoneurons were labeled only by anti-ChAT ( Fig 3c–3e ) . In the facial nucleus , we observed only one population of medium-sized motoneurons , presumably α-motoneurons , that were labeled both by anti-ChAT and anti-NeuN ( Fig 3f–3h ) . To quantify these observations , we calculated the area of each motoneuron and the average intensity of anti-ChAT and anti-NeuN labeling within the labeled area . We observed two clusters of neurons in the trigeminal motor nucleus , putatively corresponding to α- and γ-motoneurons ( Fig 3i and 3k ) . Approximately one-third of the motoneurons fell into the putative γ-motoneuron cluster , consistent with spinal motoneuron pools that innervate muscles with spindles [48] . In the facial motor nucleus we observed a unimodal distribution of motoneuron sizes and anti-NeuN intensities , putatively corresponding to α-motoneurons ( Fig 3j and 3l ) . These results imply that the innervation of intrafusal fibers by the lateral facial motor nucleus represents at most a small fraction of its total output , and are consistent with past reports that most facial muscles lack proprioceptive signaling [7–11] . Together , these anatomical analyses of neuronal endings , muscle fibers , and motoneuron types imply that classic propriception makes a negligible contribution to the encoding of vibrissa self-motion . Given the relatively poor proprioceptive innervation of the vibrissa musculature , re-afferent activation of trigeminal mechanosensory afferents , including lanceolate and Merkel ending neuron types , is a likely source of the sensory signal of whisking phase . We thus monitored neuronal activity in two of the target nuclei in the brainstem for these neuron types [15 , 16] , nuclei PrV and SpVIr , that provide the majority of the ascending projections to thalamus ( Fig 1 ) . These nuclei anchor the lemniscal and paralemniscal pathways , respectively , and the literature is unequivocal about the presence of vibrissa touch responses in both nuclei . We recorded single and multi-unit activity in nucleus PrV ( 25 putative single unit and 31 multi-unit spiking signals ) and nucleus SpVIr ( 14 putative single unit and 10 multi-unit spiking signals ) ( Methods ) . As illustrated by the example of Fig 4 , the spike rates of units in nucleus PrV are substantially modulated on a cycle-by-cycle basis during rhythmic whisking in air ( Fig 4a and 4b ) . To quantify this modulation of the spike rate , we isolated individual whisk cycles ( Eq 1 and Eq 2 ) and aligned spike events relative to the instantaneous phase in the whisk cycle ( Fig 4c , d ) [49] . We next computed the distributions of whisking phases and of whisking phases at which spikes occurred ( Fig 4e ) . From these distributions , we estimated the spike rate as a function of phase in the whisk cycle , ( black line in Fig 4f ) and fit a sinusoid rate function ( Eq 4 ) to the data as a means to parameterize the modulation depth ( Eq 5 ) and preferred phase . For the unit in Fig 4 , the majority of spikes , i . e . , 373/404 ( 92% ) of spikes across 303 whisks , occurred during the retraction phase of the whisk cycle , when the vibrissae were moving in the caudal direction ( Fig 4f ) . Tactile receptive fields were established for a subset of the recorded units by briefly anesthetizing the rat with isoflurane and manually stimulating different vibrissae ( Methods ) . The unit in the example of Fig 4 was located among many units that responded to vibrissa E3 . The firing rates of additional example units as a function of phase in the whisk cycle , along with their local receptive fields , are shown in Fig 5 . These include units in the sub-region of nucleus PrV that corresponds to the macro-vibrissae ( Fig 5a ) and units in sub-regions that correspond to the skin and fur around the mouth and nose ( Fig 5b ) . Furthermore , we observed units in nucleus SpVIr that were significantly , albeit modestly , modulated by whisking ( Fig 5c ) . As a population , 49/56 PrV units ( 88% ) and 16/24 SpVIr units ( 67% ) were significantly modulated by whisking ( Kuiper test , p < 0 . 05 ) . Units in nucleus PrV tended to fire more spikes when the animal was whisking as opposed to not whisking ( Wilcoxon signed rank test , p = 1 . 0 x 10−5 ) , whereas spike rates were not significantly different between whisking and not whisking in nucleus SpVIr ( Wilcoxon signed rank test , p = 0 . 39; Fig 5d ) . We characterized the sinusoidal fits of spike rates across all units ( Figs 4f and 5a–5c ) by two measures . The first measure is the modulation depth , MWhisk ( Eq 5 ) , which reports the fraction of the unit’s response that is locked to whisking . The second measure is the signal-to-noise ratio , SNRWhisk , over a time interval ( T ) chosen to be the average period between whisks for head-fixed rats , i . e . , T = 165 ms ( Eq 6 ) [39 , 50] . We observe a greater modulation depth for lower mean spike rates , with a SNRWhisk that peaks at a mean rate of <λ> ~20 Hz ( Fig 6a ) . As a population , units in brainstem nucleus PrV were more strongly modulated than those in nucleus SpVIr ( Wilcoxon ranked sum test , p = 0 . 03 ) , with median MWhisk values of 1 . 0 versus 0 . 6 for nucleus PrV versus SpVIr , respectively . Furthermore , units in nucleus PrV had a greater SNRWhisk than those in nucleus SpVIr ( Wilcoxon ranked sum test , p = 0 . 0016 , with median values of SNRWhisk = 1 . 6 versus 0 . 8 for nucleus PrV versus SpVIr , respectively . Different units preferentially spiked at different phases of the whisk cycle , denoted ϕPreferred ( Eq 4 ) . All phases are represented for units in both nuclei PrV and SpVIr ( Fig 6b ) . There is a significant bias in the preferred phase across all units in nucleus PrV , with a vector average <SNRWhisk> = 0 . 6 and <ϕPreferred> = 4 . 9 radians ( Hotelling’s one-sample test; p = 0 . 02 ) ; this phase corresponds to retraction from the fully retracted position . There was no bias for units in SpVIr ( Hotelling’s one-sample test; p = 0 . 3 ) [51] . In toto , these data show that self-motion is represented along the primary nuclei of the lemniscal and paralemniscal pathways , but more robustly along the lemniscal pathway . To determine the encoding of self-generated whisking in the thalamic nuclei that receive inputs from PrV and SpVIr ( Fig 1 ) , we recorded spiking activity of individual neurons using the juxtacellular configuration in VPM and PO thalamus ( 74 neurons ) . Occasionally , extracellular recordings of nearby units were obtained on the same micropipette; these units had negative initial deflections , as opposed to the initial positive spike deflections of the juxtacellularly recorded neurons ( 3 of 71 records ) . Similarly , we recorded spiking activity of individual neurons using the extracellular or juxtacellular configuration in ZIv ( 15 neurons ) . We next consider the spiking dynamics of individual neurons in VPM and PO thalamus , as well as in ZIv , in response to self-generated whisks ( Figs 7–9 ) and external vibrissa deflections with air-puffs ( Fig 10 ) . As illustrated by the example of Fig 7 , in which the neuron was located among units that responded to vibrissa C4 , neurons in VPM thalamus are substantially modulated on a cycle-by-cycle basis during whisking ( Fig 7a and 7b ) . The analysis of the spike rate as a function of phase in the whisk cycle for thalamic neurons ( Fig 7c–7f ) proceeded similarly to that for units in the brainstem ( Figs 4 and 5 ) . For the neuron in Fig 7 , the majority of spikes occurred during the protraction phase of the whisk cycle . The spike rate from this neuron was particularly well described by a sinusoidal modulation as a function of phase ( Fig 7f ) . Additional example neurons from VPM thalamus , the adjacent sub-region in PO thalamus , and ZIv , along with their anatomical locations of the recording sites , are shown in Fig 7 , S1 , S2 and S3 Figs . Qualitatively , neurons in the sub-region of VPM thalamus that corresponds to the macro-vibrissae ( Fig 8a ) , as well as units in sub-regions that correspond to the skin or fur around the mouth and nose ( Fig 8b ) , were modulated . PO thalamus also contained a minority of neurons that were modulated ( Fig 7c ) , while modulation appeared absent in neurons in ZIv ( Fig 8d ) . As a population , neurons in VPM and PO thalamus tended to fire more spikes when the animal was whisking as opposed to not whisking ( Wilcoxon signed rank test , p = 10−9 and p = 0 . 04 , respectively ( Fig 8e ) . This is consistent with past results [52] . Neurons in ZIv tended to fire fewer spikes when the animal was whisking ( Wilcoxon signed rank test , p = 0 . 02 ) ( Fig 8e ) . Overall , neurons in VPM thalamus tended to have significantly higher spike rates than in those PO thalamus during whisking epochs ( Wilcoxon ranked sum text , p = 0 . 0057 ) , but not during non-whisking epochs ( Wilcoxon ranked sum text , p = 0 . 15 ) . Similarly to the analysis for units in nuclei PrV and SpVIr , we characterized the population response for neurons in VPM and PO thalamus and ZIv in terms of the modulation depth , MWhisk ( Eq 5 ) , and the signal-to-noise ratio , SNRWhisk with T = 165 ms ( Eq 6 ) . The majority of neurons in VPM thalamus were significantly modulated by whisking phase ( 49/57; Kuiper test p < 0 . 05 ) ( Fig 9a ) , whereas only a minority of PO neurons were significantly modulated ( 4/17; Kuiper test p < 0 . 05 ) ( Fig 9a ) and no neurons in ZIv were significantly modulated ( 0/15; Kuiper test p < 0 . 05 ) . Of the VPM neurons located among units that had receptive fields corresponding to the micro-vibrissae or peri-mystacial fur , 12/15 of these neurons were significantly modulated . As in the case for brainstem ( Fig 6b ) , different VPM neurons preferentially spiked at different phases of the whisk cycle . All phases are represented for neurons in VPM thalamus ( Fig 9b ) , but with no significant bias in the preferred phase ( Hotelling’s one-sample test; p = 0 . 72 ) . In toto , these data show that self-motion is represented in thalamic nuclei of the lemniscal and paralemniscal pathways but , as with the case of brainstem , only robustly along the lemniscal pathway . We computed the perpendicular distance between the Chicago sky blue spot and the VPM/PO border for each labeled recording site , based on cytochrome-oxidase stained sections . The location of the VPM/PO border , determined by visual inspection , was estimated to be accurate to approximately 80 μm ( S4 Fig and S4 Data ) . There does not appear to be a clear systematic relationship between the signal-to-noise ratio for whisking and the distance to the border between VPM and PO thalamus at this spatial resolution . Neurons with high values of SNRWhisk occur in VPM thalamus both close to the border as well as deeper in the nucleus ( Fig 9c ) . To further clarify whether there is a potential segregation of function within VPM thalamus , we reconstructed the locations of the labelled recording sites in three dimensions ( Fig 9d and 9e ) . Again , there is no clear spatial relationship between the location of a neuron within VPM and its SNRWhisk . The lack of topography would imply that self-generated motion and touch are signaled within the same anatomical pathway . To determine whether the same neurons respond to ex-afferent and re-afferent stimuli , we next consider how the same neurons along the lemniscal and paralemniscal pathways respond to external deflections of the vibrissae . The case for touch-based responses in the VPM thalamus , along the lemniscal pathway , is unequivocal . However , the case for touch-based responses in PO thalamus , along the paralemniscal pathway , is the subject of conflicting reports as to whether external stimuli can drive neurons in PO thalamus independent of feedback activation from deep layers in cortex . As past work involved anesthetized animals [37 , 38 , 53–55] , we undertook a re-analysis of the response of neurons in VPM and PO thalamus along with the somatosensory region of ZIv ( Fig 1 ) . As illustrated by the examples of Fig 10a–10c , neurons in all three areas were modulated by air-puff deflections to multiple vibrissae and peri-mystacial fur , with neurons in VPM thalamus responding vigorously , those in PO thalamus the least responsive ( Fig 10b ) , and those in ZIv responding with short latency , precisely timed spikes ( Fig 10c ) . Across the population , 49/54 VPM neurons ( 91% ) , 11/17 PO neurons ( 65% ) , and 12/15 ( 80% ) ZIv neurons were significantly modulated by air-puffs ( p < 0 . 05 ) ( Fig 10d ) . These data imply that nucleus SpVIr indeed drives ascending targets and that neurons in PO thalamus are responsive to stimulation in alert rats . We next consider the responses of these same neurons to self-motion of the vibrissae ( inserts in Fig 10a–10c ) . Consistent with the notion of a single anatomical pathway for re-afferent whisking and ex-afferent touch , the majority of VPM units that were modulated by self-generated whisking tended to also be modulated by air-puff deflections . Of the neurons in VPM thalamus , 42/54 ( 78% ) were significantly modulated by both air-puffs and whisking , five were modulated by whisking only , and seven were modulated by air-puffs only . Yet there does not appear to be a relationship between the fidelity of modulation for VPM neurons that are significantly modulated by both whisking and air-puffs , as measured by the correlation between signal-to-noise ratio for whisking and the peak modulation upon air-puff ( Fig 10e ) , ( r = 0 . 05 with p = 0 . 76 for VPM units ) .
Whisking-phase responses observed in VPM thalamic neurons in the present study substantially extend the results of past studies performed with both alert [34] and anesthetized [27 , 61] rats . We observe phase-dependent spiking modulation throughout the depth of VPMdm thalamus , which presumably comprises units in both the “head” and “core” regions of the barreloids [62] . This finding is consistent with results in which artificial whisking was induced by electrical stimulation of the facial nerve in anesthetized rats [27]; however , we find that units are tuned to all phases of the whisk cycle rather than to protraction onset . These broader distributions of preferred phases , which are observed in both PrV and VPM thalamus ( Figs 6b and 9b ) , are consistent with the range of phase preferences observed somatosensory cortex during natural whisking [12 , 13 , 20 , 63 , 64] . We were unable to assess whether there is a finer systematic map of the encoding of self-motion on the scale of individual barreloids [62 , 65] . Interestingly , in addition to the barreloids , we observe modulation in phase with whisking in some units that encode distortions to the skin or fur outside of the vibrissa follicle in both PrV and VPM thalamus ( Figs 5b , 6a , 8b , 8e and 9 ) . The observation that the majority of whisking responses are encoded within the lemniscal pathway raises the question of how phase-dependent touch signals , which were previously observed in somatosensory cortex [20] , arise from the observed thalamic inputs . There are at least two potential schemes that could produce these cortical phase-dependent touch signals ( Fig 11 ) . One scheme is that whisking and touch are encoded by different populations of peripheral mechanoreceptors and central neurons . In this scheme , thalamic neurons that predominantly encode the whisking signal could change the slope of the gain function of cortical neurons , i . e . , the proportionality of spike rate to input current [20] , in a phase-dependent manner ( Fig 11a ) , analogous to heterodyne detection [23] . Contrary to previously proposed hypotheses [20 , 27] , our data indicate that paralemniscal inputs are unlikely to be the source of this cortical gain modulation . However , lemniscal units that encode skin or fur distortions during whisking , which we observed in nucleus PrV and VPM thalamus , could in principle contribute to a re-afferent signal of vibrissa position that is independent of vibrissa touch ( Figs 5b and 8b ) . It remains to be determined whether such signals can influence phase-dependent touch responses in the barrels of somatosensory cortex . A more parsimonious scheme is that the same mechanoreceptors , PrV neurons , and VPM thalamic neurons encode both whisking and touch signals . In this scheme , a gain function with an accelerating nonlinearity [24] could enhance the spike rate at the peak of the whisking signal relative to other positions ( Fig 11b ) , in analogy to homodyne detection [23] and the effect of a threshold nonlinearity [66] . Based on the present results , units that encode both whisking and external vibrissa deflections could provide the relevant inputs to somatosensory cortex ( Fig 10a and 10e ) . According to this scheme , if touch occurs at preferred phase of the whisk cycle , the response is enhanced , while touch at the non-preferred phase leads to a diminished response . Such non-linear gain functions could be present at multiple stages along the sensory processing stream , including at the mechanoreceptors themselves . In fact , modulation of touch by self-motion can occur even if self-motion signal alone is sub-threshold , and the resulting threshold nonlinearity can further enhance the difference between touches at different phases . The potential role of the paralemniscal pathway in sensing vibrissa motion is controversial [27 , 38 , 53 , 67–70] . The majority of neurons in nucleus SpVIr are similarly tuned to upward vibrissa deflections of many vibrissae in anesthetized rats [71] , but are only weakly tuned to phase during whisking relative to neurons in nucleus PrV ( Figs 5 and 6 ) . Neurons in PO thalamus respond only weakly to external vibrissa deflections as a consequence of feed-forward inhibition from the output of ZIv neurons in ketamine-anesthetized rats [36] . Electrical stimulation of vibrissa motor cortex inhibits activity in ZIv , which disinhibits neurons in PO thalamus and thereby increases its responsiveness to deflections [69] . This observation led to the hypothesis that whisking-related activity in primary motor cortex [49 , 72] might gate PO thalamus so that it is sensitive to whisking . Our data suggest that while the overall firing rates of ZIv neurons decrease slightly during whisking , this decrease is not sufficient to elicit whisking-phase dependent responses in PO thalamus . The lack of phase-dependent responses in the majority of PO units in our study is consistent with a past report [38] but inconsistent with results obtained with electrically induced whisking in urethane-anesthetized rats [27] . Nonetheless , it is interesting that PO thalamic neurons have been shown to respond to vibrissa movements in the latter condition . In this respect , it remains possible that PO thalamic neurons are able to respond in a similar manner to SpVIr during a currently unknown behavioral context . In the absence of proprioception ( Figs 2 and 3 ) and corollary discharge [12] , encoding of self-generated vibrissa movement through re-afferent activation of mechanoreceptors is a means for the animal to compute the position of its vibrissae [17] . This can be used to modulate the sensory response to touch depending on phase in the whisk cycle ( Fig 11 ) . Why does self-generated movement appear to be represented differently in the vibrissa system than in the limbs , where proprioceptive and cutaneous signals are encoded in separate thalamocortical pathways [73 , 74] ? One possible explanation is that the limbs , which support the body , are likely to carry a variable load . Accurate positioning therefore requires sensory information related to muscle length and force that is independent of tactile sensation . This may also be true for jaw muscles , which are innervated by muscle spindle fibers [43] and corresponding γ-motoneurons ( Fig 3 ) [46 , 75] . The vibrissa muscles , on the other hand , support only a small , relatively constant load that consists solely of the vibrissae , which readily flex upon the application of external forces [76 , 77] . While proprioception appears to exist in the extraocular muscles [78 , 79] , other facial muscles that carry a small , relatively constant load are thought to be devoid of proprioceptive innervation [7–11] . We can only conjecture that facial expression control may follow similar mechanisms . In the case of other facial movements in which self-motion is encoded by exo- as opposed to endo-receptors , any position-dependent signal may serve as a reference signal for computing sensation in terms of sensor position .
Fifty-four female Long Evans rats , 250 to 350 g in mass ( Charles River ) , were used for combined anatomical , behavioral , and electrophysiological experiments . All behavior and electrophysiological data were obtained from head-restrained rats [80 , 81] . Rats were transcardially perfused with 0 . 1 M phosphate buffered saline ( PBS ) followed by 4% ( w/v ) paraformaldehyde in PBS . Whole rat heads were post fixed for 4 to 12 h at 4°C . Muscles were dissected off of the fixed heads and cryoprotected in 30% ( w/v ) sucrose in PBS for 8 to 12 h at 4°C . Both mystacial pad and masseter muscles were sectioned tangentially at a thickness of 60 μm with a sliding microtome . Sections were incubated in 2% ( w/v ) goat serum ( S-1000 , Vector ) block for 30 min and then the primary antibody rabbit anti Neurofilament H 1:500 ( Ab 1991 , Millipore ) overnight at room temperature . For fluorescent staining , secondary antibodies raised in goat were used ( rabbit anti-594 , A-11012 , Life technologies ) . For dark product staining , sections were incubated in biotinylated rabbit secondary antibody ( BA-1000 , Vector ) for 90 min followed by processing with an ABC kit ( PK-6100 , Vector ) and the SG peroxidase kit ( SK-4705 , Vector ) . Sections were either initially counterstained with cytochrome oxidase or a solution of 0 . 25% ( w/v ) Eosin Y in 79% ethanol and 21% water . Mystacial pad and masseter muscles were frozen in blocks of OCT ( 25608–930 , Tissue-Tek ) and sectioned transversely at a thickness of 10 μm with a cryostat . Sections were directly mounted on slides to maintain the integrity and orientation of the muscle fibers . They were left to dry for a minimum of 1 h . Slides were rehydrated and , sequentially , incubated in Mayer’s Hematoxylin Solution ( MHS15-500 , Sigma-Aldrich ) for 8 min , washed with running tap water for 5 min , differentiated in a 1% ( v/v ) hydrochloric acid in distilled water for 30 s , further washed with running tap water for 2 min , “blued” in a saturated lithium carbonate solution ( 1 . 4% [w/v] lithium carbonate in distilled water ) for 30 to 60 s , washed for 5 min in running tap water , rinsed by dipping 5 to 7 times in 95% ( v/v ) ethanol in water , counterstained with a 0 . 25% ( w/v ) Eosin Y solution in 79% ethanol and 21% tap water for 2 min , finally dried in air , and cover slipped using mounting media ( 06522 , Sigma Aldrich ) . Confocal stacks of images of spindle fibers were obtained with a Leica Sp5 . Dark product , hematoyxlin , and eosin stained slides were imaged with a slide scanning microscope ( Nanozoomer 2 . 0 HT , Hamamatsu ) . Fibers were counted using Photoshop ( CS4 , Adobe ) . Rats were perfused and fixed and the brains were extracted and sectioned at a thickness of 30 μm , as above . Sections containing trigeminal and facial motor nuclei were incubated overnight in anti-ChAT ( 1:100 AB144P , Millipore ) and anti-NeuN ( either 1:100 MAB377 , Millipore , or 10 μg/mL of a custom anti-NeuN directly conjugated to Alexa 594 ( Chemicon [82] ) . Sections were then rinsed and incubated for 90 min in anti-goat Alexa 488 ( 1:200 A11055 , Invitrogen ) and anti-mouse Alexa 647 ( 1:200 A31571 , Invitrogen ) , rinsed again , mounted , and coverslipped . Slides were scanned as described above . Motoneurons in the trigeminal and facial motor nuclei that contained a DAPI-stained nucleus were manually outlined based only on the anti-ChAT label ( green channel ) using Neurolucida software . The area and average intensity of the anti-NeuN label ( red channel ) within the outlined perimeter was then calculated . Vibrissae were clipped to approximately 2/3 of their original length and vibrissa position was monitored simultaneously with neuronal spiking activity under two behavioral conditions . First , as the rats were coaxed to whisk in air by presenting food or bedding from their home cage [83 , 89] . Second , as vibrissae were deflected externally by brief puffs of air applied to the face [81 , 83] . We monitored vibrissa position with a Basler A602f camera and a white light emitting diode backlight [50] . We chose a spatial resolution of 120 μm/pixel , a field of 360 × 250 pixels , a frame rate of 250 Hz , and a trial time of 10 s . The pixel intensity in the image was thresholded and the mean position of the full set of vibrissae was tracked by computing the center of mass of the thresholded pixels in each frame . The data were then converted into whisking angle versus time , denoted θ ( t ) . Lastly , a Hilbert transform was used to decompose the whisking angle , θ ( t ) , into the phase within the whisk cycle , ϕ ( t ) , with θ ( t ) =θAmplitudecos[ϕ ( t ) ] + θMidpoint ( 1 ) where θAmplitude and θMidpoint are slowly varying parameters and the whisking frequency , fwhisk , is given by [49]: fWhisk=12π dϕ ( t ) dt . ( 2 ) Lastly , we recall that the vibrissae tend to move in phase with one another during free-air whisking [49]; thus the phase , but not the amplitude or midpoint , of all vibrissae may be taken as identical .
|
Animals interrogate the world around them with actively moving sensory organs , resulting in a blend of sensory inputs: one input is from the object under study , while the second is from self-generated movement of the sensor . The detection of an object thus depends on the ability of the animal to distinguish among internally versus externally generated sensations . Nervous systems employ various signaling mechanisms to reference inputs from a sensory organ relative to its position . A well-known example is proprioception , in which receptors in the limb muscles and joints are used to infer the position of tactile sensors in the hands . In this case , the signals of limb position are encoded in areas of the brain that are distinct from those encoding touch . Here , we investigate the analogous problem of encoding the position of the vibrissae , or whiskers—essential orofacial sensorimotor organs in rodents . In contrast to the case for limbs , we find that vibrissa position is encoded along the same neuroanatomical pathway as vibrissa touch . The seeming ambiguity that results from the mixed representation of position and touch can be resolved by a nonlinear neuronal input-output relation that demodulates touch with respect to vibrissa position . This scheme enables the rodent to determine where an object is located relative to its body axis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Vibrissa Self-Motion and Touch Are Reliably Encoded along the Same Somatosensory Pathway from Brainstem through Thalamus
|
Division of labor and task specialization explain the success of human and insect societies . Social insect colonies are characterized by division of labor , with workers specializing in brood care early and foraging later in life . Theory posits that this task switching requires shifts in responsiveness to task-related cues , yet experimental evidence is weak . Here , we show that a Vitellogenin ( Vg ) ortholog identified in an RNAseq study on the ant T . longispinosus is involved in this process: using phylogenetic analyses of Vg and Vg-like genes , we firstly show that this candidate gene does not cluster with the intensively studied honey bee Vg but falls into a separate Vg-like A cluster . Secondly , an experimental knockdown of Vg-like A in the fat body caused a reduction in brood care and an increase in nestmate care in young ant workers . Nestmate care is normally exhibited by older workers . We demonstrate experimentally that this task switch is at least partly based on Vg-like A–associated shifts in responsiveness from brood to worker cues . We thus reveal a novel mechanism leading to early behavioral maturation via changes in social cue responsiveness mediated by Vg-like A and associated pathways , which proximately play a role in regulating division of labor .
Division of labor—the distribution of work among group members—is a characteristic trait of many social species , including humans , other social mammals , and social insects . An important component of division of labor is task specialization , which increases the efficiency in performing a task and hence contributes to the evolutionary success of division of labor [1–4] . The occurrence and benefits of task specialization , however , might be context specific and are occasionally questioned [5–11] . The assignment of workers to their respective roles is rarely regulated by the queen or other dominant group members and is only in a few cases genetically determined [12] . Instead , division of labor is typically self-organized , and specialization of workers in specific tasks is affected by a variety of factors , including morphology , age , fecundity , and corpulence . For instance , in many ant and bee species , hard-wired developmentally regulated physiological and/or morphological castes can be found , resulting in highly specialized workers , e . g . , ant soldiers [2 , 13] . In species with a monomorphic worker caste , task specialization is often not a lifetime choice but typically undergoes multiple changes over a worker’s life: recently hatched , corpulent , and fertile workers stay on the brood pile and tend the brood . Later in life , following a shrinkage of lipid reserves and ovaries , workers switch to extranidal tasks such as foraging [2 , 14–20] , suggesting that the behavioral progression from internal to external tasks is rigid and unidirectional . However , workers in many social insect taxa retain a certain behavioral flexibility . For example , in primitively eusocial species , such as some wasps and bumble bees , worker task choice is only weakly linked to age [21–27] ( but see [28 , 29] ) , and workers can frequently switch between inside tasks and foraging [28] . In ants and honey bees , foragers can switch back to inside tasks in case of a sudden increase in brood-care demands [30–35] . In the reverse case—i . e . , a removal of foragers—young workers accelerate their behavioral progression from inside to outside tasks ( e . g . , [32 , 35] , but see , e . g . , [36 , 37] ) . Such high behavioral flexibility is inconsistent with the concept of behavior being directly linked to age and physiological status , as these factors are impossible ( in the case of age ) or difficult ( in the case of physiological status [38] ) to adapt . Instead , theoretical considerations based on mathematical models led to the development of two prominent models explaining task choice and division of labor in social insects . The foraging for work ( FFW ) model states that division of labor is generated via the spatial distribution of tasks in more or less distinct areas of the nest . As workers differ in their typical spatial position and seek open jobs in their vicinity , workers take over different duties [39–41] . The observed age-dependent choice of tasks can thus be explained by young workers hatching from the centrally located brood pile and physically transposing older workers to the periphery , where they then encounter different tasks [42] . The response threshold ( RT ) model is based on the idea that task choice is controlled by a set of internal RTs regulating the responsiveness of a worker toward different task-related stimuli [43 , 44] . To induce a behavioral response in a worker , the task-related stimulus has to exceed the internal RT . These thresholds are influenced by gene networks and cascades , which , in turn , are affected by the aforementioned physiological parameters such as age . In line with that , brood carers and foragers exhibit broad differences in gene expression and splicing [32–34 , 45–50] . Given that the expression of behaviorally linked genes is influenced by physiology—determined by , e . g . , age , fertility , and corpulence—typical patterns of task allocation can be explained by the RT model [44] . Unlike a solely physiology-based behavioral progression of workers , both the FFW and the RT model can explain typical patterns of age polyethism while simultaneously allowing for behavioral flexibility: if the demand for a certain task—e . g . , brood care—increases , both the likelihood to encounter such open jobs ( FFW model ) and the stimulus intensity associated with this demand ( RT model ) will rise , and even workers with high thresholds for brood care–related tasks , such as foragers , will switch to take over this job . Recent studies on honey bees provide experimental evidence for the importance of this stimulus–RT–gene expression association in influencing task choice . In particular , forager bees exhibit a high expression of the Amfor gene [51] , which increases their responsiveness toward sucrose solution and results in elevated nectar-foraging activity [52] . The role of foraging and associated pathways in division of labor in other social Hymenoptera , however , appears to be less clear [53–56] . In contrast to foraging , brood care activity is likely influenced by the insulin-like growth factor 1 ( IGF-1 ) pathway , an important endocrine network associated with fertility and behavior in insects [57–59] . Well-characterized members of this network are vitellogenin ( Vg ) genes and the associated juvenile hormone ( JH ) : in honey bees , high expression of the single Vg copy is linked positively to brood care behavior and negatively to the onset of foraging [60–62] . When Vg is down-regulated , JH titers increase and result in reduced brood care , precocious foraging , and a forager-like gene expression profile [60–64] . Although the link between Vg and brood care behavior is well documented in bees and was even expanded to noneusocial systems [65] , little is known about how changes in the expression of Vg and downstream effects contribute to alterations in behavior and task choice . One suggested mechanism is that a down-regulation of Vg is involved in increasing gustatory responsiveness and might thereby translate into precocious foraging in honey bees [66] . Moreover , it is still unclear to what extent functional annotations of and conclusions based on the honey bee Vg can be expanded to other social Hymenoptera such as ants , bumble bees , and wasps . Firstly , in contrast to honey bees , the association between Vg and JH is less clear in other social insects , suggesting fundamental changes in the underlying networks regulating behavior , e . g . , in some ants [67–69] . Secondly , Vg underwent several duplication events followed by diversification and subfunctionalization in ants and termites , potentially resulting in different Vg orthologs taking over different functions in physiology—namely , fertility and aging—and behavior , including division of labor [46 , 70–75] . Thirdly , 3 new clusters of Vg-like genes have recently been discovered in many social and solitary species [73] . Despite their structural similarity to the conventional Vgs , their function is unknown but has been suggested to be linked to inflammation and oxidative stress response [76] . Fourthly , a clear picture of Vg copy numbers , their phylogenetic relationships , and functions is lacking , and nomenclature of the different Vgs in different organisms is rather eclectic . The identification of gene pathways and networks involved in the regulation of behavior requires controlling for confounding physiological factors to disentangle the effects of physiology and behavior on gene expression [32] . We thus experimentally manipulated the age structure of ant colonies by removing either all old or all young workers in a full factorial design regarding task ( brood carer , forager ) , age ( young , old ) , and fertility ( fertile , infertile ) [34] . This allowed us to investigate the independent contribution of each factor on behavior as well as the underlying gene expression patterns [34] . Among the genes in which expression was linked to behavior—in this case , brood care behavior—Vg-like A [73] stood out as a highly promising candidate to influence task choice and division of labor , as it exhibited the strongest expression difference between brood carers and foragers ( false discovery rate [FDR] = 5 . 45 × 10−18 ) , and its expression was independent from age ( FDR = 0 . 99 ) and fertility ( FDR = 0 . 99 ) [34] . Here , we use a series of RNA interference ( RNAi ) -induced Vg-like A knockdowns followed by behavioral assays , physiological and chemical measurements , and tissue-specific quantification to functionally characterize this gene in T . longispinosus and to elucidate its role in task choice and , by that , in division of labor . In particular , we investigated ( i ) how worker behavior is influenced by age and task demands , ( ii ) whether Vg-like A influences behavior and behavioral progression and how Vg-like A interacts with worker age , ( iii ) whether the Vg-like A–associated behavioral regulation is modulated via responsiveness to task-related stimuli ( as predicted by theory ) , ( iv ) the tissue-specific expression of Vg-like A and other Vg-like and Vg orthologs and whether these are affected by Vg-like A knockdown , and ( v ) the phylogenetic position of Vg-like A relative to other Vgs and Vg-like genes within Hymenoptera , including both solitary and social insect taxa . Answering these questions will contribute to our understanding of the mechanistic underpinnings of behavioral plasticity and how these mechanisms respond to and interact with a worker’s social environment . Furthermore , investigating the role of Vg orthologs and their associated pathways in the regulation of behavior in other systems than the honey bee offers the unique possibility to reconstruct the mechanistic underpinnings of different evolutionary routes toward eusociality .
We manipulated colony demography in field-collected colonies of the ant T . longispinosus by removing either ( i ) all newly hatched ( termed “young” ) workers in the lab , ( ii ) all workers at least 1 year old ( “old” ) , or ( iii ) half of each age cohort as a control . Based on subsequent behavioral observations of 12 workers per nest , we detected the typical age-dependent division of labor in our control colonies consisting of both young and old workers . In particular , the “frequency of brood care” was influenced by an interaction between “age” and “colony treatment” ( generalized linear mixed model [GLMM]: family = quasi-Poisson , χ2 = 3 . 8 , p = 0 . 049; Fig 1A ) , and brood care was mainly conducted by young workers in control colonies ( Model estimate: z = 4 . 5 , p < 0 . 0001 ) . The “frequency of foraging , ” in contrast , was independent from this interaction ( GLMM: family = quasi-Poisson , χ2 = 0 . 7 , p = 0 . 419; Fig 1B ) and only affected by “age” ( GLMM: χ2 = 15 . 6 , p < 0 . 0001 ) , with old workers foraging more often than young workers irrespective of “colony treatment” ( GLMM: χ2 = 0 . 1 , p = 0 . 742 ) . When removing all young workers from a colony , old workers responded with elevating their “frequency of brood care” to a level close to significance ( Model estimate: z = 2 . 0 , p = 0 . 051 ) and up to the level of young workers ( Model estimate: z = 0 . 7 , p = 0 . 509 ) . In the reverse case , under old worker removal , young workers did not increase their “frequency of foraging” ( Model estimate: t = 0 . 2 , p = 0 . 831 ) . These age-dependent differences in plasticity were mirrored in the spatial location of workers , which were influenced by an interaction between “age” and “treatment” ( permutational multivariate ANOVA [PERMANOVA]: F = 4 . 0 , p = 0 . 008 ) . In particular , old workers moved to the inside of the nest when all young workers were removed ( PERMANOVA: F = 8 . 0 , p = 0 . 0001 ) , whereas young workers did not relocate to the nest periphery in the reverse case ( PERMANOVA: F = 0 . 1 , p = 0 . 959 ) . In honey bees , the care of adult nestmates has been described as an intermediate step in the behavioral progression from brood care to foraging [77] . In line with this , we found adult nestmate care—i . e . , feeding , grooming , or carrying of adult workers—to be preferably taken over by old workers ( GLMM: χ2 = 3 . 8 , p = 0 . 049; Fig 2A ) . We conducted a series of knockdown experiments to functionally annotate Vg-like A and to reveal the impact of the pathway and network involving Vg-like A on task choice , behavioral progression , and division of labor . We first conducted a short-term ( i . e . , 7-day ) Dicer-substrate small interfering RNA ( dsiRNA ) -mediated whole-colony knockdown of Vg-like A by providing dsiRNA fragments dissolved in sucrose solution for oral uptake . This resulted in a 62 . 9% down-regulation of Vg-like A in the fat body of brood carers ( see below ) . Vg-like A down-regulation reduced brood care behavior in young ( i . e . , recently hatched ) workers and field-collected brood carers but not in other behavioral castes , which generally conducted little brood care ( Fig 3A ) . This reduction in brood care cannot be explained by an overall decrease in behavioral activity due to potentially harming effects of the Vg-like A knockdown , as the “frequency of inactivity” was independent from “treatment” ( GLMM: χ2 = 1 . 3 , p = 0 . 255; S1 Fig ) . As we provided dsiRNA for the entire colony , a reduction in brood care could potentially be explained by Vg-like A knockdown–induced alterations in larval rather than worker behavior . We tested this by performing individualized brood care tests with a full factorial design regarding worker and larva dsiRNA treatment . The amount of brood care provided within 5 minutes was only influenced by “worker treatment” ( GLMM: χ2 = 5 . 5 , p = 0 . 019; Fig 3B ) but independent from “larval treatment” ( GLMM: χ2 = 1 . 8 , p = 0 . 176 ) or an interaction between both factors ( GLMM: χ2 = 0 . 8 , p = 0 . 361 ) . Despite the reduction in brood care , we did not detect an effect of “treatment” on the “frequency of foraging” ( GLMM: treatment: χ2 = 0 . 0 , p = 0 . 935; caste: χ2 = 68 . 1 , p < 0 . 0001; interaction: χ2 = 0 . 2 , p = 0 . 997; S2 Fig ) . The only indication , though weak , for an ongoing transition toward outside tasks was a reduced light aversion of Vg-like A− brood carers ( t test: t = 2 . 1 , p = 0 . 040 , Fig 3C ) . As Temnothorax workers have a life expectancy of up to 7 years [78] , much higher than the lifespan of honey bee workers [79] , we tested whether a prolonged Vg-like A knockdown induces foraging in brood carers and repeated our experiment with two changes in the setup: Firstly , we increased the knockdown period from 7 to 33 days ( named “long-term knockdown” ) . Secondly , as we were specifically interested in Vg-like A knockdown–associated accelerated behavioral maturation , we exclusively observed 3-month-old brood carers . Again , a reduction in brood care occurred ( GLMM: χ2 = 31 . 1 , p < 0 . 0001; S3 Fig ) , but still no increase in foraging activity ( GLMM: χ2 = 0 . 1 , p = 0 . 794; S4 Fig ) . Furthermore , fecundity—measured as the mean ovary length and the ratio of yolk-enriched/transparent eggs—was , as expected , higher in queens than in workers ( GLMM: mean ovary length: χ2 = 215 . 7 , p < 0 . 0001; yolk-enriched/transparent egg ratio: χ2 = 136 . 7 , p < 0 . 0001; Fig 4 ) but independent from Vg-like A treatment ( GLMM: mean ovary length: χ2 = 0 . 2 , p = 0 . 639; yolk-enriched/transparent egg ratio: χ2 = 0 . 2 , p = 0 . 677 ) or an interaction between “caste” and “treatment” ( GLMM: mean ovary length: χ2 = 0 . 8 , p = 0 . 370; yolk-enriched/transparent egg ratio: χ2 = 0 . 0 , p = 0 . 967 ) . A comparison of cuticular hydrocarbon ( CHC ) profiles of brood carers and foragers of both treatments revealed a strong caste effect ( PERMANOVA: pseudo-F = 8 . 072 , p = 0 . 001 , S5 Fig ) —i . e . , the two behavioral castes differ in chemical profile—but no effect of Vg-like A knockdown ( PERMANOVA: pseudo-F = 0 . 618 , df = 1 , p = 0 . 70 ) or an interaction between “caste” and “treatment” ( Permanova: pseudo-F = 0 . 532 , df = 1 , p = 0 . 67 ) on cuticular chemistry . However , young Vg-like A− brood carers increased adult nestmate care ( Fig 2A ) , a task found before to be typically taken over by old workers . Thus , our results suggest that a down-regulation of Vg-like A accelerates behavioral progression from brood care to adult nestmate care but not all the way to foraging . The most likely proximate mechanism underlying the observed behavioral progression is a shift in responsiveness from brood- to adult worker–related stimuli . To test this hypothesis , we compared the responsiveness of young ( 30-day-old ) and old ( 1-year-old ) workers to CHC extracts of larvae and adult workers under control and Vg-like A knockdown conditions ( Fig 2B ) . In the control with unmanipulated Vg-like A expression , young workers and old workers differed in their preference ( Binomial GLMM: z = 3 . 6 , p = 0 . 0002; Fig 2C ) , but this difference vanished under Vg-like A knockdown ( Binomial GLMM: z = 1 . 0 , p = 0 . 318 ) . In response to a Vg-like A down-regulation , young workers changed their preference to adult worker odors , thereby resembling old workers ( Binomial GLMM: z = 2 . 5 , p = 0 . 013 ) , whereas old workers did not change their behavior ( Binomial GLMM: z = 1 . 1 , p = 0 . 282 ) . We quantified the expression of Vg-like A and all other Vg and Vg-like genes in the brain , fat body , and ovaries of T . longispinosus brood carers via quantitative real-time PCR ( qPCR ) . We dissected workers and pooled the tissues of 5 workers from 22 colonies , resulting in 11 samples for each tissue for the RNAi treatment and control , respectively . Vg-like A expression was influenced by an interaction between “tissue” and “treatment” ( GLMM: χ2 = 20 . 8 , p < 0 . 0001 ) . In particular , Vg-like A was down-regulated by 62 . 9% in the fat body ( Model estimate: t = 5 . 3 , p < 0 . 0001 ) , the tissue in which it was most highly expressed ( fat body versus brain , model estimate: t = 8 . 2 , p < 0 . 0001; fat body versus ovaries , model estimate: t = 8 . 0 , p < 0 . 0001 ) , but not in the brain and in the ovaries ( Fig 5 ) . The down-regulation of Vg-like A was not apparent in the whole-body samples ( Wilcoxon test: W = 40 , p = 0 . 442 ) . The expression of CVg , MVg2 and MVg3 , and Vg-like B and Vg-like C was not altered by Vg-like A knockdown in either of the 3 tissues ( all p-values > 0 . 05 , Fig 5 for statistical details ) . However , each of these genes showed clearly tissue-specific expression , with , e . g . , the myrmicine Vgs ( MVg2 and MVg3 ) being mainly expressed in the fat body ( MVg2: fat body versus brain , model estimate: t = 5 . 2 , p < 0 . 0001; fat body versus ovaries , model estimate: t = 4 . 9 , p < 0 . 0001; MVg3: fat body versus brain , model estimate: t = 5 . 4 , p < 0 . 0001; fat body versus ovaries , model estimate: t = 5 . 1 , p < 0 . 0001 ) . The BLAST searches of our Vg contigs resulted in highly ambiguous annotations , possibly due to the highly conserved structural domains [73] . To obtain reliable Vg cluster membership , we used Vg sequences of 2 publications [72 , 73] as queries for an extensive BLAST search of Vg copies in a total of 34 insect genomes . A number of Vg copies , as well as Vg-likes , have previously not been annotated in several taxa ( complete list of Vg copies per species in S1 Table ) . Moreover , some genome annotations contained “fused” protein sequences of two Vgs combined into one ( see S2 Table for details ) . We split these “fused” proteins into 2 protein sequences . We used our BLAST results and sequences from the published alignments to construct an extensive maximum likelihood tree with RaXml [80] containing 5 distinct Vg clusters ( Fig 6 ) . The Vg-like genes ( according to [73] ) form 3 distinct clusters , which are present in all hymenopterans , with some of them occurring in nonhymenopteran taxa such as mosquitos , butterflies , and beetles . Vg-like C was found in Hymenoptera only . The Vg genes form 2 distinct clusters , with CVg being present in all investigated insects , whereas the second cluster , MVg , occurs only in ants of the subfamily Myrmicinae . Within each gene cluster , the expected taxonomic groups are recovered , with the formicine , myrmicine , and ponerine species clustering together ( Supporting information ) as well as the apoid bees . The number of CVgs , MVg , and Vg-likes is highly variable between taxa , pointing to species-specific duplications and losses . The T . longispinosus contig of interest for this study grouped to the Vg-like A cluster and was annotated as such .
The likelihood that a social insect worker takes over an open task is influenced by a complex interaction of age , physiology , and gene expression influencing internal RTs for task-related stimuli . The expression of these genes typically follows age-dependent physiological changes but—under certain conditions—can be modified in response to increased demand for a certain task . Vgs are known for their role in fertility and division of labor in social insects [60–67 , 71] . Here , we show that in the ant T . longispinosus , typical patterns of age-dependent division of labor can be found , with young workers tending the brood and old workers taking over the care of adult nestmates and foraging . Old workers were more plastic in their behavior and could return to brood care tasks if necessary , whereas young workers failed to accelerate their behavioral progression toward foraging . Our detailed functional annotation of Vg-like A—a gene highly expressed in brood carers , independent of their age and fertility status [34]—showed that Vg-like A , or the networks it is embedded in , is involved in mediating cue responsiveness and may hence play a role in the regulation of division of labor . In particular , Vg-like A down-regulation caused young workers to decrease their investment into brood care and to switch to the care for adult nestmates , a behavior typically exhibited by workers older than 1 year . These nestmate care workers are probably functionally similar to middle-aged honey bee workers that quit their brood care duties [81] and exhibit a broad behavioral repertoire of various intranidal tasks , including nest building , nectar receiving , and guarding [82] . The only gene previously found to influence behavior by modifying cue responsiveness is Amfor in honey bees , which regulates the onset of foraging by lowering the threshold for sucrose solution , a food-related stimulus [51 , 52] . In contrast to the Amfor-associated initiation of foraging , we provide evidence for early behavioral progression being regulated by an interplay between the pathways involving Vg-like A , worker age , and the relative responsiveness to specific chemical stimuli of brood and adult workers . Based on gene expression patterns [34] and our finding that the expression of Vg-like A was linked to the initiation of brood care behavior , we conclude that , across all workers in a colony , brood carers exhibit the highest expression of Vg-like A . In unmanipulated colonies , typically the youngest workers take over brood care and are thus characterized by a high expression of Vg-like A , leading to a high sensitivity to brood stimuli and a high investment into brood care ( Fig 1 , [34] ) . After the hatching of a new worker generation in the summer , the recently hatched workers take over brood care duties and consequently replace the previous brood carers . The eclosion of new workers thus results in a temporary oversupply of brood carers and a reduction of the intensity of brood care–associated stimuli . Moreover , when young workers start emerging from the brood pile , previous brood carers are physically displaced from the center to the nest periphery , where the intensity of brood care stimuli is even lower [40] . As a consequence , the likelihood that previous brood carers are recruited to brood care tasks drops significantly once new workers start hatching [82 , 83] . Potentially facilitated via experience-based feedback loops [82–84] and at least partly influenced by a reduction in the expression of Vg-like A , chemical responsiveness is shifted from brood- to adult-worker cues . These former brood carers then direct their caring behavior toward adult nestmates , resulting in a behavioral repertoire including , e . g . , intranidal hygenic behaviors and antennation of returning foragers at the nest entrance . This “push away” of workers from the brood pile is accompanied by a constant “pull out” of workers to outside tasks—e . g . , by foragers dying in the field—and results in a behavioral progression from brood care to adult nestmate care and from adult nestmate care to foraging . Such a push-pull model of behavioral progression has previously been proposed in honey bees [82] . We now expand this model by showing that the transition is internally influenced via Vg-like A and its associated pathways . These findings provide new evidence for proximate mechanisms underlying 2 prominent models of division of labor: Vg-like A–associated pathways increase responsiveness toward brood stimuli and can either explain the preferred area a worker is seeking a job in ( FFW model [39] ) or the preferred target of a certain behavior ( RT [14 , 44] ) . We present clear evidence that the progression from brood- to adult-nestmate care is at least partly controlled by a gene network including Vg-like A , whereas we do not know which genes control the subsequent steps of the behavioral progression . Foragers of the carpenter ant Camponotus aethiops behave more aggressively toward intruders than brood carers , which has been attributed to increased sensitivity toward nonnestmate stimuli [85] . In the ponerine ant Harpegnathos saltator , the neuropeptide corazonin has recently been shown to reduce Vg expression and to induce hunting behavior [86] . Whether corazonin takes over a similar role in Temnothorax ants remains unclear , as the RNA of this neuropeptide was not differentially expressed between brood carers and foragers in T . longispinosus [34] . Moreover , the Vgs of the two ponerine ants H . saltator and Dinoponera quadriceps form a monophyletic group , which is basal to all Vg-like genes , and they might therefore fulfill different functions . After experimentally removing all recently hatched young workers , old workers compensated this loss by switching back to brood care . Vg-like A was more highly expressed in brood carers compared to foragers , independent of their age and fertility status [34] . Hence , a potential proximate mechanism underlying the behavioral reversal includes the up-regulation of Vg-like A , which was found in old reverted brood carers , compared to old workers who remained foragers ( FDR = 0 . 0005 ) . These reverted brood carers would then redirect their caring behavior from nestmates back to eggs , larvae , and pupae . When removing all old workers from the nest , young workers failed to increase the frequency of nestmate care and foraging activity . This points to a constraint of young workers to accelerate their behavioral progression , e . g . , via a down-regulation of Vg-like A . This is surprising because it cuts off the colony from resources . High thresholds for stimuli associated with tasks such as foraging might have prevented the young workers’ shift to outside tasks . Young workers have a higher life expectancy , fertility , and lipid reserves than old workers [14 , 15 , 17] and are thus more valuable for the colony . This is especially true for Temnothorax workers , which can live for several years [78] and thus have a high residual lifespan after hatching . To avoid the unnecessary exposure of valuable inside workers to elevated external mortality , selection should implement multiple thresholds that must be reached before workers switch to risky outside tasks . On a proximate level , young workers might only shift to foraging when they are pushed ( by the emergence of new workers ) and pulled by the shortage of resources . However , as young workers have large lipid reserves , they might respond to colony hunger slowly . It is therefore likely that old workers are more sensitive to cues , which cause them to switch back to inside tasks , than young workers to outside tasks . Different orthologs of CVg and their pathways have extensively been studied , especially in honey bees , and are not only involved in fertility [87] but also in the regulation of brood care , the onset of foraging , longevity , immunocompetence , and the regulating of ageing [60–62 , 88 , 89] . Contrary to the studies on conventional Vgs , we provide first experimental evidence that Vg-like A–associated changes appear to be highly specialized in regulating responsiveness to worker and brood chemical cues . The knockdown of Vg-like A did not result in precocious foraging or forager-like CHC profiles , and we found no evidence for Vg-like A influencing fertility in workers or queens . The involvement of Vg-like A in immunocompetence and longevity ( as has been suggested for honey bees [76] ) still needs to be elucidated , although its expression was independent from worker age [34] . Moreover , our study confirms earlier ones indicating the existence of multiple copies of Vgs , including Vg-likes in ants [46 , 70–73] . Vg nomenclature has been rather inconsistent , with Vg genes and copies being consecutively numbered in successive genome annotations without clear structure . These inconsistencies hampered between-study comparisons of Vg genes , especially between Vg types of the “conventional” Vgs within ants . Our extensive BLAST search across all currently available ant genomes plus a number of additional social and solitary hymenopterans and other insects now resolves this and will help to clarify Vg classification in the future . Our phylogenetic tree provides the following novel insights: ( a ) Vg-likes ( according to [73] ) form 3 distinct clusters , with Vg like A and B being monophyletic , and they are present in all investigated insect species; ( b ) conventional Vgs form 2 distinct clusters with CVg , also being present in all investigated insect species , whereas the second cluster , MVg , is only found in ants—more specifically , only in myrmicine species . This finding suggests that MVg might have originated by duplication in an ancestor of the myrmicine ants but was consequently lost in some myrmicines . ( c ) The number of Vgs and Vg-likes varies strongly between ant , bee , and wasp taxa , which points to frequent gene duplications but also the recurrent loss of Vg genes . However , this finding could potentially be in part explained by incomplete genome assemblies . Although expression or functional data for most of the Vg and Vg-like genes are not available , the tissue-specific expression of these genes in T . longispinosus shown here and evidence from other ants [46 , 70 , 71 , 73] indicate different functions and caste-specific expression of the various Vg genes and may vary even between species . In summary , we show how the expression of Vg-like A and its downstream network influences task choice of ant workers by modulating RTs to 2 social stimuli . We herewith discovered a mechanism that explains how age-dependent behavioral progression and behavioral flexibility can be achieved at the same time . The pathways involving Vg-like A might therefore have an important , but yet overlooked , impact on worker behavior and division of labor . Vg-like A is not only present in the genomes of social insects but also in those of solitary wasps and bees and in non-Hymenoptera taxa , including beetles and mosquitos . Hence , an across-taxa functional annotation of Vg-like A might shed more light on the evolution of brood care behavior in insects in general . Moreover , as the expression of Vg-like A in the fat body controlled the behavioral progression , our findings indicate that future research on the proximate basis of behavior should not only focus on the brain but also on other tissues and their cross communication .
In June 2014 , 38 monogynous colonies of the ant T . longispinosus with an average colony size of 29 . 0 ± 1 . 5 workers were collected at the E . N . Huyck Preserve , Rensselearville , New York , United States of America , which also provided the collection permit . T . longispinosus colonies of this population have a synchronized annual brood production , with larval development taking about 1 year and new workers emerging July–August . At the time of collection , the brood of the previous year had not yet hatched; hence , all collected workers were at least 1 year old ( termed “old” in the following ) . In our laboratories at the Johannes Gutenberg-University in Mainz , Germany , colonies were separately relocated into nests consisting of 2 glass slides separated by a piece of plexiglas providing a cavity of 4 . 9 cm × 1 . 1 cm × 0 . 3 cm . These slide nests were placed in 3-chambered boxes with a moistened plaster floor . Colonies were maintained at 14 h:10 h Light:Dark photoperiod and a +22 °C:+18 °C temperature regime to facilitate closing of the new worker generation . Ant colonies were fed twice a week with honey and pieces of crickets . In 24 randomly chosen colonies , all old workers were labeled with a thin metal wire ( 0 . 02 mm , Elektrisola ) around the postpetiole . In the remaining 24 colonies , all young workers were labeled . Twenty-eight days after the hatching of the young workers was completed , colony demography was manipulated in 14 colonies per treatment group . Treatments included the removal of either ( i ) all freshly hatched workers in the lab ( termed “young” in the following ) , ( ii ) all old workers , or ( iii ) half of each age cohort as a control . This manipulation aimed at forcing young workers to start foraging by the removal of the old foragers—and old workers to switch back to brood care—through the removal of young brood carers . Workers were then given 21 days to adapt to this manipulation and to reorganize their division of labor . To investigate the investment of single workers into brood care , nestmate care , and foraging , 6 foragers ( observed being outside or in the nest entrance ) and brood carers ( observed actively conducting brood care when nest was opened ) were individually marked with a thin colored metal . It has been shown in T . longispinosus and another ant species that a single observation is sufficient to group workers into brood carers and foragers differing in behavior [17 , 90] , gene expression [34] , CHC composition ( S5 Fig ) , and life expectancy . Furthermore , spatial location has already been shown to predict behavior in Temnothorax ants [41] . Over the following 3 days , marked workers were scanned 10 times per day—i . e . , a total of 30 times—by recording their position and behavior ( S3 Table ) . We adhered to a period of at least 30 minutes between scans to increase independence of successive observations . Based on these data , relative brood care , nestmate care , and foraging activity was calculated for each worker . Brood care activity was defined as the “sum of antennating , grooming , feeding , or carrying of eggs , larvae , or pupae , divided by the total number of observations , ” nestmate care activity as the “sum of antennating , grooming , feeding , or carrying of adult workers , divided by the total number of observations , ” and foraging activity as the “number of times being observed outside the nest , divided by the total number of observations , ” as ant workers mostly leave the nest to search for food . To test whether these three behaviors depended on age and/or demography treatment , 3 GLMMs were run , including relative “brood care activity , ” “nestmate care activity , ” or “foraging activity” as response variable and “age” ( young , old ) , “demography treatment” ( 1 age cohort only , both age cohorts ) , and their interaction as explanatory factors . “Colony ID” was added as a random factor because multiple workers per colony were scanned . We used a quasi-Poisson distribution and log linkage function to correct for overdispersion . A total of 48 full-body transcriptomes were collected from the “Age polyethism and behavioral flexibility” experiment . Here , we shortly summarize the methods and bioinformatic approaches; additional details are given in [34] . To identify genes exclusively associated with behavior , we sampled 24 brood carers and foragers each from both age cohorts and fertility levels . Samples were then sequenced on an Illumina HiSeq 2500 , and raw reads were assembled using CLC Workbench ( Qiagen ) followed by a MIRA meta-assembly . We then ran a generalized linear model in edgeR v3 . 4 ( Bioconductor ) to detect differently expressed genes associated with “behavior” ( brood carers versus forager ) , whereas age ( young versus old ) and fertility ( fertile versus infertile ) were added as blocking factors . cGMP-dependent protein kinase showed no caste- , age- , or fertility-dependent expression . For these experiments , a total of 80 additional colonies were collected in June 2015 at the E . N . Huyck Preserve , NY , United States of America and kept under the previously described laboratory conditions . To assess phenotypic changes after RNAi-mediated Vg-like A knockdown , 2 brood carers , 2 inside workers ( found inside the nest but not contributing to brood caring ) , 2 guards ( found in the nest entrance ) , and 2 foragers were individually labeled with colored metal wire . In addition , 2 recently hatched workers were labeled to investigate age-dependent effects of RNAi-mediated gene knockdown . Ant colonies were moved to 25 °C and starved for 5 days to increase hunger , resulting in a more efficient uptake of dsiRNA . Open reading frame of the Vg-like A contig was extracted using NCBI ORFfinder , and three 25-bp-long dsiRNA fragments targeting different regions of the open reading frame ( total contig length: 5511 bp; length of open reading frame: 4551 bp ) were designed and synthesized by IDT , USA ( S4 Table ) . Using multiple short fragments targeting the mRNA near the 5′ and 3′ end has been shown to increase knockdown efficiency [91 , 92] . As a control , we fed dsiRNA with no homologous region in the T . longispinosus transcriptome , though similar in length and nucleotide composition to the dsiRNA , which targets Vg-like A . For none of the 3 Vg-like A target sequences , we detected off-target binding in silico in the T . longispinosus transcriptome . These dsiRNA fragments also poorly align to all other Vg copies , except Vg-like A , which makes off-target effects unlikely ( S5 Table ) . Fifteen μl of sucrose solution ( 0 . 102 g sucrose per 1 ml nuclease-free water ) including 0 . 05 μg/μl per dsiRNA fragment were fed to the colony . As dsRNA remains stable in an ant’s crop for at least 24 hours [93] , feeding solutions were renewed every day . Previously published protocols on administering dsRNA via feeding suggested to feed dsRNA in a concentration of 2 μg/μl . However , given the size differences between the large workers of C . floridanus ( used in [93] ) and the much smaller T . longispinosus ( this study ) , the increased efficiency of the short dsiRNA compared to large dsRNA fragments , and to avoid overdosage , we decided to feed our dsiRNA in a concentration of 1 μg/μl . This concentration was additionally corrected for differences in fragment size ( [93]: 400 bp . This study: 25 bp ) , i . e . , divided by 20 , resulting in a final concentration of 1/20 = 0 . 05 μg/μl . Per treatment ( control , Vg-like A knockdown ) , 20 colonies were used . Nest scans were started after 3 days of feeding . Each colony was scanned 5 times per day for 4 days , i . e . , a total of 20 scans . As we were unable to ensure that each worker received the same dosage of dsiRNA , we scanned several workers per colony and behavioral caste . During each scan , the position and behavior of each individually labeled ant was recorded ( S3 Table ) . dsiRNA in sucrose solution was continuously fed during the observation days . To explore RNAi-induced changes in brood-caring behavior , a GLMM was run with “brood care frequency” ( definition: see first paragraph of the Material and methods section ) as response variable , “treatment” ( Vg-like A− , control ) and “caste” ( brood carer , in-nest worker , guard , forager , young worker ) as well as their interaction as explanatory variables , and “colony ID” as a random factor . The potential RNAi-mediated changes in “nestmate care , ” “foraging frequency , ” and “inactivity” ( defined as the number of observations during which an individual was inactive ) were tested with similar models . Feeding dsRNA to knock down a gene on a colony level can affect gene expression in larvae , suggesting that the dsRNA fragments are spread among colony members [93] . To test whether the observed changes in brood care could be explained by Vg-like A knockdown–induced alterations in larval behavior rather than by changes in worker behavior , a total of 65 brood care tests were conducted . Two brood carers per colony ( a total of 36 colonies , half of them Vg-like A− , the other half control ) were placed separately in a petri dish ( diameter: 2 cm ) containing a larva from a different colony , which was either part of the knockdown or control group before . The number and type of the interactions of the worker with the larva were recorded every 15 seconds for 5 minutes . This full factorial design allowed disentangling of the effect of worker and larval treatment on brood care behavior . For the statistical analysis , a GLMM was run including “brood care frequency” as response variable , “treatment of the worker” and “treatment of the larva” as well as their interaction as explanatory variable , and “colony ID of worker” and “larva” as random factors . A total of 32 brood carers ( 16 control and 16 Vg-like A− workers , all from different colonies ) were transferred to a petri dish ( diameter: 2 cm ) . Half of the petri dish was covered with black paint to create 2 distinct , equally sized darkened and lighted sections . The focal ant was placed in the center of the petri dish , and her position ( i . e . , which side of the petri dish ) was recorded for 1 hour every 2 minutes resulting in a total of 30 observations per individual . A Welsh two-sample t test was used to compare the preference for the lighted side of the petri dish ( number of observations that the individual was on this side ) between control and Vg-like A− workers . Twenty colonies per treatment ( collected in 2015 ) were fed with Vg-like A dsiRNA or nonsense dsiRNA every second day for a total of 33 days . Nest scans following the protocol described in the “Age polyethism and behavioral flexibility” section were conducted between day 30 and day 33 . To investigate effects of long-term knockdown of Vg-like A , a series of GLMMs were run , including either “frequency of brood care , ” “foraging , ” “inactivity , ” or “nestmate care” as response variables; “treatment” ( control , Vg-like A knockdown ) as explanatory variable; and “colony ID” as a random factor . The application of dsiRNA over 33 days did not result in increased mortality of workers ( Binomial GLM: χ2 = 1 . 8 , p = 0 . 180 ) , brood ( Binomial GLM: χ2 = 0 . 3 , p = 0 . 603 ) , and queens ( Binomial GLM: χ2 = 0 . 1 , p = 0 . 747 ) . After the last nest scan ( day 33 ) , 20 brood carers and foragers of control and Vg-like A− colonies—i . e . , a total of 80 workers—were individually frozen in glass vials for analysis of CHC profiles . CHCs were extracted and analyzed as described in [94] . Changes in the composition of the CHC profile were assessed using a PERMANOVA within the PRIMER software package ( Quest Research ) . After the long-term knockdown , all remaining queens and 1 individually labeled brood carer per colony were dissected , and length of all ovarioles was measured to the closest μm , using the Leica software , and the number of transparent and white eggs was counted . Apart from mean ovary length , we calculated “number of yolk-enriched eggs / ( number of yolk-enriched eggs + number of transparent eggs ) ” to assess fertility . We used 2 GLMMs , including either “mean ovary length” or the “yolk-enriched egg / transparent egg” ratio as response variable , “treatment” ( Control , Vg-like A knockdown ) and “caste” ( queen , worker ) and their interaction as explanatory factors , and “colony ID” as a random factor . In 29 colonies collected in July 2016 at the E . N . Huyck Preserve , NY , United States of America , 3 old ( i . e . , >1 year old ) and 3 recently hatched young brood carers each were labeled with a thin metal wire , and Vg-like A was knocked down for 30 days , following the protocols described in the “Vg-like A regulation of behavior and behavioral progession” section . Old brood carers were identified by actively conducting brood care shortly before we started feeding dsiRNA . After that , 2 inside workers and 2 larvae were removed from the colony and freeze-killed for 24 hours at −20 °C , and CHCs were extracted in hexane for 10 minutes . These 4 extracts were then entirely transferred to 1 filter paper ( diameter: 1 cm ) each and provided as CHC samples in 2 subsequent behavioral tests per colony . For these tests , 1 old and 1 young brood carer per colony were individually transferred to a petri dish ( diameter: 4 cm ) and offered CHC samples of a worker and a larva from the same colony . For 10 minutes , interactions with both CHC samples were recorded . We then ran a binomial GLMM including the “number of interactions with each sample” as response; “age of the tested individual” ( old , young ) , “treatment” ( control , Vg-like A knockdown ) , and their interaction as explanatory variables; and “colony ID” as a random factor . In each experiment , all behavioral data were collected by a single researcher . While recording the data , the observer was blind regarding ( i ) how the colony demography was manipulated , ( ii ) whether the colony was from the control or knockdown group , ( iii ) whether the individually labeled workers were young or old , and ( iv ) whether the individually labeled worker was classified as a brood carer or forager prior to the observation . All statistical analyses were ( if not specifically mentioned ) run in R , version 3 . 2 . 4 , loaded with the packages car , lme4 , and MASS , and GLMMs were modified using stepwise removal of nonsignificant interactions or main effects . Eight knockdown and control workers were individually homogenized in Trizol for qPCR analysis . RNA was extracted using RNeasy Mini Kit ( Qiagen ) , and cDNA was synthesized with QuantiTect Rev . Transcription Kit ( Qiagen ) . We then amplified Vg-like A as well as alpha tubulin ( AT ) and GAPDH as housekeeping genes ( HKGs , S6 Table ) to correct for different quantities of input material using SensiFAST SYBR Hi-ROX Kit ( Bioline ) . We used the RNAseq dataset that revealed Vg-like A as an interesting candidate gene [34] to show that the expression of AT and GAPDH is independent from whether a worker serves as a brood carer or forager ( AT: FDR = 0 . 333; GAPDH: FDR = 0 . 907 ) , whether the worker is young or old ( AT: FDR = 0 . 830; GAPDH: FDR = 0 . 578 ) , and whether the worker is fertile or infertile ( AT: FDR = 0 . 999; GAPDH: FDR = 0 . 999 ) . Moreover , the expression of Vg-like A was unlinked to the expression of AT ( Pearson: p = 0 . 228 ) and GAPDH ( Pearson: p = 0 . 259 ) , suggesting that a knockdown of Vg-like A does not result in expression changes in either of the 2 HKGs . The expression of AT and GAPDH was tightly correlated ( Pearson: p = 0 . 0004 ) . To further test whether AT and GAPDH are expressed similarly among brood carers and foragers , we used a total of 8 colonies of T . longispinosus , extracted RNA of 1 brood carer and 1 forager per colony , and constructed cDNA libraries . We quantified DNA content of each cDNA library and diluted each library to 0 . 0002 μg/μl and quantified AT and GAPDH using 10 μl of each diluted library , using qPCR . We ran 2 GLMMs to test whether the expression of AT or GAPDH depends on “behavioral caste” ( brood carer versus forager ) , using “colony ID” as a random factor: The expression of AT ( GLMM: p = 0 . 847 ) and GAPDH ( p = 0 . 861 ) was not different between both castes . Moreover , we used Bestkeeper [95] to show that both genes exhibit stable expression patterns ( AT: SD = 0 . 45; GAPDH: SD = 0 . 4 ) . We additionally analyzed our tissue-specific qPCR data using Bestkeeper to test whether both HKGs exhibit stable expression across tissues . AT was equally expressed in all tissues ( SD = 0 . 91 ) , and GAPDH showed a weak tissue-specific expression ( SD = 1 . 04 ) . qPCR results were then analyzed using ΔΔCT method and Wilcoxon test to compare relative expression of the targeted gene between treatment and control workers . Vg-like A–specific primers amplify a fragment near the spliced-off 3′ end , which increases the likelihood to detect an RNAi-induced down-regulation [92] . To assess tissue-specific expression , we labeled 5 brood carers in a total of 22 colonies . In 11 colonies , we knocked down Vg-like A for 7 days , following the protocol described in the “Vg-like A regulation of behavior and behavioral progession” section . The remaining 11 colonies were fed with nonsense dsiRNA . Five brood carers per colony were dissected , and ovaries , fat body , and brains , respectively , were pooled in Trizol to minimize the effect of individual variation . Samples were then processed and analyzed as described in the “Tissue-specific expression of Vg and Vg-like genes in T . longispinosus” section . To further investigate potential effects of Vg-like A dsiRNA application on the expression of other Vg copies , we also quantified VgC , MVg2 , MVg3 , Vg-like B , and Vg-like C , using qPCR ( S6 Table ) . Tissue-specific qPCR data were analyzed using the ΔΔCT method followed by 2 GLMMs per gene ( 1 for each HKG ) , including relative expression as a response variable; “tissue” ( brain , fat body , ovaries ) , “treatment” ( control , Vg-like A knockdown ) , and their interaction as explanatory factors; and “colony ID” as a random factor . To further assess tissue specificity in the expression of each gene , we ran 1 GLMM per gene on the control colonies , including relative expression of the respective gene as a response and “tissue” as an explanatory factor . “Colony ID” was added as a random factor . As there were no qualitative differences between both HKGs , only results for AT were shown ( results for GAPDH are summarized in S6 Fig ) . Vg orthologs have previously been grouped into Vgq ( associated with reproduction ) and Vgw ( associated with brood care ) [72] or Vg-like and conventional Vgs [52] . Our BLAST searches of the T . longispinosus Vg contigs were highly ambiguous . We therefore conducted BLAST searches of the Vg sequences published by Oxley and colleagues [50] and Morandin and colleagues [52] to identify Vg genes in all ant species with a currently available genome plus a number of social and nonsocial hymenopterans and other insects ( S1 Table ) and T . longispinosus contigs [34] . In several species , we detected “fused” Vg protein sequences—i . e . , the genome annotation identified a single Vg gene—although it actually contained 2 open reading frames . We split these fused proteins in their 2 according subparts . All previously published Vg sequences plus Vg sequences obtained by BLAST searches were used to construct an alignment using the web-based Clustal Omega tool ( https://www . ebi . ac . uk/Tools/msa/clustalo/ ) . A maximum-likelihood tree was constructed with RAxML ( 8 . 2 . 11 [75] ) , using the PROTGAMMAAUTO to automatically identify the best-fitting protein model ( JTT ) , and 1 , 000 bootstraps . For additional versions of the phylogenetic tree , including full species names and NCBI reference sequence ID system labels , see S7 and S8 Figs .
|
In social insects such as ants and bees , workers specialize in different tasks . This specialization is thought to be regulated via response thresholds to task-specific cues , which vary between workers conducting different tasks . Whether a worker takes care of the brood , cares for other workers , or leaves the nest to search for food is influenced by age , fat content , and the expression of associated genes . In the ant Temnothorax longispinosus , workers specializing in brood care are younger and exhibit a high expression of the gene Vg-like A . Here , we demonstrate that young workers reduce brood care activity upon down-regulation of Vg-like A . Simultaneously , they increase care for adult nestmates , a behavior typically exhibited by older workers . We show experimentally that Vg-like A down-regulation alters perception of social cues: a shift in the responsiveness from chemical cues from brood to adult worker underlies the behavioral switch to nestmate care . Hence , the expression of Vg-like A and its associated pathways influences task choice in ants and is involved in the regulation of division of labor . Copies of Vg-like A are present in other ants and further social and solitary insects , and future studies will reveal the role of this gene in these organisms .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] |
[
"invertebrates",
"honey",
"bees",
"sociology",
"social",
"sciences",
"animals",
"social",
"systems",
"animal",
"behavior",
"zoology",
"animal",
"sociality",
"bees",
"foraging",
"lipids",
"hymenoptera",
"ants",
"behavior",
"fats",
"gene",
"expression",
"insects",
"arthropoda",
"biochemistry",
"eukaryota",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2018
|
Vitellogenin-like A–associated shifts in social cue responsiveness regulate behavioral task specialization in an ant
|
Transmission of zoonotic cutaneous leishmaniasis ( ZCL ) depends on the presence , density and distribution of Leishmania major rodent reservoir and the development of these rodents is known to have a significant dependence on environmental and climate factors . ZCL in Tunisia is one of the most common forms of leishmaniasis . The aim of this paper was to build a regression model of ZCL cases to identify the relationship between ZCL occurrence and possible risk factors , and to develop a predicting model for ZCL's control and prevention purposes . Monthly reported ZCL cases , environmental and bioclimatic data were collected over 6 years ( 2009–2015 ) . Three rural areas in the governorate of Sidi Bouzid were selected as the study area . Cross-correlation analysis was used to identify the relevant lagged effects of possible risk factors , associated with ZCL cases . Non-parametric modeling techniques known as generalized additive model ( GAM ) and generalized additive mixed models ( GAMM ) were applied in this work . These techniques have the ability to approximate the relationship between the predictors ( inputs ) and the response variable ( output ) , and express the relationship mathematically . The goodness-of-fit of the constructed model was determined by Generalized cross-validation ( GCV ) score and residual test . There were a total of 1019 notified ZCL cases from July 2009 to June 2015 . The results showed seasonal distribution of reported ZCL cases from August to January . The model highlighted that rodent density , average temperature , cumulative rainfall and average relative humidity , with different time lags , all play role in sustaining and increasing the ZCL incidence . The GAMM model could be applied to predict the occurrence of ZCL in central Tunisia and could help for the establishment of an early warning system to control and prevent ZCL in central Tunisia .
Cutaneous leishmaniasis ( CL ) is a neglected tropical disease widespread in the Middle East , the Mediterranean basin and North Africa [1 , 2] . Most of the cases occur in the arid and semi-arid regions [3] . The circumstances of the dynamic leishmaniasis disease are often complex and varying according to environmental , demographic and human behavioral factors [4] . For cutaneous leishmaniasis , the parasite is transmitted by infected female sand fly vectors . Meanwhile rodents serve as the reservoir hosts [5–7] . The clinical form of cutaneous leishmaniasis occurring in humans , causes skin lesions and permanent scars , mainly on the face , arms and legs . In Tunisia , the incidence of cutaneous leishmaniasis climbed from only one endemic governorate in 1983 to 15 endemic governorates in 2014 [8] . In fact , data from 10 governorates show that in 2014 , 23% of the population is at risk . While a total of 57 591 cases were reported during the period from 1998 to 2007 [8] . Three clinical-epidemiological forms of cutaneous leishmaniasis identified in Tunisia , vary across regions [9] . In the North of Tunisia , Sparodic Cutaneous Leishmaniasis ( SCL ) , one of the three forms of CL , is induced by Leishmania infantum MON-24 . In central Tunisia , ZCL caused by Leishmania major MON-25 is considered dominant and distributed across arid zones [10] . The third form of CL: Chronic Cutaneous Leishmaniasis ( CCL ) , generated by Leishmania killicki MON-8 , was detected in the South-East of Tunisia [11] . According to Kallel et al . [12] , the original foci were spread ( for ZCL from the Center to the North and the South; for CCL from the South-East to the North; and for SCL from the North to the Center of the country ) . ZCL remains primarily a disease responsible for considerable morbidity and disfigurement [2] . The disease is endemic , essentially in the rural areas of southern and central regions of Tunisia where infrastructure is limited and sanitation is inadequate [13–15] . In these regions , the climate is favorable to the development of sand fly species , and consequently to the transmission of the disease; so the population is exposed while managing farm activities [16] . However , the association between climate factors and ZCL incidence was not clearly elucidated . The way climate factors influence transmission of the disease became the main research concern . In fact , recent studies mainly used time series analysis to assess the relationship between climatic ( or environmental ) factors and daily ( or monthly ) records of the disease . Diverse models were used , and the choice of a model may have large influence on quantifying climate effects . The most used model of time series analysis in the field of epidemiology , is the Autoregressive Integrated Moving Average ( ARIMA ) analysis [17 , 18] . This model consider the order of data points and adjacent points in time [19]; which allows an estimation of autocorrelation and trend . These models adjust for variation due to previous observations , trend over time and variation in the observation that cannot be predicted from previous observations [20 , 21] . Although ARIMA process can be a robust technique for improving the quality of prediction , it assesses linearly the relation between the response variable and the predictor covariates . However , recent studies have identified non-linear relations of climate factors on ZCL incidence . Failure to model relations correctly can lead to model misspecification that can affect the error structure of the model . Studying non-linear time series analysis is still limited compared to linear series . Generalized additive models ( GAM ) [22] have been used in environmental epidemiology [23–26] . GAMs are semi-parametric regression methods that relate the response variable to smoothed functions of potential explanatory variables via a link function [22] . Unlike parametric methods that impose the form of the trend to be used in the model , GAMs allow data to decide about this trend . In environmental epidemiological studies , the response variable may also be correlated . It is necessary to embody autocorrelation of the dependent variable when modeling . However , few works were interested in nonparametric regression when correlated observations are detected [27] . In this paper , we first aimed to model the relationship between the incidence of ZCL with the underlying predictor factors using the generalized additive models and its extension , the generalized additive mixed models considering the autocorrelation; and second , to predict the occurrence of the disease based on the best-fit model .
The whole project was approved by the ethical committee of Pasteur Institute in Tunis , Tunisia . But in this paper we presented the results of an ecological study with monthly number of ZCL cases and climate variables . Our study area covers three districts of Bir Badr , Hichria and Zefzef . These districts are selected all around the salt pan "Garaat Njila" located in the governorate of Sidi Bouzid , central Tunisia ( Fig 1 ) . Such areas are characterized by a semi-arid climate and a long-lasting emergency of the disease . In our work monthly ZCL records were used , from July 2009 to June 2015 . Data were obtained from an active system of epidemiologic surveillance , implemented in Sidi Bouzid , central Tunisia . All new cases of people seeking treatment at primary health care facilities and other cases notified among patients' neighbors and families by the active research of the nursing stuff were included in this surveillance . Also , all schools in this area have been asked to check and notify all ZCL cases among students . Moreover , members of the research team performed a community-based active ZCL surveillance , and notified cases in schools with a clinical suggestive form of cutaneous leishmaniasis were diagnosed by physicians and nurses from the health care facilities . Parasitologic diagnosis of ZCL lesions was carried out only for a group of patients using direct examination , skin culture , PCR TagMan and PCR high-resolution melting . No laboratory exam was undertaken by dint of the good knowledge of the disease by the medical staff , the population in this region , the high sensitivity and specificity of clinical diagnosis . The number of monthly ZCL notifications was accounted according to the date of the lesion onset for the period between July 2009 and June 2015 after being reported on a standardized sheet . Data were anonymized and for this study , we only count the number of cases monthly . There is no information about patients . The whole project was approved by the ethical committee of Pasteur Institute in Tunis , Tunisia , but in this paper we presented the results of an ecological study with monthly number of ZCL cases and climate variables . We used 2009–2014 counts for the development of the predictive models and set aside 2014–2015 counts for independent validation of the prediction . The bioclimatic variables used for this study were monthly data between July 2009 and June 2015 , recorded from a private station implemented in the study area . The variables collected were: average minimal temperature , average maximal temperature , average of averages temperatures , cumulative precipitation , average relative humidity , average wind speed , maximum wind speed , and average rodent density estimated according to their activity . Our study used real data of ZCL notifications , environmental and bioclimatic variables from July 2009 to June 2015 . We used a generalized additive model ( GAM ) and a generalized additive mixed model ( GAMM ) with natural cubic splines . These models were used to assess the relation between ZCL incidence and climate factors ( temperature , rainfall , relative humidity , wind speed and rodents' density ) . These covariates were included into the model as fixed effects . However , autoregressive terms were considered as random effects .
There were 861 notified ZCL cases over the study period , from July 2009 to June 2014 in the three districts of Sidi Bouzid , central Tunisia . The peak of ZCL mainly occurred from August to December during the same epidemiological year ( Fig 2 ) . Also , this result is stressed by the seasonality test which rejected the equality of months ( Kruskal-Wallis = 46 . 57 , df = 11 , p < 0 . 001 ) . In Fig 3 , a temporal variation of the ZCL incidence , is revealed during the whole study period , with an outbreak of 143 cases in October 2013 , during the fifth epidemic season . Large values on the incidence of the disease were also seen during the second and the third epidemic seasons , but very low values were recorded in the first and fourth epidemic seasons . Cross-correlation analysis ( Table 1 ) showed that several bioclimatic parameters were significantly associated with ZCL cases at various monthly lags . The most significant associations ( within all environmental and bioclimatic variables ) were for average temperature ( Tavg ) at 2- , 3- and 4-months lag , cumulative rainfall ( rainf ) lagged one month , relative humidity ( relative_humidity ) delayed 4 months , and rodent's density ( rodens ) at 2-months lag . The probability distribution of the ZCL incidence required for GAM model is slightly over-dispersed ( Mean = 14 . 15; Standard deviation = 24 . 91 ) . A Quasi-Poisson distribution fitted adequately the data . We began with building GAM models to estimate the pattern of each influential variable on ZCL incidence . Then , we used the generalized cross validation ( GCV ) score to compare the statistical performances of different models . The lowest GCV value yields to the best fit model . Results from the best-fit GAM model with a Quasi-Poisson distribution showed a significant associations between ZCL incidence and accumulated rainfall lagged 1 month , average temperature lagged 4 months , relative humidity with 4 months lag and rodent's density lagged 2 months ( lowest GCV = 2 . 23; deviance explained = 97 . 8% , dispersion parameter = 1 . 06 ≈ 1 ) ( Table 2 ) . So , the GAM model chosen to fit the data is given below: E ( ZCLi ) =μi=β0+f1 ( monthi ) +f2 ( Tavg4i ) +f3 ( relativehumidity4i ) +f4 ( pluv1i ) +f5 ( rodens2i ) where ZCLi is the ZCL cases in the ith observation . The terms f1 to f5 are the smoothing functions , and β0 is the intercept ( Table 2 ) . The estimated effects of environmental and climate variables on ZCL incidence are shown in Fig 4 , they revealed different patterns . All climate factors were statistically significant in a highly non-linear way . Fig 4A showed that during the same epidemiological season , months are characterized by a non-linear smooth with a positive effect for month 1 ( January ) , a negative effect for months 2–7 ( February to July ) , and an increase effect for months 8–12 ( August to December ) . The month's period of negative effect coincides with the beginning of warm season which would be a favorable period for phlebotomies' activities . Average temperature delayed 4 months ( Fig 4B ) had a wiggly association with ZCL incidence ( edf = 7 . 4 ) . A positive effect was seen for temperatures ranging between 5°C—12°C and between 15°C—20°C . A negative effect was noted for temperatures ranging between 12°C—15°C and over 20°C . Relative humidity with 4 months lag had a non-linear association ( edf = 4 . 3 ) with ZCL incidence ( Fig 4C ) . An increase effect for relative humidity was seen from 30 to 45% , and a negative effect over 45% . The association between rodent's density 2-months lag and ZCL was wiggly ( edf = 4 . 25 ) and is characterized by a general positive effect , except in the range of 20 to 30 , where the effect is negative ( Fig 4D ) . Monthly cumulative rainfall lagged 1 month had also a nonlinear association with ZCL incidence ( edf = 6 . 3 ) and was found to have an increase effect reaching a peak at 10 mm ( Fig 4E ) . Epidemic season was not significant in our model and had a non-significant linear effect on ZCL incidence . So , it was removed from the model . From Fig 5 , the ACF and PACF plots of GAM model showed all lags fell within ±0 . 2 confidence bands implying that the GAM model might be an appropriate model . However , autocorrelated observations are not considered in this model . Here , the autocorrelation part was added to the best-fit GAM model as a random effect using the generalized additive mixed model . From the GAM model , the pattern of the relation between bioclimatic factors and ZCL incidence was assessed . However , the correlation between the observations of the response variable was not evaluated . So , the autocorrelation function of ZCL cases was checked ( Fig 6 ) , and revealed significant dependences . At this stage , we retained the significant variables from the best fit GAM model , and used its extension , GAMM , to consider the autocorrelation of observations and whiten the errors . So , we included Autoregressive Moving Average ( ARMA ) processes in the error structure . According to the autocorrelation function and the partial autocorrelation function plots of the dependent variable , the error structure might be an AR ( 1 ) or ARMA ( 1 , 1 ) . Fig 7 showed that ARMA ( 1 , 1 ) process has residuals at low fitted values and could be retained as the best-fit GAMM model . The results of the prediction analysis , using the GAMM model and carried out using data from July 2014 to June 2015 , were drawn with the original values ( Fig 8 ) . Results showed that prediction from GAMM approach gives a good prediction accuracy . The Pearson correlation value between predicted and original values of number of cases was 0 . 81 ( IC = 0 . 46–0 . 94 , p < 0 . 001 ) . During the validation period , most monthly original values fell within the 95% confidence interval , presenting the obvious seasonal variation during months from September to December ( Table 3 ) .
Although cutaneous leishmaniasis caused by L . major is considered as one of the most important diseases in Tunisia , few studies have been conducted studying the complicated relationships between the transmission of the disease and climatic and environmental variables . Previous studies conducted in Tunisia in 2000 and 2009 [38 , 39] revealed a significant relationship between Mediterranean visceral leishmaniasis and climatic factors . In such relationships , it is often very difficult to find a suitable mathematical function for fitting the relationship [40] . In 2012 , Toumi et al . [41] used the autoregressive integrated moving average ( ARIMA ) models to demonstrate seasonality during the same epidemiologic year . They also applied Negative-Binomial generalized additive model ( GAM ) and generalized estimating equations ( GEE ) to quantify the relationship between temperature , rainfall , humidity and ZCL in central Tunisia . This study used monthly data from January 1991 to December 2007 , it did not include wind speed and rodent's density in their GAM and GEE models . They reported that only humidity and rainfall lagged 12–14 months were significant predictors of ZCL cases in Sidi Bouzid , and that average temperature was not statistically a significant predictor of ZCL incidence [41] . In our study , we focused on the ZCL cases from July 2009 to June 2015 and we estimated lagged effects of diverse bioclimatic and environmental variables , including minimum temperature , maximum temperature , average temperature , cumulative rainfall , relative humidity , average wind speed , maximum wind speed , and rodent's density , on the incidence of ZCL using monthly data . We estimated significant effects of monthly average temperature , cumulative rainfall , relative humidity and rodent's density on ZCL incidence after accounting for distributed lag effects . We found that cross-correlation between ZCL incidence and independent variables was not very high , although some were statistically significant . But , we retained the significant lagged variables in association with the incidence . In fact , the cycle of transmission of the disease is considered complex since the existence of three seasonality patterns: the first one stressed the importance of climate changes in the study region caused by the construction of dams and irrigation projects . Second , the climate variability may affect the density of rodents' reservoirs which is highly affected by the availability of chenopods , a plant exclusive food source of rodents , in the region [42] . So , transmission is better in warm season ( May to September ) . The last seasonality is characterized by the onset of the disease and its development that occurs in cold season ( October to May ) [43] . Then , we identified the relationship between ZCL and the selected environmental and bioclimatic factors in Sidi Bouzid , central Tunisia . To explain this association , a Quasi-Poisson GAM regression was chosen to be the model adopted that integrates parametric and non-parametric terms . GAM is specifically designed to analyze data when the impact of the predictors on the outcome is not necessarily linear . The results of our research stated the robustness and flexibility of GAM to reveal meaningful curvatures in exploratory analyses and the good quality of fitting . We found that the outbreak of ZCL was associated more with local environmental diversity than with climate factors . In fact , our results showed that rodent's density and weather variables could be used to predict ZCL transmission . To our best knowledge this is the first study incorporating rodent's density in the model , together with climate variables . Our GAM model showed that the significant effect of ZCL incidence was associated with monthly cumulative rainfall lagged 1 month . In fact , rainfall would increase the density of the halophytic plant , chenopods , that constitute the food of rodent's reservoirs [6 , 7] . Consequently the reservoir density increases , and affects the ZCL transmission . We also found that ZCL was associated with rodent density with 2 months lagged effect jointly with relative humidity lagged 4 months . Relative humidity has an influence on the survival of sandfly eggs and adults , the biting behavior of female adult sandfly , and laying of eggs . At the same temperature , egg hatchability of Phlebotomus papatasi increases as the relative humidity rises . The optimum relative humidity is 75% for saving eggs . When the humidity is too low , laying eggs will be affected , and adult mosquito mortality will increase . Increasing humidity will also facilitate feeding for the adult sandfly , enhancing its survival . The results of our GAM model showed that average temperature lagged 4 months has significant effect on ZCL occurrence . According to some studies [44 , 45] , temperature is one of the factors that determine the abundance of mosquitoes and the prevalence of mosquito-borne diseases . Despite the flexibility of our model that provided a better assumption of the nature of relationships between each bioclimatic factors and the number of ZCL cases , one limitation has to be pointed out . The short-term time series used in this work may not validate our predictive model . However , we need to extend the number of observations up to 10 years to validate a robust predictive model . Such recommendations are essential to improve the model and may help governors to detect earlier an outbreak occurrence and reduce , as well as possible , number of ZCL cases . A predictive model using spatiotemporal data needs to be the next goal to be accomplished in further studies towards the construction of an early warning system ( EWS ) of ZCL in Sidi Bouzid , central Tunisia . To this end a sustained surveillance and monitoring efforts of ZCL and climate factors are needed to provide time series sufficiently long for developing and evaluating forecasting models . To conclude , a complex relationship between environmental , bioclimatic factors and ZCL occurrence was found in central Tunisia . Additive models offer flexible modeling tools for regression problems . Understanding the role of the environmental and bioclimatic factors in ZCL occurrence can help to guide government policy-makers towards the creation and implementation of more effective policies to tackle the disease , and has important implications for prevention measures .
|
Zoonotic cutaneous leishmaniasis is a human vector-borne disease caused by the parasite Leishmania major and is well spread in rural areas where human resources in public health and infrastructure are limited . The cycle of transmission of the disease is complex because of the impact of climate change . In this study we evaluated the impact of bioclimatic factors on the transmission of the disease in three districts of Sidi Bouzid , central Tunisia . We found that the occurrence of zoonotic cutaneous leishmaniasis is mainly related to average temperature with 4 months lags , rodents' density lagged 2 months , relative humidity with 4 months lags and cumulative rainfall lagged 1 month . We also confirmed that our best-fit model predict well the occurrence of the disease .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"atmospheric",
"science",
"geographical",
"locations",
"tropical",
"diseases",
"vertebrates",
"parasitic",
"diseases",
"animals",
"mammals",
"seasons",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"humidity",
"africa",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"mathematical",
"and",
"statistical",
"techniques",
"protozoan",
"infections",
"people",
"and",
"places",
"rodents",
"tunisia",
"meteorology",
"earth",
"sciences",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"wind",
"amniotes",
"statistical",
"methods",
"organisms"
] |
2017
|
Modeling zoonotic cutaneous leishmaniasis incidence in central Tunisia from 2009-2015: Forecasting models using climate variables as predictors
|
Influenza viruses resistant to antiviral drugs emerge frequently . Not surprisingly , the widespread treatment in many countries of patients infected with 2009 pandemic influenza A ( H1N1 ) viruses with the neuraminidase ( NA ) inhibitors oseltamivir and zanamivir has led to the emergence of pandemic strains resistant to these drugs . Sporadic cases of pandemic influenza have been associated with mutant viruses possessing a histidine-to-tyrosine substitution at position 274 ( H274Y ) in the NA , a mutation known to be responsible for oseltamivir resistance . Here , we characterized in vitro and in vivo properties of two pairs of oseltaimivir-sensitive and -resistant ( possessing the NA H274Y substitution ) 2009 H1N1 pandemic viruses isolated in different parts of the world . An in vitro NA inhibition assay confirmed that the NA H274Y substitution confers oseltamivir resistance to 2009 H1N1 pandemic viruses . In mouse lungs , we found no significant difference in replication between oseltamivir-sensitive and -resistant viruses . In the lungs of mice treated with oseltamivir or even zanamivir , 2009 H1N1 pandemic viruses with the NA H274Y substitution replicated efficiently . Pathological analysis revealed that the pathogenicities of the oseltamivir-resistant viruses were comparable to those of their oseltamivir-sensitive counterparts in ferrets . Further , the oseltamivir-resistant viruses transmitted between ferrets as efficiently as their oseltamivir-sensitive counterparts . Collectively , these data indicate that oseltamivir-resistant 2009 H1N1 pandemic viruses with the NA H274Y substitution were comparable to their oseltamivir-sensitive counterparts in their pathogenicity and transmissibility in animal models . Our findings highlight the possibility that NA H274Y-possessing oseltamivir-resistant 2009 H1N1 pandemic viruses could supersede oseltamivir-sensitive viruses , as occurred with seasonal H1N1 viruses .
Since its emergence in early spring of 2009 , 2009 pandemic influenza A ( H1N1 ) viruses have been circulating worldwide [1] . Although most infected individuals have exhibited an uncomplicated , mild respiratory disorder , the pathogenicity of this virus is clearly higher than that of seasonal influenza viruses in animal models [2]–[4] and humans , including those who do not have underlying illness [5] . Although two classes of anti-influenza drugs – M2 ion channel blockers ( amino-adamantines; amantadine and rimantadine ) and neuraminidase ( NA ) inhibitors ( oseltamivir and zanamivir ) – are licensed , the 2009 H1N1 pandemic viruses , including the earliest isolate , are already amino-adamantine-resistant [1] . By contrast , most of the currently circulating pandemic viruses are susceptible to NA inhibitors [2] , [6] , and therefore , pandemic influenza patients are treated with NA inhibitors in many countries . This widespread administration , however , raises concerns over the emergence and global spread of NA inhibitor-resistant 2009 H1N1 pandemic viruses . Studies with seasonal H1N1 , H3N2 , and highly pathogenic avian H5N1 viruses revealed that single amino acid substitutions at several positions in or around the NA active site confer resistance to viruses against NA inhibitors [7]–[11] . Among these NA substitutions , a histidine-to-tyrosine substitution at position 274 ( N2 numbering , H274Y ) is one of the best characterized oseltamivir-resistant markers . Recently , the NA H274Y substitution has been detected in sporadic cases of oseltamivir-treated and -untreated patients infected with 2009 H1N1 pandemic viruses [12]–[14]: however , the pathogenicity and transmissibility of viruses possessing this NA substitution remain unknown . To better assess the risk of the 2009 H1N1 pandemic viruses that possess the NA H274Y substitution , we studied the in vitro and in vivo properties of two NA H274Y-possessing isolates that emerged independently: A/Osaka/180/2009 ( H1N1; O180r ) and A/Vietnam/HN32060/2009 ( H1N1; VN32060r ) .
To test the in vitro sensitivity to various NA inhibitors of 2009 H1N1 pandemic viruses possessing the NA H274Y substitution , we measured the NA activity of O180r and VN32060r exposed to oseltamivir carboxylate ( the active metabolite of oseltamivir ) , zanamivir , or R-125489 ( the active metabolite of the experimental NA inhibitor CS-8958 ) and determined IC50 values . These IC50 values were compared with those of A/Osaka/164/2009 ( O164s ) and A/Vietnam/HCM9727/2009 ( VN9727s ) , both of which do not possess the NA substitution ( Supplementary Table S1 ) and thus served as control strains ( Table 1 ) . O164s and O180r were isolated on similar dates , circulating in the same community and were genetically similar . Likewise , VN9727s and VN32060r were similar with respect to their isolation history ( temporally and geographically ) and their genetic properties . The IC50 values of O180r and VN32060r were about 500-fold higher than those of O164s and VN9727s for oseltamivir , while the sensitivity to zanamivir and R-125489 was comparable for all viruses tested . To further assess the in vitro sensitivity of oseltamivir-resistant 2009 H1N1 pandemic viruses to the NA inhibitors , we investigated the growth kinetics of the two pairs of oseltaimivir-sensitive and -resistant viruses in cultured cells in the presence of oseltamivir carboxylate , zanamivir , or R-125489 ( Supplementary Fig . S1 ) . While the growth of O164s and VN9727s was severely impaired by oseltamivir carboxylate , O180r and VN32060r replicated efficiently even in the presence of oseltamivir carboxylate . By contrast , both zanamivir and R-125489 inhibited the replication of all of the viruses tested . These results indicate that the NA H274Y substitution reduces the sensitivity of 2009 H1N1 pandemic viruses to oseltamivir , as has been shown with seasonal H1N1 , H3N2 , and highly pathogenic avian H5N1 viruses [7]–[11] , [15] , since the other amino acid variations among the viruses tested ( Supplementary Table S1 ) did not account for their reduced oseltamivir sensitivity: the lower growth of O180r compared to that of O164s in the absence of antivirals ( Supplementary Fig . S1 ) might be due to these additional amino acid differences . To evaluate the effect of the NA H274Y substitution on the replication of the 2009 pandemic viruses in vivo , we infected BALB/c mice with either 102 or 103 plaque-forming units ( PFU ) of O164s , VN9727s , O180r , or VN32060r and determined virus titers in the lungs of mice on days 3 and 6 post-infection ( pi ) ( Fig . 1 , see “Control” samples ) . We found that oseltamivir-resistant viruses replicated significantly better than oseltamivir-sensitive viruses at both day 3 and day 6 pi with one exception: on day 6 pi , the virus titer in the lungs of mice infected with 103 PFU of VN9727s was significantly higher than that of VN32060r-infected mice . These results indicate that the NA H274Y substitution does not negatively affect virus replication in the mouse lung in most settings . Next , we tested virus sensitivity to anti-influenza drugs in mice ( Fig . 1 ) . One hour pi , mice received oseltamivir , zanamivir , CS-8958 , or an experimental viral RNA polymerase inhibitor favipiravir ( also known as T-705 ) . When mice were treated with oseltamivir ( 80 mg/kg/day dose ) , the replication of the NA H274Y-possessing viruses O180r and VN32060r was not inhibited , whereas the virus titers of their oseltamivir-sensitive counterparts O164s and VN9727s were significantly reduced . These results support the results of the in vitro NA inhibition assays and growth kinetics in MDCK cells indicates that the NA H274Y substitution confers oseltamivir resistance to viruses in mice . Although the efficacy of zanamivir against an oseltamivir-resistant virus ( O180r ) was not prominent when mice were infected with 103 PFU of viruses ( Fig . 1 ) , at a lower infectious dose ( 102 PFU ) of viruses , its efficacy was apparent . CS-8958 and favipiravir significantly reduced the replication of all viruses tested , indicating that these 2009 H1N1 pandemic isolates , including oseltamivir-resistant strains , are as susceptible to these experimental antiviral drugs as one of the initial isolates A/California/04/2009 ( H1N1 ) [2] . To assess the transmissibility of oseltamivir-resistant 2009 H1N1 pandemic viruses , on day 1 pi we co-housed naïve ferrets in perforated cages next to ferrets inoculated with 106 PFU of O164s , VN9727s , O180r , or VN32060r and collected nasal swabs from inoculated and contact ferrets on days 1 , 3 , 5 , 7 , 9 , and 11 after infection and co-housing , respectively , for virological analysis ( Table 2 ) . For all viruses tested , all contact ferrets were infected with viruses via respiratory droplets from infected ferrets , although the detection of O180r in contact ferrets was delayed by up to two days in all three pairs . Sequencing of the NA gene of the oseltamivir-resistant viruses ( i . e . , O180r and VN32060r ) revealed that all of the viruses isolated from the contact ferrets retained the oseltamivir resistance-conferring tyrosine at position 274 in NA . Collectively , there were no substantial differences in transmissibility between oseltamivir-sensitive and -resistant viruses in ferrets . To evaluate the pathogenicity of NA H274Y-possessing oseltamivir-resistant 2009 H1N1 pandemic viruses , we infected ferrets with 106 PFU of O164s , VN9727s , O180r , or VN32060r and monitored them daily for changes in body temperature and weight , and for clinical signs . On day 6 pi , nasal turbinates , trachea , and lungs were removed for pathological analysis . All viruses tested caused respiratory symptoms such as sneezing , although no marked changes in body temperature or weight were observed ( data not shown ) . We found no remarkable pathological differences in the extent and manifestation of disease between ferrets infected with oseltamivir-sensitive or -resistant viruses ( Fig . 2 ) . All ferrets examined had severe inflammatory changes in their respiratory tracts: rhinitis , tracheitis and bronchopneumonia with bronchadenitis ( Fig . 2A and B ) . In addition , we found limited bronchadenitis lesions with viral antigens in four ferrets , two of which were infected with oseltamivir-sensitive viruses ( either O164s or VN9727s ) while the other two were infected with oseltamivir-resistant VN32060r ( Fig . 2C and D ) . These results indicate the comparable pathogenicity of the oseltamivir-resistant viruses to their oseltamivir-sensitive counterparts in ferrets . These results also suggest that the oseltamivir-resistant 2009 H1N1 pandemic viruses have the potential to be readily transmitted among humans without loss of pathogenicity . Here , we investigated antiviral sensitivity , pathogenicity , and transmissibility of 2009 H1N1 pandemic influenza viruses possessing the substitution NA H274Y , which is known to reduce the sensitivity of influenza viruses to oseltamivir in cultured cells , mice , and ferrets . Viruses with this NA substitution exhibited reduced sensitivity to oseltamivir with comparable replication , pathogenicity , and transmissibility compared to their oseltamivir-sensitive counterparts: these findings need to be validated in humans . The widespread administration of oseltamivir , and to a lesser extent zanamivir , will clearly contribute to the emergence of NA inhibitor-resistant viruses that retain optimal replication fitness and transmissibility in humans . Given our finding that the oseltamivir-resistant 2009 H1N1 pandemic viruses are as pathogenic and transmissible as their drug-sensitive counterparts , specific criteria for the use of NA inhibitors may need to be reconsidered . Oseltamivir has been used extensively in patients with 2009 pandemic influenza , especially those who are immunocompromised . Approximately one-third of the oseltamivir-resistant pandemic strains have been isolated from immunocompromised patients ( http://www . who . int/csr/disease/swineflu/frequently_asked_questions/antivirals/resistance/en/index . html ) , a possibility we predicted early in this pandemic [16] . Oseltamivir-resistant viruses readily emerge in immunocompromised patients partly because viruses can replicate to higher titers and for longer periods in immunocompromised hosts compared to those who are immunocompetent hosts [17] . Since the oseltamivir-resistant 2009 pandemic viruses are as pathogenic and transmissible as their drug-sensitive parents , immunocompromised influenza patients require special care - increased drug doses , combination therapy , and possibly isolation . In particular , combination therapy has the potential to reduce the emergence of drug-resistant mutants [18] and provide synergistic effects on the inhibition of virus replication [19]–[21] with no increase in adverse events [22] . Adopting such measures would help prevent the situation experienced with the oseltamivir-resistant seasonal H1N1 virus , which emerged during the 2007-2008 influenza season in Europe [23] , spread , and superseded previously circulating oseltamivir-sensitive viruses [24] , [25] . Finally , our in vitro and in vivo experiments suggest that the experimental drugs CS-8958 and favipiravir may be promising candidates to combat 2009 H1N1 pandemic viruses .
We used two pairs of H1N1 pandemic virus clinical isolates; A/Osaka/164/2009 ( O164s ) , A/Osaka/180/2009 ( O180r ) , A/Vietnam/HCM9727/2009 ( VN9727s ) , and A/Vietnam/HN32060/2009 ( VN32060r ) . All of the viruses were isolated in Madin-Darby canine kidney ( MDCK ) cells , which were maintained in Eagle’s minimal essential medium ( MEM ) supplemented with 5% newborn calf serum ( Sigma , St . Louis , MO ) and cultured at 35°C in 5% CO2 . Tyrosine at position 275 in the NA of O180r and VN32060r was confirmed by molecular cloning and sequencing of 18 individual clones each , while the rest of the viral genome sequences were determined by direct sequencing of amplified DNA . Oseltamivir ( oseltamivir phosphate ) and oseltamivir carboxylate ( the active metabolite of oseltamivir; GS4104 ) were prepared from Tamiflu ( Roche Laboratories Inc . , Basel , Switzerland ) . Zanamivir , CS-8958 ( an experimental NA inhibitor ) , and R-125489 ( the active metabolite of CS-8958 ) were synthesized at Daiichi Sankyo Co . Ltd . according to published procedures [26] . Favipiravir ( an experimental broad-spectrum viral RNA polymerase inhibitor , also known as T-705 ) , was synthesized at Toyama Chemical Co . , Ltd . NA activity of viruses in the presence of NA inhibitor was measured by an NA inhibition assay as described previously [8] , [27] , [28] . Briefly , diluted viruses were mixed with various concentrations of NA inhibitor in 2-[N-morpholino]ethanesulfonic acid containing CaCl2 and incubated for 30 min at 37°C . A fluorescent substrate methylumbelliferyl-N-acetylneuraminic acid ( Sigma ) was added to this mixture , which was then incubated for one hour at 37°C . After adding NaOH in 80% ethanol to the mixture to stop the reaction , the fluorescence of the solution was measured at an excitation wavelength of 360 nm and an emission wavelength of 465 nm and the 50% inhibitory concentration ( IC50 ) was calculated . MDCK cells were infected with viruses at a multiplicity of infection of 0 . 0001 . One hour later , oseltamivir carboxylate , zanamivir , R-125489 ( 10 µM each ) , or nothing ( control ) was added to the cells . At 8 , 16 , 24 , 32 , 40 , and 48 h post-infection ( pi ) , the supernatants of the infected cells were harvested and subjected to plaque assays in MDCK cells to determine virus titers . Six-week-old female BALB/c mice ( Japan SLC Inc . , Shizuoka , Japan ) were anesthetized with isoflurane and intranasally infected with 102 or 103 plaque-forming units ( PFU ) ( 50 µl ) of viruses . One hour pi , six mice per group were administered the following antiviral compounds: ( 1 ) oseltamivir phosphate: 80 mg per kg per 400 µl ( divided into two oral administrations per day ) for 5 days; ( 2 ) zanamivir: 0 . 8 , 8 , or 25 mg per kg per 50 µl in one daily intranasal administration for 5 days; ( 3 ) CS-8958: 0 . 75 mg per kg per 50 µl in one intranasal administration; ( 4 ) Favipiravir: 60 or 300 mg per kg per 400 µl ( divided into two oral administrations per day ) for 5 days; ( 5 ) or distilled water orally ( 200 µl ) and PBS intranasally ( 50 µl ) . Three mice per group were euthanized on days 3 and 6 pi and virus titers in the lungs were determined by plaque assays in MDCK cells . Three-to-four-month-old male ferrets ( Marshall Farms , Wolcott , NY ) and six-to-eight-month-old female ferrets ( Triple F Farms , Sayre , PA ) were used; animals from Triple F Farms were used for the transmission study of O164s and O180r , while those from Marshall Farms were used for the rest of studies . All of the animals , which were serologically negative for currently circulating human influenza viruses ( including pandemic H1N1 viruses ) by haemagglutination inhibition ( HI ) assay , were anesthetized with ketamine and xylazine ( 5 mg and 0 . 5 mg per kg of body weight , respectively ) and intranasally infected with 106 PFU ( 500 µl ) of viruses . On day 6 pi , three ferrets per group were euthanized and their nasal turbinates , trachea , and lungs subjected to virological and pathological analyses . For the transmission study , on day 1 pi , three naïve ferrets per group were co-housed in a perforated cage adjacent to an inoculated ferret; ferrets were separated by a perforated divider and did not have direct contact . All ferrets were monitored daily for changes in body temperature and weight , and clinical signs . Nasal swabs of inoculated and contact ferrets were collected on days 1 , 3 , 5 , 7 , 9 , and 11 after infection and co-housing , respectively , and virus titers were determined by plaque assays in MDCK cells . Nasal turbinates , lungs , and trachea of euthanized ferrets were preserved in 10% phosphate-buffered formalin . Tissues were then processed for paraffin embedding and cut into 5 µm-thick sections . One section from each tissue sample was subjected to standard hematoxylin-and-eosin ( H&E ) staining , while another was processed for immunohistological staining with an anti-influenza virus rabbit antibody ( R309; prepared in our laboratory ) that reacts comparably with all viruses tested . Specific antigen-antibody reactions were visualized by 3 , 3'-diaminobenzidine tetrahydrochloride staining using a Dako EnVision system ( Dako Co . Ltd . , Tokyo , Japan ) . Our research protocol for the use of mice and ferrets followed the University of Tokyo's Regulations for Animal Care and Use , which was approved by the Animal Experiment Committee of the Institute of Medical Science , the University of Tokyo ( approval number: 19–28 ) . The committee acknowledged and accepted both the legal and ethical responsibility for the animals , as specified in the Fundamental Guidelines for Proper Conduct of Animal Experiment and Related Activities in Academic Research Institutions under the jurisdiction of the Ministry of Education , Culture , Sports , Science and Technology , 2006 .
|
Although most of the currently circulating 2009 pandemic influenza A ( H1N1 ) viruses are susceptible to neuraminidase ( NA ) inhibitors ( oseltamivir and zanamivir ) , oseltamivir-resistant mutants have sporadically appeared . Yet , the pathogenicity and transmissibility of these oseltamivir-resistant 2009 H1N1 pandemic viruses remain unknown . Here , we compared the pathogenicity and transmissibility of two pairs of oseltamivir-sensitive and -resistant viruses in mouse and ferret models . We found that the oseltamivir-resistant viruses efficiently replicated in the lungs of mice treated with oseltamivir or even zanamivir . Further , we demonstrated that these oseltamivir-resistant viruses caused lung lesions in ferrets and efficiently transmitted between ferrets , as did their oseltamivir-sensitive counterparts . Overall , our results suggest the possibility that oseltamivir-resistant viruses could spread among humans without loss of pathogenicity .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/mechanisms",
"of",
"resistance",
"and",
"susceptibility,",
"including",
"host",
"genetics",
"virology/new",
"therapies,",
"including",
"antivirals",
"and",
"immunotherapy",
"virology/animal",
"models",
"of",
"infection",
"virology/antivirals,",
"including",
"modes",
"of",
"action",
"and",
"resistance"
] |
2010
|
Characterization of Oseltamivir-Resistant 2009 H1N1 Pandemic Influenza A Viruses
|
Heterosis has been widely used in agriculture , but the molecular mechanism for this remains largely elusive . In Arabidopsis hybrids and allopolyploids , increased photosynthetic and metabolic activities are linked to altered expression of circadian clock regulators , including CIRCADIAN CLOCK ASSOCIATED1 ( CCA1 ) . It is unknown whether a similar mechanism mediates heterosis in maize hybrids . Here we report that higher levels of carbon fixation and starch accumulation in the maize hybrids are associated with altered temporal gene expression . Two maize CCA1 homologs , ZmCCA1a and ZmCCA1b , are diurnally up-regulated in the hybrids . Expressing ZmCCA1 complements the cca1 mutant phenotype in Arabidopsis , and overexpressing ZmCCA1b disrupts circadian rhythms and biomass heterosis . Furthermore , overexpressing ZmCCA1b in maize reduced chlorophyll content and plant height . Reduced height stems from reduced node elongation but not total node number in both greenhouse and field conditions . Phenotypes are less severe in the field than in the greenhouse , suggesting that enhanced light and/or metabolic activities in the field can compensate for altered circadian regulation in growth vigor . Chromatin immunoprecipitation-sequencing ( ChIP-seq ) analysis reveals a temporal shift of ZmCCA1-binding targets to the early morning in the hybrids , suggesting that activation of morning-phased genes in the hybrids promotes photosynthesis and growth vigor . This temporal shift of ZmCCA1-binding targets correlated with nonadditive and additive gene expression in early and late stages of seedling development . These results could guide breeding better hybrid crops to meet the growing demand in food and bioenergy .
A filial F1 hybrid often outperforms inbred parents in growth and fitness , a phenomenon known as heterosis or hybrid vigor . Since early twentieth century , heterosis has been applied in breeding and production of maize and many other crops , vegetables , and some animals [1–3] . However , the molecular bases for heterosis are poorly understood . Traditional “dominance” and “overdominance” models cannot explain the complexity of heterosis . It is suggested that these allelic models should be updated to account for gene expression and regulatory networks that are often altered in hybrids relative to their parents [3–5] . Notably , gene expression changes are dynamic in crop hybrids . For example , maize F1 hybrids exhibit additive expression , non-additive expression [6 , 7] , and all possible modes of gene expression [8] . These different observations of expression changes could be associated with a few major regulatory genes that cause a cascade of effects on downstream metabolic and physiological pathways , leading to heterosis [5] . Consistent with this concept , recent studies have discovered a link between altered circadian clock regulation and increased levels of photosynthetic and metabolic activities and biomass in Arabidopsis allotetraploids ( interspecific hybrids ) [9] and Arabidopsis thaliana intraspecific hybrids [10 , 11] . These results collectively indicate that the expression of central circadian clock genes is epigenetically altered in hybrids , which in turn increases expression levels of downstream genes involved in energy and metabolic pathways , promoting carbohydrate metabolism during the day and night . The more starch accumulates during the day; the more starch can be degraded at night to stimulate growth , leading to biomass heterosis [5] . In A . thaliana , the circadian clock consists of central oscillator components CIRCADIAN CLOCK-ASSOCIATED 1 ( CCA1 ) and LATE ELONGATED HYPOCOTYL ( LHY ) and their reciprocal regulator TIMING OF CAB EXPRESSION 1 ( TOC1 ) , also known as PSEUDORESPONSE REGULATOR1 ( PRR1 ) [12 , 13] . When the endogenous clock matches the external diurnal cycle , CO2 fixation , photosynthetic activities , and fitness are increased [14] . Disrupting clock function causes decreased growth and fitness . For example , a CCA1 overexpression mutant ( CCA1-OX ) lacks circadian rhythms and displays reduced photosynthesis and fitness [15] . The double mutant cca1 lhy accumulates less starch and is unable to properly set the rate of starch degradation to match the length of night [16] . An important mechanism by which the clock improve fitness is temporal regulation of energetically costly activities , also known as gating [17] . Gating is apparent for pathogen responses , temperature responses , growth control , shade avoidance , and phytohormone signaling [18–25] . CCA1-OX plants are arrhythmic and largely lack these behaviors . Maize performs C4 photosynthesis , which is anatomically and biochemically distinct from C3 photosynthesis in plants like Arabidopsis [26] . Photosynthetic activities in maize are subject to diurnal regulation [27] . Sucrose accumulation rates increase during the day , reach a high level at 15:00 hours , and continue to increase until dusk . Starch mobilization occurs beginning at dusk , and all of the starch is depleted by the end of the night . Mobilization of starch at night likely promotes growth , which , after temperature correction , is greatest at night [28] . This diurnal regulation of carbohydrate metabolism is consistent with diurnal gene expression in maize leaves . Approximately 10% of ~13 , 000 transcripts examined display circadian expression patterns [29] . The majority of cycling genes peak their expression at subjective dawn and dusk , similar to other plant circadian systems . In another study , 23% of expressed transcripts exhibit a diurnal cycling pattern in leaves [30] . By contrast , in developing ears only core circadian clock genes and a handful of other transcripts are diurnally regulated . This suggests tissue-specific circadian regulation in maize , as observed in Arabidopsis [10 , 31] . The circadian effect on biomass heterosis is established early during embryo development [10] . Furthermore , developmental shifts in gene expression can also influence biomass accumulation . In sorghum , SbPRR37 activates expression of several downstream genes that repress flowering in long days [32] . SbPRR37 expression is dependent on light and regulated by the circadian clock . In short days , SbPRR37 is not expressed during the evening-phase , allowing sorghum to flower . This suggests an integration of circadian clock with flowering time that alters biomass production . Quantitative trait loci ( QTLs ) for key agronomic traits in crop plants have been associated with likely clock-associated genes [33] . For example , yield-related QTLs are mapped to clock-related and light signaling genes in the super-hybrid rice [34] . Flowering QTLs are associated with circadian clock regulators , GIGANTEA in soybean [35] and CONSTANS in sorghum [36] . A recent study finds that the allelic variation of clock loci in tomato varieties is linked to their domestication [37] . In maize , agronomic traits are controlled by many small-effect QTLs [38] , and several clock regulator homologs are identified as a priori candidate genes among genome-wide association QTLs [39] . These studies support the notion that the circadian clock mediates flowering , growth , and heterosis in crop plants including maize as in Arabidopsis [40] . However , the effect of the circadian clock on biomass production in maize hybrids and their inbred lines is largely unknown . Here we examined biomass accumulation in maize hybrids and their inbred parents , and investigated the function of maize CCA1 homologs ( ZmCCA1a and ZmCCA1b ) . Cloned ZmCCA1 coding sequences were used to test their functions in both Arabidopsis and maize . Recombinant ZmCCA1 proteins were used to study their binding affinity to promoter fragments in vitro , and polyclonal antibodies against ZmCCA1 were used to investigate ZmCCA1-binding targets and genes in planta . The results indicate altered temporal binding activities of ZmCCA1s in hybrids relative to inbred parents , associated with increased expression levels of carbon fixation genes , as well as increased rates of carbon fixation and biomass accumulation . The data suggest a molecular mechanism that reprograms circadian-regulatory networks in maize hybrids , promoting growth vigor .
Seedling growth vigor was reported to correlate with larger seeds in F1 hybrids [41] . This hypothesis was not supported by seed size analysis: seed weight was similar in maize inbred lines ( B73 and Mo17 ) and their reciprocal F1 hybrids ( B73XMo17 , BM , and Mo17XB73 , MB ) ( by convention , the female parent is listed first in a genetic cross ) ( Fig 1A ) . In spite of similar seed size , seedling growth heterosis occurred soon after germination ( Fig 1B ) , and leaf areas were significantly greater in the F1 hybrids than in the parents ( Fig 1C ) . F1 hybrids showed biomass heterosis 5 days after planting ( 5 DAP ) ( Fig 1D ) , and leaf length of the hybrids was significantly greater than the best parent from the emergence of the first true leaf 5 DAP until 14 DAP ( Fig 1E ) . Biomass and plant height heterosis was significantly different between the reciprocal crosses , and vigor was higher when Mo17 was used as the maternal parent . However , this difference in height decreased and was no longer obvious after two weeks ( Fig 1D and S1 Fig ) . These data suggest early establishment of seedling growth heterosis and subsequent maintenance of this heterosis during seedling development . Seedling growth rates were associated with increased levels of photosynthetic activities including CO2 assimilation , which is under diurnal regulation as observed in maize inbred lines [27] . Positive net CO2 assimilation increased from zeitgeber time 0 ( ZT0 = dawn ) , peaked at ZT8 , declined through the remaining light period , and ceased at ZT14 ( Fig 2A ) . Notably , the amplitude of CO2 assimilation significantly increased in both hybrids relative to the mid-parent value ( MPV ) , especially during the morning phases . This increase in photosynthetic rate of the hybrids should be related to the expected effect of greater leaf area on whole-plant carbon gain relative to the inbreds . The increased photosynthetic activities promote starch synthesis and mobilization . In a maize leaf , the middle-tip region represents a source where carbohydrates are synthesized , whereas the base region represents a sink where carbohydrates are stored , utilized in growth , and mobilized [26 , 42] . In the source region , starch amount was significantly more in the hybrids than the MPV , but the sucrose content was similar ( Fig 2B ) . In the sink region , both starch and sucrose accumulated significantly more in the hybrids than the MPV . Little photosynthetic activity is present in the sink region of a maize leaf [43]; thus starch accumulation is likely caused by the increased CO2 assimilation in the source region during the day , with carbohydrates being mobilized to the sink region at night . More sucrose synthesis and mobilization in the hybrids provides more carbohydrates for seedling growth , suggesting a role for diurnal regulation of carbon metabolism in growth heterosis . Photosynthetic activities and starch metabolism are diurnally regulated in maize [27 , 28] as in Arabidopsis diploids , hybrids and allopolyploids [5] , suggesting a role for the circadian clock in promoting seedling growth in maize . To test this , we investigated CCA1 homologs and their functions in maize . Being an ancient tetraploid [44 , 45] , maize has paralogous duplicates of many clock gene homologs [30] . The two homologs , designated ZmCCA1a ( ZM2G474769 ) and ZmCCA1b ( ZM2G014902 ) and located on chromosomes 10 and 4 , respectively [30] , are closely related to CCA1 and LHY in Arabidopsis ( S2A Fig ) . In the maize gene model ( B73 RefGen_v2 ) , ZmCCA1a is a truncated version; in the ~20-kb upstream , there is another homologous gene ( ZM2G175265 ) that encodes a MYB DNA-binding domain protein . The short transcript ( ZM2G474769 , ZmCCA1a ) could be a splicing variant and was used for expression and other analyses in this study . In Arabidopsis , CCA1 lacking the MYB DNA-binding domain acts as a dominant negative factor to fine-tune the period length [46] . The two maize genes exhibit diurnal expression patterns , although ZmCCA1b was expressed at higher levels than ZmCCA1a [30] , in most of 17 maize tissues examined [47] ( S2C Fig ) , indicating that ZmCCA1b may play a larger role than this truncated ZmCCA1a . Expression of ZmCCA1a and ZmCCA1b was diurnally regulated and peaked at ZT3 ( Fig 3A ) , whereas GIGANTEA1 ( gi1 ) , ZmPRR59 , and ZmTOC1a [30 , 33] , peaked at ZT12 ( S6A–S6C Fig ) . Interestingly , all clock genes analyzed were nonadditively expressed ( deviated from MPV ) in the hybrids at time-points that coincided with nonadditive photosynthetic and metabolic activities during the day ( Fig 2 ) . In the hybrids , both ZmCCA1a and ZmCCA1b were up-regulated in the morning phase , while gi1 and ZmPRR59 , a PRR5 homolog in Arabidopsis , were up-regulated at ZT9 . However , ZmTOC1a was down-regulated in the middle of the day and at night . These data indicate altered expression of circadian clock genes in the maize hybrids as in Arabidopsis allotetraploids [9] and intraspecific hybrids [10 , 48] . Altered CCA1 expression may affect abundance of CCA1 protein that binds to the evening element ( EE ) or CCA1-binding site ( CBS ) in the promoters of other clock and output genes [9 , 49] . Electrophoretic mobility shift assays ( EMSA ) showed that ZmCCA1b but not ZmCCA1a ( excluding the MYB DNA-binding domain ) bound competitively and specifically to the EE and CBS elements of ZmTOC1a , gi1 and ZmPRR59 promoter fragments in vitro ( Fig 3C ) . The binding signal depended on EE or CBS elements as well as the recombinant ZmCCA1b ( rZmCCA1b ) protein concentration . When the EE or CBS was mutated ( M ) or no EE or CBS ( N ) was present in the endogenous promoter fragments , the binding activity of rZmCCA1b was undetectable . As a control , maltose-binding protein ( MBP ) showed no binding activities . Like the control , rZmCCA1a showed no binding activity . The absence of the MYB DNA-binding domain in truncated ZmCCA1a likely explains undetectable target binding . These data suggest that ZmCCA1b is a functional protein that binds to the promoters of circadian clock genes via cis-acting EE and CBS elements . To test the function of maize CCA1 homologs , we generated Arabidopsis CCA1 , maize ZmCCA1a or ZmCCA1b overexpression ( OX ) lines in the transgenic A . thaliana ( Col-0 ) plants that also express a luciferase ( LUC ) reporter driven by the promoter of CHLOROPHYLL A/B BINDING PROTEIN 2 ( CAB2 ) ( CAB2:LUC ) . These lines designated CCA1-OX , ZmCCA1a-OX , and ZmCCA1b-OX , respectively ( Fig 3D ) . While the wild-type and transgenic control ( Vec-1 ) lines showed rhythmic CAB2:LUC expression , overexpressing CCA1 abolished CAB2:LUC expression rhythmicity under constant light ( S3A Fig ) , as previously reported ( Wang and Tobin , 1998 Cell ) . Interestingly , in the ZmCCA1b-OX lines the CAB2:LUC expression rhythmicity was dampened , and the timing of peak expression delayed under the constant light . Overexpressing ZmCCA1a in the ZmCCA1a-OX lines had a smaller effect on CAB2:LUC expression rhythmicity , and the expression peak was lower than that in the ZmCCA1b-OX lines ( Fig 3D and S3A Fig ) . The weaker ZmCCA1a-OX phenotype may be associated with the lack of binding activity in the truncated ZmCCA1a to the EE or CBS of gene promoters ( Fig 3C ) . As a result of disrupting the clock gene functions , aerial biomass of CCA1-OX and ZmCCA1b-OX lines was significantly lower than the wild-type or the transgenic control ( Fig 3E ) , consistent with the growth disadvantage in clock gene mutants and overexpression lines [14 , 15] . Biomass in the ZmCCA1a-OX lines was not significantly different from the controls , suggesting that a full-length clock gene is required for growth vigor . Expressing ZmCCA1a or ZmCCA1b under the Arabidopsis CCA1 promoter partially rescued the early period phenotype of CAB2:LUC in the cca1-11 mutant ( S3B and S3C Fig ) . Under constant light , the period of CAB2:LUC rhythms was shorter in the cca1-11 mutant ( 23 . 9 ± 0 . 1 h ) than in the wild-type ( WS ) ( 25 . 4 ± 0 . 3 h ) . Expressing Arabidopsis CCA1 in the cca1-11 mutant , as a control , fully rescued the short period phenotype of CAB2:LUC ( 25 . 4 ± 0 . 1 h ) . Expressing CCA1:ZmCCA1b ( 24 . 8 ± 0 . 1 h ) or CCA1:ZmCCA1a ( 24 . 2 ± 0 . 1 h ) in the cca1 mutant also lengthened the period relative to cca1-11 ( 23 . 9 ± 0 . 1 h ) . These data suggest that the circadian clock function between Arabidopsis and maize CCA1 homologs is conserved . ZmCCA1b complemented the mutant phenotype , while ZmCCA1a did not , suggesting that DNA binding activities conferred by the MYB DNA-binding domain are required for the clock function . To test if ZmCCA1b regulates growth vigor in maize , we expressed ZmCCA1b driven by the constitutive UBIQUITIN promoter-intron cassette ( UBI ) [50] in maize B104 line . Multiple independent transgenic lines were generated; two lines , designated OX1-3 and OX10-1 , were analyzed in both greenhouse and field conditions ( Fig 4 ) . These lines were maintained as heterozygotes to prevent transgene silencing and were subsequently genotyped to identify the lines carrying the transgene for analysis . In the greenhouse , plant height was dramatically reduced in the OX lines ( Fig 4A ) . The OX1-3 line exhibited a more severe phenotype than the OX10-1 line ( Fig 4B ) . Reduction in plant height was associated with reduced lengths of early nodes ( from #3 to #10 ) , while the total node number was unaffected ( Fig 4C ) . The OX1-3 line showed more severe reduction in plant height than the OX10-1 line , which is consistent with higher ZmCCA1b expression levels in the former than in the latter ( Fig 4D ) . Moreover , ZmCCA1 were expressed at higher levels at ZT3 than at ZT15 , suggesting oscillation of ZmCCA1 transcripts in the overexpression transgenic plants . In addition , the amount of total chlorophyll and chlorophyll a and b was reduced in the OX1-3 line in both greenhouse and field conditions , with more severe reduction in the greenhouse than in the field . In the field , the OX1-3 line also showed reduced height ( Fig 4F ) , but the reduction in node length occurred in the later nodes ( #7 to #16 ) ( compare Fig 4G with 4C ) . The phenotype was less severe in the field than in the greenhouse . A major difference between the greenhouse and field conditions is the light intensity , suggesting a role of light compensation for altered circadian regulation in growth vigor . Alternatively , enhanced metabolic activities such as increased endogenous sugar levels can provide metabolic feedback to the circadian oscillator , as observed in Arabidopsis [51] . These predictions remain to be investigated . Together , these data suggest that disrupting circadian regulation reduces chlorophyll content and growth vigor in maize as in Arabidopsis ( Fig 3E ) [9 , 14] . These data indicate that ZmCCA1s , like their homologs in Arabidopsis , act as central clock regulators in maize . We predicted that altered ZmCCA1 expression in the F1 hybrids plays a role in maize heterosis . To test how ZmCCA1s affect expression of output genes and growth vigor in maize , we raised antibodies against the N-terminus MYB DNA-binding domain ( S2B Fig ) , which could recognize ZmCCA1b and a full-length ZmCCA1a if the MYB DNA-binding domain is included . Western blot analysis indicated that accumulation of ZmCCA1s reached high levels in the morning , peaking at ZT3 , and decreased during the day ( S4A Fig ) . Using the antibodies , we tested genome-wide ZmCCA1-binding profiles in the F1 hybrids and their parents at ZT3 , ZT9 and ZT15 using chromatin immunoprecipitation ( ChIP ) followed by deep sequencing ( ChIP-seq ) . Sequencing reads were mapped onto the B73 reference genome ( AGPv2 ) and the Mo17 genome , respectively [44] ( see Materials and Methods ) . The paired-end reads that were uniquely and concordantly mapped were normalized among replicates and all genotypes using a “down-sampled” approach [52] and subject to the peak calling pipeline ( model-based analysis of ChIP-seq , MACS2 , p-value < 0 . 001 ) [53] ( S4B Fig and S1 Table ) . The peaks that were present in both replicates ( Pearson correlation coefficient = 0 . 97 , based on read coverage over consecutive 1-kb bins ) in one or more genotypes were used for further analysis . The number of peaks ranged from 1 , 874 to 3 , 364 with a total of 10 , 136 peaks in all four genotypes ( S4C Fig ) ( S2 Table ) . Compared to total genomic features , the peaks were enriched in coding , upstream including 2-kb promoter and 5’ untranslated region ( UTRs ) , and downstream flanking sequences ( S5A Fig ) . De novo motif analysis identified two top-scoring motifs , namely , AAAATA , an EE ( AAATATCT ) variant , and AAGAAA , a CBS ( AAAAATCT ) variant , which represented 81% and 65% of total binding peaks , respectively ( Fig 5A ) . The sequence variants may suggest divergence of canonical EE and CBS sequences between Arabidopsis ( eudicot ) and maize ( monocot ) over 150–200 million years of evolution [54] . Further analysis using the TOMTOM tool in MEME [55] found a Dof-binding motif ( AAAGC ) , which was statistically significantly similar to the non-classified motif ( p-value = 0 . 019 ) ( Fig 5A ) . The Dof-transcription factor genes are involved in tissue-specific expression , light signaling , and carbon fixation in maize [42 , 56] , suggesting that ZmCCA1 interacts with other transcription factors to mediate light-regulated gene expression . These peaks were associated with target genes within 10-kb sequences using the filtered gene set ( FGS ) in MaizeGDB ( http://archive . maizegdb . org/cgi-bin/termrefs . cgi ? id=2366450 ) . The number of ZmCCA1-binding target genes ranged from 1 , 406 to 2 , 511 in each genotype , resulting a non-redundant set of 4 , 319 target genes ( S2 Table ) , which is consistent with the observation that 10–20% of maize genes exhibit diurnal expression [29 , 30] , although not all diurnally expressed genes are regulated by ZmCCA1 . ZmCCA1s bound to EE and CBS elements of putative circadian clock homologs ( S5C Fig ) , as shown in the EMSA results ( Fig 2C ) . These putative clock genes were nonadditively expressed at certain time points in the hybrids relative to the parents ( S6A and S6B Fig ) . These data suggest a role for ZmCCA1-binding activities in circadian-mediated gene expression . In the reciprocal crosses , more ZmCCA1-binding peaks were found in BM than in MB ( S6C Fig ) , which may suggest a parent-of-origin effect of circadian clock function , consistent with that in Arabidopsis intraspecific hybrids [10] . Interestingly , the proportion of ZmCCA1-binding peaks in the hybrids was significantly shifted towards ZT3 ( Fisher’s exact test , p-value < 4 . 3E-15 at ZT3; p-value < 4 . 2E-05 at ZT9 ) , while the proportion in the inbreds was relatively unchanged between ZT3 and ZT9 ( Fig 5B ) . The ZmCCA1-binding target genes , associated with ZmCCA1-binding peaks , were partitioned into shared among all genotypes and specific to either hybrids or inbreds and among all time-points ( S7A Fig ) or at each time-point ( S7B Fig ) . Among the shared set of ZmCCA1-binding target genes ( 418 ) ( significant overlaps; Fisher’s exact test , p-value < 1 . 59E-179 ) , the proportion was significantly more at ZT3 in the hybrids than in the inbred lines ( Fisher’s exact test , p-value < 7 . 2E-12 ) but fewer at ZT9 in the hybrids than in the inbred lines ( Fisher’s exact test , p-value < 0 . 03 ) ( Fig 5C ) . The temporal shift of ZmCCA1-binding target genes in the inbred lines were not statistically significant ( Fig 5C ) , but the shift was significant for the targets that were not shared among all genotypes ( S6D Fig ) . Similarly , among the ZmCCA1-binding target genes that were specific to hybrids or inbreds , the temporal shift was significantly in the hybrids but not in the inbreds ( S6E Fig ) . These data suggest that the ZmCCA1-binding activities of target genes in the hybrids have shifted towards the early morning . However , this does not exclude a possibility of many targets that are preferentially bound by ZmCCA1s in the hybrids compared to the inbreds . To test the biological function of the temporal shift , we classified ZmCCA1-binding target genes into GO groups ( S3 Table ) . The shared target genes in both hybrids and inbreds were enriched in the genes in electron transport chain , generation of precursor metabolites and energy , photosynthesis , and light reaction among all time-points ( S7A Fig ) or in different time-points ( S7B Fig ) . This indicates a major role of ZmCCA1s in modulating photosynthetic and metabolic activities , which is consistent with the circadian control of energy and metabolism in other plants [5 , 9] and mammals [57] . Moreover , the hybrid-specific target genes were significantly enriched in the genes involved in protein and cellular protein metabolic processes , indicating a role for ZmCCA1s in altered cellular metabolism that would promote growth vigor in maize hybrids . Notably , the inbred-specific target genes among all time points or in different time points were overrepresented in the genes involved in intracellular transport and protein localization ( S7 Fig ) , which could suggest a role for protein stability and movement in maintaining cellular growth and development [4] . Notably , the enrichment of carbon fixation genes in F1 hybrids was almost exclusive at ZT3 ( Fig 5D ) , whereas these genes in the inbreds also occurred in other time-points , in addition to ZT3 . The genes in GO terms of other pathways , including protein catabolic process , tRNA metabolic process and protein transport , were also enriched , but they were not associated with the phase-shift ( Fig 5D and S3 Table ) . This suggests a uniform shift in activating photosynthetic and carbon metabolism pathways to the early morning in the hybrids . Morning phased-genes associated with the temporal shift included those encoding a starch synthase III ( ZM2G121612 ) , a light harvesting complex photosystem II ( ZM2G033885 ) , a malate dehydrogenase ( ZM2G129513 ) , and a starch synthase II ( ZM2G126988 ) , a putative phospholipid-transporting ATPase ( ZM2G398288 ) ( Fig 6A–6C and 6G and S8A Fig ) . These genes were up-regulated in the morning and in the hybrids compared to the inbreds , which correlated with the shift of ZmCCA1-binding activities in the hybrids ( Fig 6D–6F and S8E Fig ) . Thus , the phase-shift of ZmCCA1-binding in the hybrids is positively associated with expression levels of these morning-phased genes . EMSA assays have confirmed the binding activities of ZmCCA1b to the promoters of ZM2G121612 , ZM2G033885 and ZM2G129513 ( S9 Fig ) . For the afternoon-phased genes , including gi1 and ZM2G412611 encoding an alpha-glucan water dikinase chloroplast precursor , their expression peaks were also shifted from ZT12 to ZT9 ( Fig 6G–6I ) , which correlated with the temporal shift of ZmCCA1-binding in the hybrids ( S8F Fig ) . Collectively , these results indicate that altered temporal binding activities of ZmCCA1s to the clock output genes in the maize hybrids are responsible for the expression rhythms of carbon fixation genes , promoting photosynthesis and metabolism . There were a few exceptions . For example , the expression shift for the genes encoding a photosynthetic reaction center protein ( ZM2G427369 ) and a photosystem II reaction center protein Z ( ZM2G394732 ) was consistent with the ZmCCA1-binding shift in the hybrids ( S8B and S8C Fig ) , but they were expressed at higher levels in most time points examined . The transcript level of the gene ( ZM2G448142 ) encoding an ATP synthase β-subunit in the hybrids was similar to that in the inbreds ( S8D Fig ) . These data suggest that other circadian clock regulators such as TOC1 and ELF3 homologs may also play a role in the diurnal regulation of gene expression in maize hybrids . The ChIP-seq binding activities of ZmCCA1s to the promoters of several carbon fixation genes were confirmed by ChIP-qPCR ( Fig 6D–6F ) and EMSA ( S9 Fig ) assays . To test if the temporal shift of ZmCCA1-binding targets is induced in the hybrids , we examined ZmCCA1-binding activities to the promoters of selected four carbon fixation genes in B104 , OX1-3 and OX1-3XMo17 by ChIP-qPCR ( Fig 7 ) . Compared to the control ( B104 ) , overexpressing ZmCCA1b in the OX1-3 increased the binding levels at the four genes tested , consistent with the levels of ZmCCA1 transcripts that are oscillating in the overexpression lines ( Fig 4D ) . Remarkably , in the F1 hybrid ( OX1-3XM ) , the ZmCCA1-binding activities were not only increased in levels but also shifted to the morning-phase ( ZT3 ) , which was not obvious in the OX1-3 line . Although not all possible genotypes were tested , the data suggest a temporal sift of ZmCCA1-binding activities in the hybrids towards early morning , which could play an important role in maize heterosis . Collectively , these results indicate that altered temporal ZmCCA1-binding activities to the clock output genes in the maize hybrids are responsible for expression rhythmicity of carbon fixation genes , promoting photosynthesis , metabolism and growth vigor . Circadian regulation of gene expression in Arabidopsis intraspecific hybrids is established during embryo development and maintained during seedling growth [10] , and biomass heterosis is established during early stages of seedling development [58 , 59] . Consistent with this notion , expression of the circadian clock genes ( ZmCCA1a and ZmCCA1b ) and their output genes in maize hybrids is developmentally regulated during seedling stages ( S8G Fig ) . Expression of ZmCCA1a , ZmCCA1b and carbon fixation genes was nonadditive in maize hybrids at 5 DAP and 8 DAP but gradually became additive at 11 DAP and 14 DAP . As a control , Cell Number Regulator 2 ( CNR2 ) , which controls cell numbers in maize [60] , was nonadditively expressed in the hybrids late in seedling development ( 14 DAP ) . These data suggest that the influence of circadian regulation on seedling heterosis in maize is likely developmentally regulated , corresponding to gene expression dynamics during maize development [42] .
Heterosis is predicted to arise from allelic interactions between parental genomes , leading to altered regulatory networks that promote the growth and fitness of hybrids [1 , 3 , 5] . Here , we demonstrate that one such regulator is the circadian clock in maize hybrids . The circadian clock genes are functionally conserved in Arabidopsis and maize . In hybrids , the maize central clock proteins target thousands of output genes early in the morning , including carbon fixation genes . The data collectively support a phase-shift model for heterosis ( Fig 8 ) . ZmCCA1-binding target genes are involved in energy and metabolism , which is established early in the seedling and subsequently maintained during growth . During establishment , morning-phased bindings of ZmCCA1s to carbon fixation genes in F1 hybrids ( relative to the inbreds ) may cause nonadditive gene expression; consequently nonadditive increases in carbon fixation rate , and , ultimately , biomass accumulation . During late stages of development , gene expression in the hybrids is shifted towards additive expression of the circadian-mediated carbon fixation and metabolic genes . This shift of different gene expression modes may explain different findings of additive expression [7] , nonadditive , and/or all modes of gene expression [8] , which have been documented in maize hybrids or hybrids with different ploidy levels [61] . The temporal shift from non-additive to additive gene expression has also been observed in A . thaliana F1 hybrids [58 , 62] . Moreover , ZmCCA1-binding target genes are dependent on genotypes ( inbreds vs . hybrids ) ( S7 Fig ) . In the hybrids , coordinated regulation of the protein and cellular metabolic genes could allow them to achieve greater protein metabolic efficiency , as many cellular metabolic processes are simultaneously engaged in production of stable or efficient metabolites , saving energy used in their synthesis and metabolism [4] . Our findings provide novel insights into a circadian-mediated mechanism for the temporal shift of morning-expressed genes in the maize hybrids to promote photosynthetic and metabolic activities , leading to biomass heterosis . Maize leaf displays a gradient with respect to cell maturation and transcript levels [42 , 43] . This developmental gradient is reflected by the differential starch and sugar accumulation between the sink and source within a leaf ( Fig 2 ) . In A . thaliana intraspecific hybrids , intermediate gene expression and higher metabolic activities lead to enhanced performance in the hybrids [58] . This early establishment of heterosis persists over the growth cycle of the hybrids , leading to higher biomass and yield [62] . In maize hybrids , the change from nonadditive gene expression in early developmental stages to additive gene expression in the late stages may serve as a transition or switch to accelerate the rate of development , while the additive gene expression maintains or stabilizes this gradient development at the accelerated rate . The transitional changes in gene expression will enhance metabolic activities that promote higher biomass accumulation and ultimately higher yield in the hybrids . Circadian clocks regulate biological processes of most organisms including plants and animals [57] . Plant growth and development , from stress responses , biomass accumulation , to seedling growth , flowering , and fruit formation , is directly controlled by clock regulators through transcriptional and post-transcriptional regulation of output processes [12 , 13] . Physiological activities , including CO2 fixation and starch metabolism , are diurnally regulated by circadian clock factors . CCA1 serves as both a transcriptional repressor and activator of target genes , including those in the core circadian oscillator and output pathways [12 , 49] . For example , CCA1 is a repressor for evening-phased genes in A . thaliana; it directly binds to promoters of the evening-phased central clock genes ( TOC1 and GI ) and represses their expression [63 , 64] . In the cca1-11 mutant , expression of evening-phased starch metabolic genes is up-regulated [9] . Meanwhile , CCA1 is an activator for morning-phased genes; it directly binds to the promoters of the morning-phased central clock genes ( PRR7 and PRR9 ) [65] and cold stress responsive genes [66] , which are repressed in cca1-11 and cca1 lhy mutants . CCA1 also regulates expression of morning-phased photosynthetic genes , including CAB genes , and their expression is suppressed in the cca1 mutant [67] . A recent study indicates that circadian-mediated stress-responsive genes could promote and predict heterosis in Arabidopsis [68] . Our study has demonstrated functional conservation between Arabidopsis and maize CCA1 homologs . Consistent with the positive role of CCA1 in morning-phased genes in Arabidopsis , ZmCCA1-bindings to the morning-phased carbon fixation genes were positively correlated with the expression of these genes in the morning ( ZT3 ) , whereas ZmCCA1 association with promoters of these genes in the afternoon ( ZT9 ) were negatively correlated with the expression of these genes . It is likely that the positive and negative roles of CCA1 are achieved through its interaction with other factors . For example , in A . thaliana , CCA1 interacts with LHY in vitro and in vivo , and the two factors function synergistically in circadian clock regulation [69] . A splice variant of CCA1 ( CCA1β ) , which lacks the MYB DNA-binding domain , interferes with the transcriptional function of CCA1 by interacting with the full-length CCA1 ( CCA1α ) or LHY to form nonfunctional heterodimers [46] . Without CCA1β , the circadian period is longer , whereas overexpressing CCA1β leads to a short-period phenotype , suggesting that interacting with spliced CCA1β fine-tunes CCA1α function . In maize , although the long-form of ZmCCA1a is not studied , the short-form of ZmCCA1a lacking the MYB DNA-binding domain could act as a dominant negative factor to fine-tune circadian rhythms in maize as in Arabidopsis [46] , as the effect of this splicing variant in Arabidopsis overexpressing and complementation tests is relatively small compared to ZmCCA1b ( Fig 3 and S3 Fig ) . Both maize ZmCCA1b and ZmCCA1a are up-regulated in the hybrids in a morning-specific manner with higher expression of ZmCCA1b , which may serve as co-regulators for each other with redundant and/or separate functions in the transcriptional regulation of output ( photosynthetic , carbon fixation , and metabolic ) genes in the hybrids . The rhythmic expression of circadian-mediate genes is likely established over 24-hour diurnal cycles based on ZmCCA1 expression rhythms ( Fig 3 ) . It is notable that the F1 hybrids share similar molecular and physiological phenotypes such as photosynthetic rate and expression of carbon-fixation genes with one of the parent B73 . However , the hybrid still performs better than the best parent , which is known as better-parent heterosis ( BPH ) [1] . This suggests that some molecular events could be more important than others in promoting growth vigor . For example , net carbon gain ( starch and sugar contents ) in the sink and source correlates better with the level of heterosis than other molecular and physiological parameters ( Fig 2B ) . Interestingly , the genes corresponding to starch metabolic pathways are regulated by the circadian clock and nonadditively expressed , which could play an important role in heterosis . The circadian-mediated phase-shift model of gene expression in maize hybrids should be further tested using quantification of genome-wide gene expression data , which will demonstrate relative effects of the earlier and extra induction of many morning-phased genes on the contribution to heterosis . In addition , the effect of these morning-phased genes on heterosis could be tested using maize transgenic lines , mutant , and/or recombinant inbred line populations . Further testing the functions of these genes will reveal mechanisms for the circadian clock to regulate unique and complex biological pathways that stimulate growth vigor in maize hybrids .
The inbred lines B73 and Mo17 and their reciprocal F1 hybrids B73XMo17 ( BM ) and Mo17XB73 ( MB ) were used for this study . All maize plants were grown at 600 μmol m−2 s−1 under 16L:8D ( hours of light:dark ) cycle with temperature 28°C ( light ) and 23°C ( dark ) except for the experiments to measure photosynthetic rates , starch and sucrose content . They were carried out in a glasshouse where natural sunlight ( maximum photosynthetically active radiation 598 ± 354 μmol m−2 s−1 , mean ± SD ) was supplemented with fluorescent lighting to achieve a 16L:8D cycle with daily maximum temperature at 26 . 7 ± 2 . 1°C ( mean ± SD ) , daily minimum temperature at 19 . 4 ± 0 . 4°C ( mean ± SD ) , relative humidity at 57 ± 7% ( mean ± SD ) , and net CO2 exchange examined during the day . The plants were placed in a randomized design and rotated on a daily basis to minimize positional effects . The growth traits included seed mass , aerial biomass , plant height , and leaf-blade length and area . For seed weight , 10 seeds randomly selected from three replicates for each genotype were weighed . For aerial biomass , aboveground seedlings for each genotype were collected at 3 , 4 , 5 and 6 days after planting ( DAP ) ( n = 4–8 ) . The biomass was weighed after desiccation for 48 h at 80°C . Plant height and leaf-blade length for each genotype were daily measured from 5 DAP to 12 DAP ( n = 5 ) . Plant height was measured as the distance from the ground to the top of seedlings . Leaf-blade length was measured as the distance from the base to the tip of the 2nd leaf . The 2nd leaf at 12 DAP for each genotype was photographed and processed for leaf area using NIH ImageJ software ( http://imagej . nih . gov/ij/ ) [70] ( n = 6 ) . All experiments were replicated three times , unless noted otherwise . Arabidopsis seedlings were plated onto Murashige-Skoog media ( Sigma , St . Louis , MO ) supplemented with 0 . 8% agar , 3% sucrose ( MS agar ) and appropriate antibiotics . After stratification in the dark at 4°C for 2 days , plates were transferred to a growth room with a 16L:8D cycle ( 80 μmol m−2 s−1 ) at 22°C . Luminescence recordings were performed on a TopCount ( Packard Bioscience , Shelton , CT ) over 7 days , as previously described [10] . The data were analyzed by fast Fourier transform–nonlinear least squares ( FFT–NLLS ) [71] using the interface provided by the Biological Rhythms Analysis Software System version 3 . 0 ( BRASS ) ( http://www . amillar . org ) . The construct for ZmCCA1b ( ZM2G014902 ) overexpression in the binary vector pTF101 . 1gw1 [72] ( generously provided by Dr . Kan Wang , Iowa State University , Ames , IA ) consisted the ZmUBI promoter-ZmUBI intron ( ZmUBIpro ) cassette immediately upstream of the ZmCCA1 coding sequence and the octopine synthase ( OCS ) terminator immediately downstream . The ZmUBIpro cassette was PCR amplified from the pANIC6A vector [73] with primers F 5’-CACCGTAAGCTTAATGCAGTGCAGCGTGACCC-3’ and R 5’-GTAAGCTTTGCAGAAGTAACACCAAACAACAGGGTG-3’ , which added the HindIII restriction site ( underlined ) onto each side of the fragment . The OCS terminator was PCR amplified from the pANIC6A vector with primers F 5’-CCTGCTTTAATGAGATATGCGAGA CGC-3’ and R 5’-CACCAAAACGACGGCCAGTGCCAA G-3’ . The ZmCCA1b coding sequence was PCR amplified with primers F 5’-CACCATGGAGGTGAATTCCTCTGGTGAGGAAAC-3’ and R 5’-TTATGTTGATGCTTCACTATCAAGACGAATCCTCTT-3’ . The ZmUBIpro cassette was subcloned into the Hind III site upstream of attR1 site in pTF101 . 1gw1 . The ZmCCA1b coding sequence was moved into with LR Clonase II according to the manufacturer’s recommendations ( Thermo-Fisher , Waltham , Massachusetts ) . The OCS terminator was subcloned into the AscI site downstream of the ZmCCA1b coding sequence . The Plant Transformation Facility ( Iowa State University , Ames , IA ) transformed the complete ZmCCA1b overexpression binary construct into the B104 maize inbred by Agrobacterium-mediated transformation according to their published protocol [74] . Pollen from T0 plants , which were confirmed to carry at least copy of the one construct , was used for crosses to the B104 inbred . Transgenic lines were maintained in a hemizygous state by crosses to the B104 inbred . Transgenic plants were screened using fluorescent probe-based endpoint qPCR . Genomic DNA was extracted from leaf tissue taken from the newest expanded leaf when plants were between V3 and V8 . The presence of ADH and BAR was assessed together in PCR reactions with primers 5’-TGTTGAGCAGATCTCGGTGAC-3’and 5’-GTTTCTGGCAGCTGGACTTC-3’ , with probe 5’-[HEX]AGGACCGGACGGGGCGGTA[BHQ1]-3’ , for the bar gene in pTF101 . 1gw1; and primers 5’-GAATGTGTGTTGGGTTTGCAT-3’ and 5’-TCCAGCAATCCTTGCACCTT-3’ , with probe 5’-[FAM]TGCAGCCTAACCATGCGCAGGGTA[BHQ1]-3’ , for ADH1 , which is a single copy gene in the maize genome [75] . Samples with PCR amplification for both ADH and BAR were scored as transgenic and those with ADH alone were scored as non-transgenic . Field trials took place at Oxford Tract in Berkeley , CA during summer 2015 . The field was sown in a randomized complete block design with two trials separated by one week , planted in late May . In the greenhouse , plants were grown under 16L:8D cycle , supplemented by LumiGrow Pro 325 LED lights ( LumiGrow , Inc . , Novato , CA ) , with 25°C days and 20°C nights . Two individual trials were replicated in greenhouse conditions . 5 plants of each family along with 5 non-transgenic siblings were maintained for each trial . In field trials , weekly height measurements were taken beginning at the V8 stage until final height was reached . In greenhouse trials , weekly height measurements were taken starting when plants were at the V7 stage until final height was reached . Height was measured from prop roots to the collar of the last fully expanded leaf . Node length was measured at the end of the field or greenhouse trial after plants had reached maturity . Measurements were taken from the upper prop roots to the final visible node , or the base of the tassel . Plant tissue for chlorophyll analysis was harvested using a Harris Uni-Core 2 . 5 mm biopsy punch ( Ted Pella , Redding , CA ) . 4 punches were taken midway from the base to the tip of the leaf , equidistant from the mid-vein and the leaf edge . Tissue was immediately placed in 1 mL DMSO and incubated at 65°C for 30 min . Absorbance at 645 nm and 663 nm was used to determine chlorophyll content based on Arnon’s equation [76] . CCA1 protein sequences are AtCCA1 ( At2g46830 ) , AtLHY ( At1g01060 ) , SbCCA1 ( Sb07g003870 ) , OsCCA1 ( Os08g0157600 ) , PnLHY1 ( AB429410 ) , PnLHY2 ( AB429411 ) , BraA . LHY . a ( Bra030496 ) , McCCA1 ( AY371287 ) and GmLCL1 ( Glyma11g15580 ) , At: Arabidopsis thaliana; Sb: Sorghum bicolor; Os: Oryza sativa; Pn: Populus nigra; BraA: Brassica rapa; Mc: Mesembryanthemum crystallinum; and Gm: Glycine max . The protein sequences were aligned using the ClustalW module [77] . The phylogenetic tree was constructed using the Neighbor-Joining method [78] . The bootstrap values were calculated with 1 , 000 replicates and shown next to the branches . The evolutionary distances were computed using the Poisson correction method [79] with the units of the number of amino acid substitutions per site . All positions containing gaps and missing data were eliminated . A total of 472 positions were analyzed using MEGA6 [80] . For gene expression analysis , aerial tissues were harvested and immediately frozen in the liquid nitrogen . Total RNA was extracted from the frozen tissues by PureLink Plant RNA reagent ( Invitrogen , Carlsbad , CA ) and treated with RQ1 DNaseI ( Promega , Madison , WI ) according to the manufacturers’ instructions . For cDNA synthesis , 1 μg of DNaseI-treated total RNA was incubated with Omniscript reverse transcriptase ( Qiagen , Valencia , CA ) in the presence of 10 μM random hexamer ( GeneLink , Hawthorne , NY ) . For qPCR , FastStart Universal SYBR Green Master ( ROX ) ( Roche Applied Science , Indianapolis , IN ) was used in the presence of gene-specific primers and template cDNAs in an ABI7500 ( Applied Biosystems , Foster City , CA ) or LightCycler 96 machine ( Roche Applied Science , Indianapolis , IN ) . The control was 18S rRNA to estimate the relative expression levels of each gene in three biological replicates . A list of primers for gene expression analysis is provided in S4 Table . Photosynthetic rate was measured on fully expanded 2nd leaves of seedlings at 12 DAP ( n = 9 ) , grown under 16L:8D cycle with natural sunlight , every 2 hours from dawn ( ZT0 ) to dusk ( ZT14 ) . A LI-6400XT portable photosynthesis analyzer ( LICOR Environmental , Lincoln , NE ) equipped with a light source ( LI-6400-40 ) and CO2 mixer ( LI-6400-01 ) was used to determine net CO2 assimilation at a reference CO2 concentration of 400 μmol mol−1 , with cuvette conditions set to match ambient conditions at the start of each measurement period . The experiment was reproduced twice , in February and April 2012 , respectively . Data from the April experiment is shown . Starch and sucrose contents were measured on source ( middle-tip of the 2nd leaf ) and sink ( base of the 2nd leaf ) of additional replicate plants grown alongside those used for photosynthesis assays . Pooled plants in three biological replicates at ZT14 were used for testing . Leaf discs were collected using an 11 mm-diameter cork borer , weighed and immediately frozen in liquid nitrogen . The frozen tissues were ground , mixed with a homogenization buffer ( 500 mM MOPS pH 7 . 5 , 5 mM EDTA , 10% ethyl glycol ) , and then filtered through Miracloth ( CalBiochem , San Diego , CA ) . After centrifugation , pellets were dissolved in DMSO to extract the insoluble carbohydrate fraction , while supernatant was transferred to a new tube as the soluble carbohydrate fraction . The starch content was measured from the insoluble carbohydrate fraction using a commercial assay kit according to the manufacturer’s instruction ( R-Biopharm , Darmstadt , Germany ) . Sucrose content was measured from the soluble carbohydrate fraction using a commercial assay kit according to the manufacturer’s instruction ( K-SUFRG Megazyme , Bray , Ireland ) and as previously described [9] . The coding sequence ( CDS ) of ZM2G474769 ( ZmCCA1a ) was amplified from B73 cDNA by the primer pair 5’-GAATTCATGCCCTTGAGCAATGAG-3’ ( EcoRІ , underlined ) and 5’-GTCGACTCATGTTGATGCTTCACTAT-3’ ( SalІ ) . The full-length ZmCCA1b ( ZM2G014902 ) CDS was amplified from B73 cDNA by the primer pair 5’- GGATCCATGGAGGTGAATTCCTCTGGC-3’ ( BamHІ ) and 5’- GTCGACTTATGTGGATGCTTCGCTATC-3’ ( SalІ ) . The ZmCCA1a or ZmCCA1b cDNA fragment was cloned into a pGEM-T ( Promega , Madison , Wisconsin ) . After sequence verification , the ZmCCA1a or ZmCCA1b CDSs was subcloned into pMAL-C2 ( New England BioLabs , Beverly , MA ) through EcoRІ/SalІ and BamHІ/SalІ restriction sites , respectively . E . coli strain Rosetta-gami B competent cells ( Novagen , Madison , WI ) was used to transform empty pMAL ( expressing maltose-binding protein , MBP ) , pMAL-ZmCCA1a or pMAL-ZmCCA1b , which were grown in 4 ml of Luria-Bertani ( LB ) media with Carbenicillin ( 100 mg/L ) at 37°C for 18 h . The overnight cultures of Rosetta-gami B cells containing pMAL , pMAL-ZmCCA1a or pMAL-ZmCCA1b construct were diluted into 1:100 in 80 ml LB media with Carbinicillin ( 100 mg/L ) and grown at 37°C to an OD600 value of 0 . 5 , when isopropyl-β-D-thiogalactoside ( IPTG ) ( 0 . 1 mM ) was added . After 20 h of additional incubation at 16°C , cells were harvested after centrifugation at 4 , 000 g at 4°C for 10 min and resuspended in 2 ml of Column buffer ( 20 mM Tris-HCl , 200 mM NaCl , 1 mM EDTA ) . After frozen at -20°C for 18 h , cells were lysed by a Bioruptor Sonicator ( Diagenode , Sparta , NJ ) and centrifuged at 20 , 000 g at 4°C for 20 min . The cleared cell lysates were diluted 1:5 with Column buffer , and loaded on amylose-coupled agarose resin columns prepared according to the manufacturer’s instruction ( New England BioLabs , Beverly , MA ) . After columns were washed with 12 volumes of Column buffer , MBP , rZmCCA1a and rZmCCA1b were eluted with elution buffer ( 20 mM Tris-HCl , 200 mM NaCl , 1 mM EDTA , 10 mM Maltose ) . After filtration by Amicon Ultra 100 K ( Millipore , Darmstadt , Germany ) , the purified MBP , rZmCCA1a or rZmCCA1b was aliquoted and stored at -80°C . ChIP was performed as previously described with following modifications [81] . Aerial tissues of three biological replicates were harvested at ZT3 , ZT9 and ZT15 and were completely submerged in fresh prepared formaldehyde buffer ( 0 . 4 M Sucrose , 10 mM Tris-HCl , 1 mM PMSF , 3% Formaldehyde , 5 mM β-mercaptoethanol ) . A vacuum was applied for 20 min; after adding glycine to a final concentration 125 mM , another vacuum was applied for 5 min . After the formaldehyde/glycine buffer was removed , the cross-linked tissues were washed with sterilized water , briefly dried with paper towels , immediately frozen in liquid nitrogen and stored at -80°C for experimental use . Sonication was performed with 10 cycles of 30 s pulses on and 30 s pulses off to achieve an average fragment size of 400-bp using a Bioruptor Sonicator ( Diagenode , Sparta , NJ ) . Samples were subjected to centrifugation at 13 , 800 g at 4°C for 10 min , and the supernatant containing chromatin was transferred to a new 1 . 5 ml tube . Immunoprecipitation ( IP ) was performed using 600 μl of sonicated chromatin with 5 μg of anti-CCA1 antibody . For each IP sample , a mock ( no antibody ) and input ( no IP ) were included . Purified DNA ( IP , mock and input ) from ChIP was resuspended in TE , pH 7 . 5 and used for qPCR and ChIP-seq library preparation . ChIP-seq libraries for ChIP and input samples from two biological replicates were constructed using the standard NEB protocol ( New England BioLabs , Beverly , MA ) using custom made adapters containing barcodes used to pool multiple samples for sequencing . ChIP DNA was subject to end repair , dA-tailing , ligation with the adapters , and amplification by 18 cycles of PCR using Next High-Fidelity 2XPCR Master Mix ( New England BioLabs , Beverly , MA ) . Pair-end ( 2X100-bp ) sequencing was performed on an Illumina HiSeq 2500 at The University of Texas at Austin Genomic Sequencing and Analysis Facility . For qPCR , purified ChIP and mock DNA was diluted 2 times , and input DNA was diluted 5 times . The diluted DNA ( 2 μl ) was used for qPCR in an ABI7500 machine ( Applied Biosystems , Foster City , CA ) using FastStart Universal SYBR Green Master ( ROX ) ( Roche Applied Science , Indianapolis , IN ) . Enrichment of the binding in IP and mock samples was normalized relative to the corresponding input sample . A list of primers for ChIP-qPCR analysis is provided in S4 Table . Raw reads were subject to quality trimming and adaptor clipping using FASTX-Toolkit ( hannonlab . cshl . edu/fastx_toolkit ) followed by removing orphan reads . The filtered pair-end reads were mapped to the maize reference genome ( Zmays_284_AGPv2 , release 5b . 60 ) and Mo17 genome ( reference-guided assembly based on the B73 genome ) [44 , 82] , using Bowtie ( version 2 . 1 . 0 ) [83] , allowing 1 mismatch in the 20-bp ‘seed’ with options ‘—score-min L , 0 , -0 . 3 -X 1000—no-mixed—no-discordant’ . Particularly , for B73 and Mo17 , the filtered pair-end reads were mapped to the maize B73 reference genome and the Mo17 genome , respectively . For F1 hybrids , the filtered pair-end reads were mapped to both B73 and Mo17 reference genomes , and the best alignment was selected for each read . Only reads concordantly mapping to the genome exactly 1 time was kept and used for peak calling . Duplicate reads were removed from the Bowtie output using samtools rmdup command [84] . To normalize sequencing depth of the mapped reads among samples from different genotypes and time-points , we adopted a normalization method based on the previously published paper [52] . In order to adjust different sequencing depths among genotype and time-points , the uniquely and concordantly mapped paired-end reads were “down-sampled” to the lowest number of samples without PCR duplicates ( S1 Table ) . For example , if B73 , Mo17 , BM and MB had 4 , 5 , 6 and 7 million reads respectively , all samples were down-sampled to 4 million reads . This will ensure that numbers of binding peaks are comparable among genotypes and time-points . The criteria for identifying targets were based on abundance of peak enrichment using ANOVA test ( p-value < 0 . 05 as a cut-off ) . When target genes showed same abundance 2 or more time-points , they were excluded from the phase-shift analysis . The latest version of model-based analysis of ChIP-seq algorithm ( MACS , version 2 . 0 . 1 ) [53] was used to identify enriched peaks on each ChIP-seq file against the corresponding input file using a mappable genome size of–g 2 . 07e+09 and cut off p-value of 1e-3 . Peaks were defined if they overlapped in two biological replicates . To make a master-peak list from the three time-points , the peaks obtained from each time-point were merged for each genotype . Integrated Genomics Viewer ( version 2 . 3 . 46 ) [85] was used to visualize duplicate-filtered input subtracted ChIP signals . For the normalization , modules of the deepTools suite ( http://deeptools . ie-freiburg . mpg . de ) [86] were used . De novo motif analysis was performed with the total master-peaks using MEME suite [87] in which JASPAR CORE ( 2014 ) plants database was selected for motif database . GO enrichment analysis was performed using the web interface of agriGO ( http://bioinfo . cau . edu . cn/agriGO/ ) [88] with false-discovery rate adjusted p-value < 0 . 05 ( Hypergeometric test ) as a cutoff . Biological process among ontology categories was used . The heatmap for GO analysis was generated using R package gplots . Peak distribution was determined with respect to the maize gene model ( release 5b . 60 ) . To create CAB2:LUC , a native Arabidopsis CAB2 ( AT1G29920 ) promoter fragment was amplified from Col-0 wild-type genomic DNA by the primer pair 5’- CTCGAGTTATATTAATGTTTCGATCATC-3’ ( XhoІ ) and 5’- CCATGGGTTCGATAGTGTTGGATTATA-3’ ( NcoІ ) and cloned into pGEM-T ( Promega , Madison , WI ) . After sequence validation , the CAB2 promoter fragment was subcloned into pFAMIR binary vector ( Basta resistance ) , which was fused with LUC CDS , via XhoІ/NcoІ restriction sites . The CAB2:LUC construct was transformed into the Col-0 and WS wild-types , respectively , using a floral dip method [89] . Homozygous T3 lines in each background were selected for the following analysis . To generate CCA1-OX , ZmCCA1-OX and ZmCCA1b-OX transgenic lines , CCA1 , ZmCCA1a and ZmCCA1b cDNAs were amplified by the following primer pairs , respectively , CCA1 5’-CTCGAGATGGAGACAAATTCGTCTGG-3’ ( XhoІ ) and 5’- GGATCCTCATGTGGAAGCTTGAGTTTC-3’ ( BamHІ ) , ZmCCA1a 5’- CTCGAGATGCCCTTGAGCAATGAG-3’ ( XhoІ ) and 5’- GGATCCTCATGTTGATGCTTCACTATC-3’ ( BamHІ ) , and ZmCCA1b 5’- CAAGCTCGAGATGGAGGTGAATTCCTCTGGC-3’ ( XhoІ ) and 5’- GGATCCTTATGTGGATGCTTCGCTATC-3’ ( BamHІ ) , and cloned into pGEM-T ( Promega , Madison , WI ) . After sequence validation , CCA1 CDS , ZmCCA1a CDS or ZmCCA1b CDS was subcloned into pF35SE binary vector ( conferring Kanamycin resistance ) through XhoІ/BamHІ restriction sites . The resulting constructs designated CCA1-OX , ZmCCA1a-OX and ZmCCA1b-OX , respectively . The construct of vector control , CCA1-OX , ZmCCA1-OX or ZmCCA1b-OX was introduced into the transgenic Col-0 plants that express CAB2:LUC . T2 lines were used for the luciferase and biomass analysis . For the cca1-11 complementation analysis with maize CCA1 homologs , a native Arabidopsis CCA1 ( AT2G46830 . 1 ) promoter fragment was amplified from WS wild-type genomic DNA by the primer pair , 5’- GAATTCGCCACGTCCTTCCTTCAATC-3’ ( EcoRІ ) and 5’- CTCGAGCACTAAGCTCCTCTACACAA-3’ ( XhoІ ) , and cloned into pGEM-T ( Promega , Madison , WI ) . After sequence validation , the CCA1 promoter fragment was subcloned into pFAMIR binary vector ( Hygromycin resistance ) , via EcoRІ/XhoІ restriction sites . To create CCA1:CCA1 , the CCA1 CDS was amplified from the WS cDNA by the primer pair , 5’- ATTTAAATATGGAGACAAATTCGTCTGG-3’ ( SwaІ ) and 5’- GGATCCTCATGTGGAAGCTTGAGTTTC-3’ ( BamHІ ) , and cloned into pGEM-T ( Promega , Madison , WI ) . To generate CCA1:ZmCCA1a , the ZmCCA1a CDS was amplified from Z . mays B73 cDNA by the primer pair , 5’- CTCGAGATGCCCTTGAGCAATGAG-3’ ( XhoІ ) and 5’- GGATCCTCATGTTGATGCTTCACTATC-3’ ( BamHІ ) , and cloned into the pGEM-T ( Promega , Madison , WI ) . To generate CCA1:ZmCCA1b , the ZmCCA1b CDS was amplified from Z . mays B73 cDNA by the primer pair , 5’- CAAGCTCGAGATGGAGGTGAATTCCTCTGGC-3’ ( XhoІ ) and 5- ACCGGATCCTTATGTGGATGCTTCGCTATC-3’ ( BamHІ ) , and cloned into the pGEM-T ( Promega , Madison , WI ) . After sequence validation , CCA1 CDS , ZmCCA1a CDS or ZmCCA1b CDS was subcloned into the pFAMIR ( Hygromycin resistance ) , which harbors the CCA1 promoter fragment , through the respective restriction sites , generating CCA1:CCA1 , CCA1:ZmCCA1a , and CCA1:ZmCCA1b constructs , respectively . The construct of CCA1:CCA1 , CCA1:ZmCCA1a or CCA1:ZmCCA1b was transformed into A . thaliana cca1-11 plants ( obtained from the Arabidopsis Biological Resources Center , ABRC , CS9865 ) , which express CAB2:LUC , generating CCA1:CCA1 cca1-11 , CCA1:ZmCCA1a cca1-11 or CCA1:ZmCCA1b cca1-11 transgenic lines , respectively . T1 transgenic lines were selected for the luciferase analysis . DNA probes were generated by annealing PAGE-purified sense and antisense oligonucleotides ( S5 Table ) . The double-stranded oligonucleotides were [32P] end-labeled using a T4 Polynucleotide Kinase according to the manufacturer’s instruction ( New England BioLabs , Beverly , MA ) . The recombinant proteins ( 0 . 5 to 2 pmol ) were mixed with 20 fmol of the radiolabeled probes , without or with variable amounts of unlabeled competitor DNA in reaction buffer ( 25 mM HEPES-KOH pH7 . 5 , 2 . 5 mM DTT , 75 mM KCl , 10% glycerol , 1 . 25 ng poly-dIdC ) . Each reaction was incubated on room temperature for 10 min without the probes and then incubated on ice for 20 min with the radiolabeled probes . The competitor concentrations were at 0 ( - ) , 0 . 5 ( 25X ) , 1 ( 50X ) and 2 ( 100X ) pmol . After the incubation , the reaction mixtures were resolved by electrophoresis on a 5% non-denaturing polyacrylamide gel . Gels were dried in a gel dryer ( Bio-Rad , Richmond , CA ) and exposed to X-ray film ( Kodak , Rochester , NY ) . Total crude protein was extracted from the aerial tissues of seedlings at 5 DAP using an extraction buffer ( 50 mM Sodium Phosphate pH 8 . 0 , 1 mM EDTA , 5 mM DTT , 10% Glycerol ) . 30 μg of total crude proteins were resolved in 10% SDS-PAGE gel and transferred to PVDF membrane ( GE Healthcare , Piscataway , NJ ) . ZmCCA1s were detected using anti-CCA1 antibody ( 1:2 , 000 ) that was generated against the N-terminus MYB DNA-binding domain ( epitope: residues 11–77 of ZmCCA1s , S2B Fig ) , followed by HRP-conjugated anti-rabbit secondary antibody ( 1:20 , 000 ) ( 12–348 Upstate , Lake Placid , NY ) . Loading control was detected using an anti-H3 antibody ( 1:5 , 000 ) ( Ab1791 Abcam , Cambridge , MA ) . Immunoreactive protein was visualized on X-ray film ( Kodak , Rochester , NY ) using SuperSignal West Pico Substrate ( Thermo Scientific , Waltham , MA )
|
All corn in the USA is grown as hybrids , which grow more vigorously and produce higher yield than their parents , a phenomenon known as heterosis . The molecular basis for heterosis remains elusive . Heterosis is predicted to arise from allelic interactions between parental genomes , leading to altered regulatory networks that promote the growth and fitness of hybrids . One such regulator is the circadian clock , which is functionally conserved in Arabidopsis and maize . Disrupting corn CCA1 expression reduces growth vigor . In corn hybrids , CCA1 proteins target thousands of output genes early in the morning , as if the hybrids wake up early to promote photosynthesis , starch metabolism and biomass accumulation . This early acting mechanism could guide breeding and selection of high-yield hybrids .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biotechnology",
"plant",
"anatomy",
"chemical",
"compounds",
"brassica",
"carbohydrates",
"organic",
"compounds",
"cereal",
"crops",
"plant",
"science",
"model",
"organisms",
"crops",
"plant",
"genomics",
"chronobiology",
"plants",
"starches",
"arabidopsis",
"thaliana",
"research",
"and",
"analysis",
"methods",
"grasses",
"crop",
"science",
"maize",
"gene",
"expression",
"plant",
"genetics",
"chemistry",
"leaves",
"agriculture",
"circadian",
"rhythms",
"heterosis",
"plant",
"and",
"algal",
"models",
"organic",
"chemistry",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"plant",
"biotechnology",
"organisms"
] |
2016
|
Temporal Shift of Circadian-Mediated Gene Expression and Carbon Fixation Contributes to Biomass Heterosis in Maize Hybrids
|
Folding of the chromosomal fibre in interphase nuclei is an important element in the regulation of gene expression . For instance , physical contacts between promoters and enhancers are a key element in cell-type–specific transcription . We know remarkably little about the principles that control chromosome folding . Here we explore the view that intrachromosomal interactions , forming a complex pattern of loops , are a key element in chromosome folding . CTCF and cohesin are two abundant looping proteins of interphase chromosomes of higher eukaryotes . To investigate the role of looping in large-scale ( supra Mb ) folding of human chromosomes , we knocked down the gene that codes for CTCF and the one coding for Rad21 , an essential subunit of cohesin . We measured the effect on chromosome folding using systematic 3D fluorescent in situ hybridization ( FISH ) . Results show that chromatin becomes more compact after reducing the concentration of these two looping proteins . The molecular basis for this counter-intuitive behaviour is explored by polymer modelling usingy the Dynamic Loop model ( Bohn M , Heermann DW ( 2010 ) Diffusion-driven looping provides a consistent framework for chromatin organization . PLoS ONE 5: e12218 . ) . We show that compaction can be explained by selectively decreasing the number of short-range loops , leaving long-range looping unchanged . In support of this model prediction it has recently been shown by others that CTCF and cohesin indeed are responsible primarily for short-range looping . Our results suggest that the local and the overall changes in of chromosome structure are controlled by a delicate balance between short-range and long-range loops , allowing easy switching between , for instance , open and more compact chromatin states .
Chromosomes are highly folded inside the interphase nucleus . Their structure has been extensively studied by light and electron microscopy and more recently by genome-wide mapping of intra- and inter-chromosomal contacts . The picture that is emerging is that the chromosomal fibre is packed in a hierarchical way on different length scales [1]–[3] . Complete chromosomes are confined to chromosome territories ( CTs ) that intermingle only to a limited extent [4]–[6] . Another well-defined level is that of the topologically associated domains ( TADs ) , which are distinct sub-chromosomal structures in the Mb range [7]–[10] . These two levels of chromosomal organisation are evolutionary conserved in metazoans . Chromosomal folding is intimately linked with genome function . TADs are functional genomic units , each containing genes that often are transcriptionally or epigenetically co-regulated [2] , [9] . Specific sequence elements between TADs , called insulators , confine long-range regulatory interactions between for instance enhancers and promoters to the individual TADs [11] . TADs coincide with DNA replication units , which show a distinct pattern of replication timing that is remarkably well conserved between mouse and man [12]–[15] . Another striking aspect of chromatin folding is the considerable cell-to-cell variation observed in populations of otherwise identical cells [16] , [17] . Overall , chromosomal architecture seems a mix of well-defined and probabilistic components . Despite the extensive information about chromosomal architecture and its importance for the functioning of the genome , underlying principles that direct chromosome folding are still elusive . What controls the dynamic folding of chromosomes ? Such fundamental insight can be captured and explored in predictive computational models [18]–[21] . Models help to identify critical experiments and make sure that proposed mechanisms are physically and thermodynamically sound . Ideally , models should correctly reproduce the hierarchical architecture of interphase chromosomes , the probabilistic aspects of chromatin folding , as well as the structural transitions that chromosomes undergo during mitosis , meiosis and cell differentiation . Importantly , models should be able to make predictions that can be experimentally tested . Since chromosomal fibres are long flexible one-dimensional structures , polymer models are a good first approximation for chromosomes . Recently , a variety of such models have been proposed based on information from genome contact maps [20] , [22]–[24] . So far , many models have not yet been thoroughly tested experimentally , which makes it difficult to assess their quality . Recently , the Dynamic Loop polymer model ( DL model ) has been proposed , based on systematic 3D fluorescent in situ hybridisation ( FISH ) measurements on primary human fibroblasts , in combination with basic polymer physics [16] , [25] . The DL model is based on first principles and its main parameter is the looping probability , describing the chance that two non-adjacent monomers of the polymer make contact , i . e . form a loop . The DL model shows that a linear polymer with randomly positioned loops with a broad length distribution correctly recapitulates three basic aspects of interphase chromatin: ( i ) the formation of chromosome territories , due to the entropy-driven segregation of looped chromosomes , ( ii ) the presence of discrete sub-chromosomal domains that differ in chromatin compaction due to local differences in looping frequency , and ( iii ) the considerable cell-to-cell variation in chromatin folding [25]–[28] . In the DL model diffusional behaviour of the chromatin fibre , which is considerable in interphase nuclei [29] , drives the dynamic formation of loops . Importantly , the DL model is compatible with results of genome-wide intra-chromosomal contact mapping experiments [30] . The model indicates that the distribution of loops along the chromosome controls chromatin folding . To unveil further principles of chromosome folding we now test predictions of the DL model by manipulating the chromosome looping frequency and measuring the effect on chromosomal structure . Two abundant nuclear proteins that are involved in chromatin looping are CTCF and cohesin [31]–[33] . Their binding sites on the genome in part co-localise indicating that they functionally cooperate [34] , [35] . Reducing the levels of CTCF and Rad21 , an essential component of the cohesin complex , in the cell has been shown to reduce the number of chromatin loops [36]–[38] . Here we analyse by quantitative 3D FISH the effect of CTCF and/or Rad21 knockdown on chromosome structure in human primary G1 fibroblasts . The DL model , similar to most other polymer models , predicts that upon reduction of the number of chromatin loops chromosomes become less compact . In striking contrast to this prediction our experiments show that they become considerably more compact after knocking down CTCF and Rad21 . These observations put major constraints on polymer models that aim to recapitulate the behaviour of interphase chromosomes . We show that the DL model can be adapted to correctly describe the observed compaction . Systematic model simulations strongly suggest that the looping regime , i . e . loop frequency along the chromosomal fibre and loop size distribution , is the key variable that controls chromosome folding . The cell is able to manipulate chromosome folding by regulating the activity and concentration of looping proteins , such as CTCF and cohesin . The model makes specific predictions that are supported by recent experiments of others , underscoring the potential of the DL model to explore underlying principles of chromosome folding . Predictions of the adapted DL model about structural transitions that chromosomes undergo during cell differentiation , mitosis and meiosis are briefly discussed .
The expression levels of CTCF and of cohesin were reduced by siRNA-mediated gene knockdown . To interfere with cohesin function we depleted its kleisin subunit Rad21 . In the absence of this subunit none of the other cohesin subunits bind to chromatin [34] . The decrease in protein levels was quantified by Western blotting of nuclear fractions . Figure 1 shows that knockdown of CTCF and cohesin individually and in combination results in a decrease of ∼80% of their cellular concentration . A variety of studies have shown that such decrease in CTCF and cohesin significantly reduces chromatin looping [39] , [40] , but has no significant effect on cell cycle progression [41] . All experiments were carried out with human primary female fibroblasts to avoid possible effects of immortalisation . Depletion of CTCF and/or cohesin had no measurable effect on the cellular and nuclear morphology and the size of the nucleus during the time course of the experiments , i . e . up to 72 hours after siRNA transfection ( Figures S1 and S2 ) . Also the viability of cells was not affected and the percentage of apoptotic cells among cells analysed did not change , as demonstrated by AnnexinV staining ( Figure S3 ) The relationship between the mean square physical distance ( MSD ) <R2> and the genomic distance g between FISH probes was determined for the same genomic regions as studied earlier in establishing the DL model , i . e . the q-arms of chromosomes 1 and 11 [16] . We concentrated on an about 3 Mb size gene-rich region gene-poor region of similar length on chromosome 1 and on a 27 Mb and a 70 Mb region on the q-arms of chromosome 1 and 11 , respectively . The latter spans the complete q-arm of chromosome 11 . Figures 2A and 2B show the analysed regions and the distribution of the BAC ( bacterial artificial chromosome ) probes used for FISH labelling . The BACs used in this study and the number of FISH probe-pairs analysed is listed in Table S1 . Figure S6 shows the positions of the BACs on the genomic maps of the relevant parts of chromosomes 1 and 11 and relates them to the human transcriptome map [42] , replication domains [43] and the HiC topological domains of similar cell types [7] . Only cells in G1 were used for these analyses . In the randomly growing cell culture S-phase cells were identified based on their incorporation of bromodeoxy uridine ( BrdU ) during DNA replication and G2 cells could be recognised by their significantly larger size [16] , [44] . Figures 2C and D show the relationship between the MSD of two FISH probes and their genomic distance in two about 3 Mb regions on chromosome 1q , one being gene-rich ( green ) and the other gene-poor ( red ) . Results from nuclei of control cells and of cells in which CTCF and cohesin were knocked down either individually or simultaneously are shown . Simultaneous knockdown of the two proteins resulted in considerable chromatin condensation , as shown by the decrease of the MSD plateau levels . The maximum average distances between the FISH probes decreased from about 2 . 7 µm to 1 µm for the gene-rich regions and from about 0 . 7 µm to close to 0 . 5 µm for the gene-poor regions . Knockdown of CTCF and cohesin individually had a considerably smaller effect . Control experiments using non-target siRNA did not reveal significant changes in chromatin compaction ( Figure S4 ) . Chromatin condensation is particularly evident for gene-rich areas , which are relatively open in control cells , and is smaller for the gene-poor regions that already are relatively compact in untreated cells . Figures 2E and F show the same type of measurements for the longer genomic stretches: 27 Mb of chromosome 1q and 70 Mb of chromosome 11q . At these longer distances the decrease of CTCF or of cohesin concentration reduced the maximum probe distance from about 3 µm to around 2 µm for chromosome 1q , indicating chromatin condensation , as found for shorter distances . Knockdown of both proteins simultaneously reduced the average distance further to close to 1 µm , indicating a three-fold compaction . For the 70 Mb region of chromosome 11q the effect of knocking down of CTCF and cohesin individually is less pronounced . The maximum average probe distance is about 3 µm , similar to the value for the chromosome 1q region in control cells . Simultaneous knockdown of CTCF and cohesin reduced this value to between 2 and 2 . 5 µm . What causes this quantitative difference in the behaviour of the two chromosomes is unclear . The average frequency of CTCF binding sites in both genomic regions is similar [45] . Importantly , depletion of the two looping proteins does not result in decondensation of chromatin , in contrast to what is predicted by most chromatin-inspired polymer models , including the DL model . Instead , after CTCF and cohesion depletion the opposite is observed , i . e . condensation of chromatin . Although depletion of CTCF and cohesin causes the analysed regions to become more compact than in control cells , there are no evident changes in chromatin density when comparing DAPI staining signals of these cells ( Figure S1 ) . To find out if compaction of chromatin in CTCF and cohesin depleted cells results in large-scale chromatin rearrangements in in the nucleus that remain unseen in DAPI staining , we analysed the radial position of probes used for 3D FISH measurements in the nucleus . No significant changes in radial distribution of sites on the q-arms of chromosomes 1 and 11 were observed ( Figures 2G and H ) , indicating that depletion of CTCF and cohesin has no effect on overall nuclear organization . The position of a locus in nuclear space and its compactness are correlated: gene-rich and open chromatin regions on average are located closer to the nuclear centre , whereas compact and gene-poor regions are located closer to the nuclear periphery [44] , [46] , [47] . Figure 2G shows that the analysed regions on chromosome 1q have approximately the same radial position . On chromosome 11 two well-defined regions can be identified ( Figure 2H ) : the probes that label the centromere-proximal 0–11 Mb of the analysed region are closer to the nuclear centre , while the probes marking the 23–76 Mb area are located more towards nuclear periphery . This 23–76 Mb area of chromosome 11 is closer to the nuclear periphery compared to the probes on chromosome 1 . In combination with measurements of relative volume of these regions carried out previously [44] . these results indicate that in control cells the 23–76 Mb domain of chromosome 11 has a more compact structure than the 0–11 Mb region on chromosome 11 and the analysed regions on chromosome 1 . Recently , we proposed a simple polymer model that explains the confinement of an interphase chromosome to its chromosome territory . In this DL model the chromosomal fibre is represented as a self-avoiding random walk polymer forming probabilistic intra-polymer contacts between non-adjacent monomers [25] . As a consequence , loops with a broad size distribution are formed . The main model parameter is the looping probability ( p ) , which is a measure for the probability that a bond ( loop ) is formed between two non-adjacent monomers is formed . Due to the proximity of monomers , short-range loops are formed more frequently than long-range ones , yielding a loop size distribution comparable to that observed in HiC experiments [30] . In the DL model a decrease in looping probability results in an increase of the volume occupied by the polymer , i . e . decondensation . Interestingly , our experimental results show that chromatin in interphase nuclei behaves in an opposite way , i . e . a decrease in chromatin looping due to depletion of CTCF and cohesin results in compaction . In the DL model decreasing the looping probability affect loops of all lengths . However , in cells CTCF seems involved preferentially in mediating short-range loops , i . e . below 1 Mb length [48] . To explore whether the observed chromatin compaction may be caused by the predominant depletion of short-range loops , we modified the DL model by imposing different looping probabilities for short-range and long-range looping ( pshort and plong , respectively ) . In simulations with this adapted DL model we used polymers with a total length of 1050 monomers ( for further details see Materials and Methods section ) . Short-range loops were defined as those spanning 50 monomers or less and long-range loops as those spanning 51 monomers or more . For comparison , for human chromosomes in sizes 50–250 Mb the cut-off between short- and long-range loops would be approximately 2 . 5–12 . 5 Mb . Subsequently , we calculated the MSD versus contour distance relationship for a matrix of combinations of short- and long-range looping probabilities . Figure 3A illustrates the typical compact conformation of an adapted DL polymer with high pshort ( 0 . 12 ) and low plong ( 0 . 04 ) probabilities . The colour code labels the monomers along the polymer according to the visible spectrum along the polymer . The inset in panel A displays the conformation for the same pshort ( 0 . 12 ) after plong has been set to zero , showing that a high probability of the short-range interactions ( pshort ) in the absence of long-range interactions results in a uniform thick fibre . Figure 3B shows a configuration of the DL polymer with low pshort ( 0 . 04 ) and high plong ( 0 . 12 ) probabilities , showing a more chaotic folding of the polymer compared to Figure 3A . Figures 4A and B exhibit examples of the calculated relationships between the MSD and the contour distance , covering a range of pshort values ( 0 . 02–0 . 05 ) , while keeping plong constant ( 0 . 03 ) ( Figure 4A ) , and of plong ( 0 . 02–0 . 05 ) values at constant pshort ( 0 . 03 ) ( Figure 4B ) . The plateau level MSD value is used as a measure for the overall compaction of the polymer: a lower MSD values indicates a stronger compaction . Remarkably , these simulations show that decreasing the long-range looping frequency plong results in expansion of the polymer , while decreasing pshort leads to compaction . The heat map of Figure 5A displays the MSD plateau level of the polymer as a function of the pshort - plong parameter space . The arrows depict the effect of decreasing the pshort value at constant plong and of decreasing the plong value at constant pshort , underscoring that compaction occurs only if the short-range looping probability is decreased , whereas a reduction of plong results in de-compaction . The qualitative explanation for this behaviour is explained in the Discussion . Evidently , the DL-model model analysis does not intend to reflect the precise behaviour of interphase chromosomes . It shows how different looping regimes may dramatically affect large scale chromosome folding . As can be seen in Fig . 5 the behaviour is qualitatively the same over a large range of the plong and pshort parameter landscape . Polymer modelling shows that the chromatin compaction we observe after knocking down CTCF and cohesin may be attributed to a specific reduction of short-range loops . The adapted DL model does not incorporate what seems to be a key aspect of chromatin architecture , namely the presence of topological sub-chromosomal domains . Several independent lines of evidence , including replication timing , HiC analysis , clustering of epigenetic marks and electron microscopy , indicate that chromosomes of higher eukaryotes consist of arrays of topological domains in the size 0 . 5 to 5 Mb range [7]–[10] . In terms of the adapted DL model parameters this means that short-range looping interactions are locally clustered at many positions along the polymer , rather than evenly distributed as assumed in the adapted DL model . We wondered whether a polymer consisting of an array of topological domains would recapitulate the chromatin compaction we observe in our experiments when reducing the looping probability . To force the adapted DL model polymer to mimic topological domains we assumed that the polymer consists of regions that alternatingly contain a 50 monomer domain with high short-range looping probability ( pshort ) and one that is devoid of short-range loops . For simplicity , in this domain-adapted DL model the two types of regions were made equal in length ( 50 monomers ) resulting in 10 highly and 11 lowly looped domains in the model polymer of again in total 1050 monomers . Long-range loops were allowed between the compact domains only with a probability plong . Figure 3C shows the characteristic domain-like structure of the domain-adapted DL polymer under conditions of high short-range looping probability ( pshort = 0 . 16 ) in combination with a low long-range probability ( plong = 0 . 02 ) . As expected , compact domains ( red in Figure 3C ) are formed and connected by non-structured linker stretches ( blue ) . When lowering the looping probability in the domains to pshort = 0 . 02 and increasing the looping between the domains ( plong = 0 . 16 ) , the less compact individual domains ( red ) cluster due to long-range looping ( Figure 3D ) . The heat map of the variation of polymer compaction in the pshort - plong parameter space is shown in Figure 5B . Comparison of Figures 5A and B shows that the behaviour of polymers with and without domain-structure is very similar after decreasing short-range or the long-range looping probabilities . The former induces compaction and the latter results in expansion of the polymer . Evidently , introducing discrete topological domains in the adapted DL polymer model does not affect its capacity to show compaction after decreasing the short-range looping probability pshort .
The 46 human chromosomes , in total consisting of 5 cm nucleosomal fibre , are packed in an interphase nucleus with a diameter in the order of 10 µm . Intra-chromosomal interactions forming chromatin loops play an important role in compacting interphase chromosomes . At the same time loops are a key component in gene regulation , for instance because they allow physical contacts between distant regulatory elements , such as promoters and enhancers [11] . Our understanding of basic principles of chromosome folding and how these affect gene regulation and other genomic functions is still limited . Polymer modelling based on experimental data sets begins to unveil general aspects of large scale ( supra Mb ) chromosome folding . The Dynamic Loop ( DL ) model shows that intra-chromosomal loops can explain the relatively compact nature of chromosome territories in interphase nuclei and why they intermingle only to a limited extent . Also , variations in looping frequency along the chromosomal fibre result in sub-chromosomal domains with different compaction [25]–[28] . All polymer models that are based on chromosome looping [16] , [49] predict that a decrease in the number of loops results in chromosome expansion . Here we test this prediction in primary human fibroblasts by reducing the concentration of two abundant proteins that are involved in loop formation , i . e . CTCF and cohesin . Knocking down individually or simultaneously the CTCF gene and a gene coding for Rad21 , an essential component of cohesin , reduces their concentration to about 20% of the initial levels ( Figure 1 ) . Others have shown that such decrease of CTCF and cohesin concentrations indeed results in a decrease in the number of chromosomal loops [36]–[38] . Remarkably , instead of the predicted expansion we observe a significant condensation of chromosome , in full contrast to the model predictions ( Figure 2 ) . These results confirm the importance of looping in controlling how chromosomes are folded . At the same time they put severe constraints on looping-based polymer models that should recapitulate the configuration of interphase chromosomes . Because the part of the binding sites on the genome of CTCF and cohesin co-localise [34] , [35] , it can be expected that knocking down of each of these proteins individually will affect similar effects , whereas their simultaneous knock down has a larger effect . This is exactly what is observed ( Figure 2C–F ) . To explain our observations we started from the DL model , which assumes that loops with a broad length distribution are randomly positioned along the chromosomal fibre . Systematic model simulations showed that selectively decreasing the frequency of short loops , without affecting longer loops , recapitulates the chromatin compaction observed in our experiments after decreasing the looping frequency by reducing the cellular concentration of CTCF and cohesin ( Figures 4A , B and 5A ) . To do so we introduced separate parameters for the probability of short loops and of long loops ( shorter than 5% of the polymer length and 5% or longer , respectively ) . Systematic simulations exploring variations of polymer compaction in the short-range vs . long-range looping probabilities show that only in a specific part of the parameter space a decrease of short-range looping frequencies results in significant condensation of the polymer . In contrast , decrease of long-range looping frequencies always results in decondensation . The observed chromatin compaction after decreasing the number of loops is counter-intuitive . The rationale is that reducing the number of short loops leads to expansion of the polymer on the short scale , resulting in an increase in volume occupied by the polymer . However , at the same time the decrease of the number of short loops leads to reduced entropic intra-polymer repulsion [27] . Consequently , parts of the polymer that are in close spatial proximity , e . g . due to the long-range interactions , intermingle stronger , which in turn makes the formation of a long-range loops more probable . This means that for constant plong the number of long-range loops ( nlong ) indirectly increases when reducing the number of short-range loops ( Figure 6 ) . As the entropy-driven volume reduction , due to this intra-polymer intermingling , exceeds the increase of volume due to the short-range polymer swelling , the volume occupied by the complete polymer decreases , i . e . it condenses . It may be argued that after de-compaction of the polymer at low length scales the probability of long range interaction increases because more such interaction sites become available . As can be seen in Fig . 5 such increase of plong would further enhance the compaction process . The above considerations predict that CTCF and cohesin are mainly involved in short-range looping . In support of this , Handoko et al . [48] showed that CTCF is predominantly engaged in the Mb-range organisation of chromosomes . Cohesin and CTCF are known to cooperate in the spatial organisation of chromosomes [34] , [35] . In line with this , decreasing the activity of CTCF and cohesin preferentially affects looping at short-range , below Mb [36]–[38] . There is ample experimental evidence that , in contrast to what the adapted DL model assumes , the distribution of short-range loops is not random . Rather , short-range looping sites seem to cluster at multiple sites along the chromosome , resulting in topological domains in the 0 . 5–5 Mb size range [7]–[10] . Examples are distinct sub-chromosomal domains that differ in replication timing and in epigenetic state [2] , [9] , [12]–[15] . To mimic these topological domains in the adapted DL model , the short-range loops were clustered at multiple domains along the polymer , intervened by areas devoid of looping . Model simulations show that this redistribution of loops has little effect on the condensation-decondensation behaviour of the polymer ( Figure 5B ) . Evidently , the domain-adapted DL polymer model reproduces the condensation of the polymer after reducing the number of short loops similar to the adapted DL model without domains ( Figure 5A ) . Interestingly , after knockdown of CTCF and cohesin the Mb-size topological domains do not dissolve , indicating that at least part if the intra-domain loops remain intact . Zuin et al [36] showed that the number of interactions between chromatin from different topological domains , i . e . long-range loops , increases . As can be seen in Figure 5 , the adapted DL model predicts that in addition to the entropic effects the increase of the frequency of large loops further contributes to polymer compaction . In recent years , several polymer models have been proposed for the folding of interphase chromatin . The ‘Strings and Binders Switch’ ( SBS ) model of Barbieri et al . [49] ) , which assumes a diffusible component responsible for loop formation by linking two monomers of the polymer , is a special case of the DL model [16] , [25] . In the DL model the properties of such ‘binders’ are incorporated in the looping probability parameter . As expected , the SBS model correctly mimics the formation of chromosome territories in the same way as the DL model . Another class of polymer models describes chromatin as a fractal polymer [22] , [23] . This type of model was proposed to explain the chromatin folding based on HiC contact maps in the 0 . 5–10 Mb range [22] , [50] . Since these models do not involve polymer looping-related parameters in their present form , they do not allow making predictions about how changes in the looping regime relate to changes in compaction . Systematic quantitative FISH measurements on chemically fixed cells , as used in this and other papers [16] , [49] , [51] , do not support fractal models . We can only speculate about the biological relevance of chromatin compaction after decreasing the concentration of the looping proteins CTCF and cohesin . Conceivably , this may be a step in the transition to metaphase chromosomes at the onset of cell division . A recent genome-wide interactome study shows that there are marked changes in intra-chromosomal contacts in metaphase chromosomes compared to the situation in interphase [52] . Whatever the mechanism of metaphase chromosome formation is , it most likely involves the disappearance of long-range loops . The inset in Figure 3A suggestively shows that complete abolishment of long-range loops results in metaphase-like structures in a state before axial condensation . Taken together , the adapted DL model is successful in predicting various key properties of interphase chromosomes: ( i ) distinct and poorly intermingling chromosome territories , ( ii ) local differences in chromatin compaction along the chromosomal fibre , and ( iii ) chromosome compaction after reducing the looping frequency . The model shows that chromosomal looping constitutes the basis for this behaviour . Relatively subtle changes in the balance between short-range and long-range looping probabilities lead to changes in local and overall chromosome folding and compaction , probably including the interphase-metaphase transitions .
Human primary female fibroblasts ( 04–147 ) were cultured in DMEM containing 10% FCS , 20 mM glutamine , 60 µg/mL penicillin and 100 µg/mL streptomycin ( Gibco , Life Technologies Corporation , Carlsbad , CA , USA ) . Cells were used up to passage 25 to avoid effects related to senescence . siRNA transfections were performed using lipofectamine RNAiMAX ( Invitrogen , Life Technologies Corporation , Carlsbad , CA , USA ) according to the reverse transfection protocol of the manufacturer . siRNAs used are listed in Table S2 . The transfection efficiency of siRNA was estimated with the BLOCK-iT Alexa FluorRed Fluorescent Control ( Invitrogen , Life Technologies Corporation , Carlsbad , CA , USA ) in three independent experiments . In each experiment 100 nuclei were scored based on DAPI signal , 99% of which showed red fluorescence and therefore had been transfected successfully . Knockdowns were verified by Western blot . Cells were resuspended in lysis buffer ( 10 mM HEPES , 10 mM KCl , 1 , 5 mM MgCl2 , 0 , 34M sucrose , 10% glycerol , 1 mM DTT , complete protease inhibitor cocktail ( Roche , F . Hoffmann-La Roche Ltd , Basle , Switzerland ) ) containing 0 . 1% Triton X-100 , incubated 10 min on ice and centrifuged 5 min at 1300×g . Pellets were resuspended and washed once . We used anti-CTCF polyclonal antibody ( 07-729 ) from Millipore ( Millipore , Temecula , CA , USA ) and anti-Rad21 polyclonal antibody ( abb992 ) from Abcam ( Abcam plc , Cambridge , UK ) , both in 1∶1000 dilutions to detect CTCF and Rad21 protein levels . As a control antibody against β-Actin was used - anti-β-Actin ( A1978 ) from Sigma-Aldrich ( Sigma-Aldrich , Saint Louis , MO , USA ) in 1∶5000 dilution . Secondary antibodies: AP-conjugated Anti-Rabbit IgG ( 111-055-003 ) and AP-conjugated Anti-Mouse IgG ( 115-055-003 ) both from Jackson ( Jackson ImmunoResearch Laboratories Inc , West Grove , PA , USA ) were used in dilution 1∶5000 . Signals were quantified using ImageJ software ( http://rsb . info . nih . gov/ij/ ) . The fraction of apoptotic cells on slides was estimated by FITC Annexin V ( BD Pharmingen , Becton , Dickinson and Company , Franklin Lakes , NJ , USA ) staining . 100 cells of control , CTCF knockdown , Rad21 knockdown and CTCF-Rad21 knockdown populations were selected by DAPI signal , the fraction of Annexin V positive cells was measured in this population . The same procedure was repeated for three independent siRNA knockdown experiments . Results obtained by AnnexinV staining were confirmed by the Tunel apoptosis assay ( Promega , Madison , WI , USA ) . BACs were selected from the BAC clones available in the RP11-collection at the Sanger Institute . Genomic distances were defined as the distance between centres of the BACs . BAC DNA was isolated using the Qiagen REAL prep 96 kit ( Qiagen , Qiagen Benelux BV , Venlo , Netherlands ) . Nick-translation was used to label the probes , either with digoxigenin or biotin ( Roche , F . Hoffmann-La Roche Ltd , Basle , Switzerland ) . FISH was carried out as described in [44] . FITC-conjugated antibodies ( Roche , F . Hoffmann-La Roche Ltd , Basle , Switzerland ) and Cy3-conjugated streptavidin ( Jackson ImmunoResearch Laboratories , Inc . , West Grove , PA ) were used to visualize the hybridization signals . DAPI ( 4′ , 6′-diamidino-2-phenylindole ) ( Sigma-Aldrich , Saint Louis , MO , USA ) was used to outline the cell nucleus . To exclude S-phase cells from the analysis the culture was incubated with bromodeoxyuridine ( BrdU ) to label replicating cells 30 min before fixation , followed by immunolabelling in combination with FISH labelling as described in [44] . Twelve-bit 3D images were recorded using Nikon A1R confocal laser-scanning microscope ( Nikon , Tokyo , Japan ) equipped with a 100×/1 . 49 NA Apo TIRF DIC objective ( Nikon ) , using a diode laser at 405 nm , an Ar laser at 488 nm and a diode-pumped solid-state laser at 561 nm to excite DAPI , FITC and Cy3 , respectively . Fluorescence was detected with the following bandpass filters: 425–475 nm ( DAPI ) , 500–550 nm ( FITC ) and 570–620 nm ( Cy3 ) . Images were scanned with a voxel size of 60×60×100 nm . All confocal images were subject to deconvolution using Huygens Professional 3 . 7 software ( Scientific Volume Imaging , Hilversum , The Netherlands ) using the measured point spread function ( PSF ) for each channel and the classical maximum likelihood estimation algorithm . The PSF was obtained by imaging Tetraspeck Fluorescent Microsphere Standards with a diameter of 200 nm ( Invitrogen ) . The signal to noise ratios and background intensities were estimated for each channel and averaged from several images . These values were used as a standard for batch processing deconvolution of all the image stacks . Automated image analysis was carried out on deconvolved datasets with the ARGOS software ( http://homepages . cwi . nl/~wimc/argos ) to identify nuclear sites labelled by BACs and to compute their 3D position in the nucleus as described in [44] . Chromatic aberration was measured via Tetraspeck Fluorescent Microsphere Standards with a diameter of 200 nm ( Invitrogen ) and corrected for in the analysis . After background subtraction , images were treated with a bandpass filter to remove noise . Subsequently , images were segmented and ensembles of interconnected voxels were regarded as the site labelled by a BAC . The centre of mass was calculated for each labelled site at a sub-voxel resolution and 3D distances between BACS were measured . To estimate the systematic measuring error we hybridized cells with a mixture of the same BAC marked with two different fluorophores and measured the distances between the two signals . Accuracy of measurements was better than 50 nm in three dimensions . The radial nuclear position pn ( pn = ro/rn ) of BAC probe indicates positioning of the centre of gravity of a FISH labelled site on the line drawn from the centre of gravity of the nucleus to nuclear envelope . pn was calculated as previously described [44]: the distance from the centre of gravity of a BAC probe and the centre of the nucleus ( ro ) was divided by the length of a line from the nuclear centre to the nuclear envelope through the centre of gravity of the BAC probe ( rn ) . pn value close to 1 indicates positioning of the BAC probe at nuclear periphery while pn value close to 0 indicates its positioning close to the centre of gravity of the nucleus . Chromatin fibres are in general modelled as un-branched polymer chains . The modelled polymers are coarse-grained versions of the real chromosomes , where each monomer represents a stretch of DNA . If the length of this stretch is much larger than the persistence length of the chromatin fibre , we can expect the polymer to be totally flexible and are allowed to neglect the influence of the local bending rigidity . With a persistence length below 250 nm [53] and a fibre packing , where a length of 10 nm = 1 kb , the monomers should represent DNA stretches with a length of at least 50 kb . In this study , we performed Monte-Carlo simulations [54] with an implementation of the Dynamic Loop model to generate chromosomal conformations . For the polymer chains that represent the interphase chromosomes we used the well-tested bond fluctuation method [55] , [56] to investigate their structure , as well as their dynamics . In the simulations a monomer of the polymer chain is randomly selected and , if possible , randomly moved to one of its nearest neighbours on the lattice . Excluded volume interactions are taken into account by preventing a lattice site to be occupied by more than one monomer . When simulating N monomers we define one Monte-Carlo step ( MCS ) to correspond to N moves , i . e . on average each monomer is translated once during a MCS . Due to the fact that the polymer conformations only exhibit slight changes from one MCS to the next , the time span when two conformations can be considered to be independent must be determined . Therefore , we calculate the autocorrelation function of the polymer's squared radius of gyration Rg2 ( t ) which is a measure for structural correlation . We obtain the estimated autocorrelation time τac by applying an exponential fit to the autocorrelation function . Finally , we set 5τac MCS as the time span above which two conformations are expected to be independent . For the interactions between the chromatin fibres we used the Dynamic Loop ( DL ) model [25] . In this model two monomers that are non-adjacent along the fibre are only allowed to interact if they are in spatial proximity , i . e . if the distance between them is below a certain cutoff . When two monomers i and j approach each other due to diffusional motion and the cutoff condition is fulfilled , a bond can be established between them with a certain probability pbond , ij , the binding or looping probability . In case the bond is formed , a lifetime tbond chosen from a Poisson distribution with mean Tbond is assigned to it , determining when the bond dissociates . Hence , the bonds can frequently change during the course of the simulation , which mimics the effect of the highly dynamic DNA-DNA interactions . In contrast to this , the bonds along the backbone of the polymer are fixed and cannot break open . For the simulations with the domain-adapted DL and with the adapted DL model , the chromosome-representing polymers have a length of 1050 monomers . We have to simulate thousands of independent chromosome conformations for each parameter set to get statistically reliable values for the different chromosome properties . Hence , the polymer length is limited to ∼1000 monomers in order to finish the simulations in a manageable timeframe . We set the length to exactly 1050 monomers because in the domain-adapted DL model , the highly active as well as the lowly active domains have a size of 50 monomers and the polymer consists of 21 domains ( 11 lowly active interspersed by 10 highly active regions ) . We also use 1050 monomers for the adapted DL model to ease comparison with the results from the domain-adapted DL model . The size of the Monte-Carlo lattice is set to a large value ( L = 500 ) to avoid interactions caused by periodic boundary conditions . The mean of the Poisson distributions that is used to determine the bond lifetime is set to 8000 MCS . In the adapted DL model , pbond , ij can take 2 different values , namely pshort if |i-j|< = 50 and plong otherwise . In the domain-adapted -DL model , the bond formation probabilities pbond , ij between the monomers are set to pshort if monomers i and j are inside the same domain , or to plong if i and j belong to different domains . We start each simulation by creating a random self-avoiding walk ( polymer where bonds are only established between adjacent monomers . We continue with a first equilibration run of 108 MCS where only homogeneous bond formation is allowed . As pbond is set to be constant along the fibre , we obtain a homogenous starting conformation , meaning a polymer with a uniform structure . In a second equilibration run of 108 MCS we allow heterogeneous looping as defined by the domain-adapted DL and the adapted DL model . When all equilibration runs are finished , the main simulation starts and the chromosomal conformation is saved every 107 MCS . This is done until at least 1000 independent conformation are generated .
|
Folding of chromosomes in interphase nuclei of higher eukaryotes is a key element in regulating gene expression . The mechanisms that control chromatin folding are largely unknown . We have shown earlier that looping is a fundamental aspect of large-scale chromatin structure . Two abundant looping proteins are known: CTCF and cohesin . Here we combine quantitative fluorescent in situ hybridisation experiments in human cells with polymer modelling to unravel mechanisms of chromatin folding . We show that chromatin becomes more compact after depletion of looping proteins . This is remarkable , since polymer models describing chromatin predict decompaction . We present a polymer model that shows that specific biologically relevant looping regimes give rise to this behaviour . Importantly , chromosome conformation capture studies of mammalian chromatin support such looping regimes . Our results indicate that the local and overall compaction of the chromatin is defined by a subtle balance between short and long range loops; this may explain cell cycle and genome activity dependent structural transitions of chromatin .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"systems",
"biology",
"imaging",
"techniques",
"cell",
"biology",
"computational",
"techniques",
"microscopy",
"mathematical",
"and",
"statistical",
"techniques",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"biophysics",
"research",
"and",
"analysis",
"methods"
] |
2014
|
Depletion of the Chromatin Looping Proteins CTCF and Cohesin Causes Chromatin Compaction: Insight into Chromatin Folding by Polymer Modelling
|
Drosophila melanogaster larvae irradiated with doses of ionizing radiation ( IR ) that kill about half of the cells in larval imaginal discs still develop into viable adults . How surviving cells compensate for IR-induced cell death to produce organs of normal size and appearance remains an active area of investigation . We have identified a subpopulation of cells within the continuous epithelium of Drosophila larval wing discs that shows intrinsic resistance to IR- and drug-induced apoptosis . These cells reside in domains of high Wingless ( Wg , Drosophila Wnt-1 ) and STAT92E ( sole Drosophila signal transducer and activator of transcription [STAT] homolog ) activity and would normally form the hinge in the adult fly . Resistance to IR-induced apoptosis requires STAT and Wg and is mediated by transcriptional repression of the pro-apoptotic gene reaper . Lineage tracing experiments show that , following irradiation , apoptosis-resistant cells lose their identity and translocate to areas of the wing disc that suffered abundant cell death . Our findings provide a new paradigm for regeneration in which it is unnecessary to invoke special damage-resistant cell types such as stem cells . Instead , differences in gene expression within a population of genetically identical epithelial cells can create a subpopulation with greater resistance , which , following damage , survive , alter their fate , and help regenerate the tissue .
The ability to regenerate is critical for tissue homeostasis in adult organisms . Many tissues such as the gut and the skin suffer constant environmental insults that result in cell death , requiring tissue-specific stem cells to proliferate and compensate for cell loss . Two key characteristics of stem cells , namely , greater resistance to killing compared to other cells and the ability to contribute to regeneration , are also ascribed to cancer stem cells or cancer initiating cells . Specifically , tumors and blood cancers are hypothesized to contain a small population of cells with greater ability to initiate new tumors than the rest . Such cancer-initiating cells would help replenish the tumor , after therapy for example , or to produce new tumors at distant sites as metastases . Cancer-initiating cells have been identified for different types of cancers , typically by virtue of differential cell surface markers that correlate with the ability of cells with such markers to initiate cancers or tumors in animal models ( for example [1 , 2] ) . Where tested , cancer-initiating cells show greater resistance to killing by cancer therapy agents , such as radiation and chemotherapy [3 , 4] . Understanding molecular mechanisms that underlie the increased resistance of cancer-initiating cells and their ability to contribute to regrowth is essential for optimizing anticancer treatments . In this regard , genetically tractable model systems in which cells show resistance to death induced by cytotoxic agents and contribute to regeneration would be valuable tools . Imaginal discs in Drosophila larvae are precursors of adult organs . Larval imaginal discs are composed of a layer of columnar epithelial cells covered by a layer of peripodial squamous cells . Imaginal discs lack a dedicated stem cell population yet regenerate efficiently . For example , surgical ablation of up to a quarter of a leg disc still allows complete regeneration [5] . Likewise , irradiation of larvae with X-ray doses that kill up to 50% of the cells is still compatible with the production of a viable adult [6] . In studies in wing and eye imaginal discs , dying cells are found to produce mitogenic signals that result in non-autonomous activation of cell proliferation locally , up to 5 cell diameters away from the dying cells . These signals operate through JNK , Wg , and Dpp in the wing disc and Hh and EGFR in the eye disc ( reviewed in [7] ) . Increased proliferation of nearby cells could help replace cells lost to cell death and , therefore , help regenerate the disc . In studies of mitogenic signals that emanate from dying cells , cell death was often confined to marked clones in order to detect effects on the neighbors . In other studies in which death-inducing stimuli was applied globally , by irradiating the whole larva , for example , the resulting cell death was not homogenous across the disc . In other words , cells within the continuous single-layer epithelium of imaginal discs show unequal propensity to die ( for example , see [8 , 9] ) . What makes some cells die while their neighbors survive in the face of identical external insult remains an open question , which may be directly relevant to , for example , radiation treatment of tumors . We report here the identification of a subpopulation of cells in Drosophila melanogaster larval wing disc , which , we found , shows consistent resistance to ionizing radiation ( IR ) - and drug-induced apoptosis . These cells reside in the region of the wing disc that normally differentiates into the hinge region of the adult wing . We found that these cells are protected from IR-induced apoptosis by Wg and signal transducer and activator of transcription ( STAT ) and participate in tissue regeneration after radiation damage . These results illustrate that it is unnecessary to invoke special damage-resistant cell types such as stem cells for regeneration . Instead , differences in gene expression within a population of epithelial cells can create a subpopulation to fulfill this role . We discuss ways in which this situation may be analogous to human cancers .
It is evident in many previous studies that there are domains within the wing imaginal disc with different propensity to undergo IR-induced apoptosis ( for example , [8 , 9] ) . In our experiments , apoptosis was detected by immunostaining for cleaved caspases Drice or Dcp1 , or by TUNEL assay ( Fig 1 and S1 Fig ) , in wing discs from feeding third instar larvae . For example , IR-induced apoptosis was reproducibly robust along the dorsal/ventral boundary of the wing pouch ( between white brackets in Figs 1B–1E , 1J and S1 ) . IR-induced apoptosis was reproducibly low in the dorsal part of the future wing hinge ( between yellow lines in Fig 1B–1E , 1J and S1 ) . This pattern of differential IR sensitivity was seen in wing discs of different sizes present among larvae of similar chronological age ( Fig 1B–1E ) and also for the entire range of X-ray doses tested , 2 , 000–8 , 000R; 4 , 000 and 6 , 000R samples are shown here . IR-resistance in this region spanned through the thickness of the disc , as seen in z-sections ( S1D and S1E Fig ) . IR-resistance was also seen in the ventral-anterior cells of the hinge ( yellow brackets in Fig 1D and 1J ) but was less pronounced and less consistent than in the dorsal hinge . IR resistance in a stripe of cells was also evident in the eye disc ( bracket in Fig 1G ) . In this tissue , resistant cells were found within the morphogenetic furrow , a domain of cells with little or no mitotic activity ( Fig 1H ) and arrested in G1 [10] . This is in agreement with published studies in many systems that show an inverse correlation between mitotic proliferation and DNA damage-induced cell death [11 , 12] . While cell cycle phasing can explain the stripe of IR-resistant cells in the eye disc , it cannot explain the IR-resistance of future-hinge cells in the wing disc because proliferation occurs throughout the whole disc at these stages in development ( Figs 1I and S1G ) . Therefore , we sought other explanations . We focused our analysis on the dorsal hinge region because it showed consistent resistance to IR-induced apoptosis ( Fig 1B–1E and 1J ) , and by different detection methods ( cleaved Dcp1 in Fig 1 , cleaved Drice in S1A Fig , and TUNEL in S1B Fig ) . Because it resembles an inverted “smile , ” we refer to this region as the “frown” hereafter . It is possible that enhanced DNA repair is the reason the frown is resistant to IR-induced apoptosis . We do not favor this possibility because the same region was also resistant to a chemical microtubule depolymerizing agent maytansinol ( S1C Fig ) , which induces robust apoptosis when fed to larvae and adults [13 , 14] . The resistance of frown cells to insults that act by different mechanisms , DNA damage and microtubule de-polymerization , argues for a general anti-apoptotic mechanism . This idea is also supported by the published data that heat-shock-induction of p53 , in the absence of any other insult , was able to induce apoptosis throughout the wing disc except in the frown ( Fig 2E in [15] ) . We conclude that cells of the frown are generally resistant to apoptosis . We note a prior study of spatial differences in IR-induced apoptosis in wing discs [16] . This study focused on the wing pouch in wandering stage larvae , where cell proliferation had decreased and apoptosis was confined to inter-vein regions . Studies here are at an earlier stage in development , during which cells are still actively proliferating . By examining the literature on gene expression in the wing disc , we identified Wg ( Drosophila Wnt-1 ) and STAT92E ( the sole STAT gene in Drosophila , “STAT” hereafter ) as genes whose expression correlates spatially with the frown . Antibody staining showed that Wg-expressing cells of the so-called inner ring ( Wg-IR , arrowheads in Fig 2A and 2I ) lined the ventral ( lower ) edge of the frown ( Fig 2C and 2D ) . The outer ring of Wg expression was less apparent at these stages but abutted the dorsal ( upper ) edge of the frown ( Wg-OR , arrows in Fig 2I ) . Strikingly , the frown overlapped almost entirely with a domain of high STAT activity ( Fig 2E–2H ) as seen by a published GFP reporter under the control of 10 STAT binding sites [17] . This domain is comprised of two folds of epithelial cells immediately dorsal ( up ) to Wg-IR [18] . Thus , the frown includes cells with high STAT activity and borders on cells with high Wg expression . The area of the ventral-anterior hinge that showed increased but variable resistance to IR-induced apoptosis ( yellow bracket in Fig 1D ) also included cells with high STAT activity ( Fig 2E and 2F ) , but the overlap is not as good as in the frown . Because of the proximity between cells with high Wg and cells with high STAT-GFP , we investigated whether each influences the expression of the other . During wing development , the formation of the dorsal hinge region with high STAT activity requires Wg [18–20] . In turn , STAT is required for the growth of this region that results in the spatial separation of Wg-IR and Wg-OR [18 , 20] . Besides these indirect interactions , Wg and STAT pathways are thought to function in parallel and not intersect in the wing disc ( reviewed in [21] ) . We confirmed these finding using conditional inhibition of each pathway ( S2 Fig ) , in agreement with the published results . We next investigated whether Wg or STAT has a role in the IR-resistance of the frown . Because of the known requirement for Wg and STAT in development , we took care to allow wing discs to develop before inhibiting each pathway conditionally ( see Fig 3 legend and Materials and Methods ) . We expressed Axin , an inhibitor of Wg signaling ( S2A Fig ) , in the P compartment ( using en-GAL4 ) or in the A compartment ( using ci-GAL4 ) , in conjunction with GAL80ts for conditional expression . In control discs , A and P halves of the frown showed similar , low levels of IR-induced apoptosis ( Fig 3A–3E ) . Induction of Axin for 24 h induced scattered apoptosis but not in the frown ( Fig 3F and 3I ) . After irradiation , apoptotic cells were observed in the half of the frown that expressed Axin but not in contralateral half ( Fig 3G , 3H , 3J and 3K ) . These results were confirmed using homozygotes of temperature-sensitive wg1-12 mutants [22] , which were incubated at 29°C for 24 h to inactivate Wg before irradiation . wgI-12 discs showed no apoptosis without IR ( Fig 3L ) but showed robust IR-induced apoptosis throughout the disc including the frown ( Fig 3M and 3N ) . Trans-heterozygote of wg1-12 with wgI-16 , an X-ray-induced hypomorphic allele , showed similar results , although attenuation of IR resistance was variable with this allelic combination ( Fig 3O and 3P ) . We conclude that Wg signaling is required for the IR-resistance of frown cells . The results with Axin suggest that the requirement is cell autonomous . We obtained similar results using TCFDN to inhibit Wg signaling . As in the case of Axin , wing discs expressing en-GAL4>UAS-TCFDN showed increased cell death in the pouch but not in the frown ( S3B Fig ) . After irradiation , cells in the posterior half of the frown underwent apoptosis while contralateral anterior cells remained resistant ( S3C and S3D Fig ) . These discs also showed greater intensity of DNA stain in the posterior half ( S3A Fig ) . We do not know the reason , but this effect would be consistent with the published data that TCFDN can drive G1 cells along the D/V boundary in the pouch into S phase [23] . Because the proliferation state itself could affect apoptosis , the TCFDN data is harder to interpret than the data with Axin or wg mutants , neither of which showed altered DNA intensity ( e . g . , Fig 3F , 3I and 3L ) . Depletion of STAT92E using a published RNAi construct ( Fig 3Q–3S ) or a temperature-sensitive allelic combination ( Fig 3T and 3U ) also increased IR-induced apoptosis in the frown , although the effect was variable with the latter treatment . We conclude that STAT92E has a role in IR resistance of the frown . To investigate the consequence of co-depleting Wg and STAT , we combined two protocols that produced the most consistent effect , STATRNAi and Axin . Under identical de-repression conditions for GAL4 , incubation at 29°C for 48 h , Axin induced more apoptosis than STATRNAi in the frown ( compare S4B to S4E Fig ) . Combining the two did not produce more apoptosis than Axin alone ( compare E and K ) . While these results are consistent with STAT and Wg functioning in a single pathway or on the same target , we caution that analysis of genetic epistasis requires complete loss-of-function alleles . Wg and STAT are essential for wing disc development and for larval viability . As such , genetic manipulations we used that allow us to analyze wing discs in third instar larvae may not cause a complete loss of function , even under strong non-permissive conditions . IR-induced apoptosis accompanies a p53-dependent increase in transcripts for pro-apoptotic genes hid , reaper , and sickle [24 , 25] . These encode SMAC/DIABLO orthologs that antagonize Drosophila inhibitors of apoptosis proteins 1 ( DIAP1 ) to allow caspase activation [26] . Genetic manipulations with the potential to reduce the pool of Drosophila SMAC/DIABLO orthologs result in reduced and delayed induction of apoptosis after irradiation . These include heterozygosity for hid and chromosomal deficiencies that remove hid and rpr or hid , rpr , and skl . Ectopic expression of hid or rpr on its own is sufficient to induce apoptosis . These and other results led to a model in which a threshold of pro-apoptotic activity may be reached by an aggregate of Hid , Rpr , and Skl to antagonize DIAP1 and induce apoptosis [26–29] . To gain insight into how cells in the frown resist IR-induced apoptosis , we monitored the expression of transcriptional reporters for DIAP1 , hid and rpr . We chose DIAP1 because STAT92E has been shown to transcriptionally activate DIAP1 to resist IR-induced apoptosis in the wing pouch [30] . In these experiments , the role of STAT in the hinge was not assayed because STAT-/- clones were not recovered in this region [30] . We saw no significant changes in diap1-lacZ reporter expression after irradiation ( Fig 4A and 4B ) . More relevant to this study , DIAP1 protein [30] or transcriptional reporter expression ( this study ) were not different between the pouch and the hinge and , therefore , could not explain the differences in IR resistance . A similar result was obtained using a hid-GFP transcriptional reporter . This reporter is comprised of 2 kb of the hid 5′ regulatory sequences and could be induced by IR in a p53-dependent manner ( Fig 4C and 4D; [31] ) . As in the case of DIAP1 , we did not observe variations in hid reporter expression that could explain differential IR-sensitivities between the pouch and the hinge . rpr-lacZ reporter expression , on the other hand , could explain the differential sensitivity of the pouch and the hinge to IR-induced apoptosis . In un-irradiated discs , rpr-lacZ expression was high in the pouch and parts of the notum but low in the hinge ( Fig 4H , with increased gain in L ) . rpr-lacZ expression increased after irradiation as expected , but the pattern remained the same as in un-irradiated discs ( Fig 4 , compare H and I ) ; that is , rpr expression was low in the hinge with or without IR . Induction of Axin did not alter rpr-lacZ expression in the hinge without IR . This is consistent with the finding that Axin did not induce apoptosis in the frown without IR ( Fig 3F ) . After irradiation , however , we could see increased rpr-lacZ expression in the posterior half of the frown ( arrow in Fig 4J , compared with the contralateral half of the same disc . This is consistent with the ability of Axin to promote IR-induced apoptosis in the frown ( Fig 3H ) . Similarly , rpr-lacZ was also increased after irradiation in wgI-12/I-12 mutants ( S5 Fig ) . Ectopic expression of rpr with patched-GAL4 driven expression of a UAS transgene was sufficient to induce apoptosis in the cells of the frown to similar levels as in the pouch ( Fig 4K ) . Collectively , these results demonstrate that cells of the frown are capable of undergoing apoptosis and suggest that they are prevented from doing so because rpr is repressed , at least in part by Wg signaling . Regions of high STAT activity in the hinge correspond closely to regions of low rpr-lacZ expression , particularly the frown ( Fig 4L–4N ) . Depletion of STAT by RNAi , which was sufficient to promote IR-induced apoptosis ( Fig 3R and 3S ) , however , had little effect on rpr-lacZ expression ( representative disc shown in Fig 4O ) . We conclude that while STAT acts to prevent IR-induced apoptosis in the frown ( Fig 3 ) , this effect was likely to be through another target besides rpr . We further addressed the role of rpr in regulating apoptosis in the frown using gain- and loss-of-function experiments . rpr87 deletes -1860 to +661bp relative to rpr transcription start [32] . In larvae homozygous for rpr87 or trans-heterozygous for rpr87 and deficiency Df ( 3L ) XR38 , which removes rpr and other genes , IR-induced apoptosis was reduced throughout the wing disc ( compare Fig 5A to 5B and 5C ) . These results demonstrate that reduction of rpr can prevent IR-induced apoptosis and support the idea that reduced rpr limits apoptosis in the frown . In reciprocal experiments , we induced UAS-rpr . After testing several GAL4 drivers , we identified 30A-GAL4 as specific to the hinge with little or no expression in the pouch ( S6 Fig ) . De-repression of UAS-rpr using GAL80ts and incubation at 29°C for 4 h did not induce apoptosis on its own ( Fig 5D ) but promoted IR-induced apoptosis in the frown ( compare Fig 5E and 5G to the “no GAL4” sibling control in Fig 5F and 5H ) . These results also support the idea that limited rpr expression prevents IR-induced apoptosis in the frown . We next addressed the functional significance of apoptosis resistance in the future hinge cells of the wing disc . We hypothesized that these cells survive apoptosis-inducing insults in order to participate in subsequent tissue regeneration , much like stem cells . To test this hypothesis , we used a published lineage tracing system that relies on GAL4-driven expression of UAS-RFP to mark cells of interest [33] . Co-expression of UAS-FLP in these cells results in stable GFP expression by “flipping-out” intervening sequences , thus providing a clonal marker that persists even if the cells changed their identity and lost RFP expression . Using the 30A-GAL4 driver described above , we saw coincidence of RFP and GFP in un-irradiated discs sampled at multiple points throughout feeding third larval instar stages ( Fig 6A and 6B ) . Quantification of GFP+RFP- areas ( cells that used to express RFP but have lost it ) as a percent of total GFP+RFP+ area for each disc showed that cell fates were stable; less than 7% of hinge cells that ever expressed 30A-GAL4 became RFP- ( Fig 6B and graphed in Fig 6J , “30A-GAL4 0R” ) . In contrast , in irradiated discs , GFP+RFP- cells were observed in the pouch 48 h after irradiation with 4 , 000–6 , 000R of X-rays . Quantification at 72 h after irradiation with 4 , 000R showed that GFP+RFP- cells occupied as much as 50% of the pouch in some discs , averaging ~27% ( Fig 6C–6E , quantified in Fig 6J ) . Furthermore , the fact that migrant cells no longer expressed RFP meant that these cells not only relocated but also lost their original identity , such that 30A-GAL4 was no longer active . Lack of cell death in the frown raised the possibility that cells with DNA damage are surviving in this region and contributing to regeneration . Published and new data help rule out this possibility . Prior analysis of the γ-H2Av , an indicator of DNA double strand breaks , showed that cells of the wing disc follow a similar time course of break and repair regardless of their location in the disc [34] . Briefly , robust induction of γ-H2Av was observed immediately after exposure to 4 , 000R of radiation but was reduced to background levels by 24 h after irradiation . We reproduced this result while also marking the hinge with 30A-GAL4>UAS-RFP; 24 h after irradiation with 4 , 000R , γ-H2Av signal was back to background levels throughout the disc ( S7 Fig focuses on part of the frown and nearby pouch and notum cells ) . We conclude that repair of DNA breaks is complete before the movement of hinge cells into the pouch becomes detectable . We note a published study that observed hinge cells invading the pouch [35] . A key difference in that study was the specific direction of apoptosis to the pouch by tissue-specific expression of hid , while hinge cells were deliberately kept alive . Here , we irradiated the entire disc , and indeed the entire larva , and allowed natural variations in apoptosis to occur . In this situation , cells of the hinge were protected ( Figs 1–4 ) and became part of the pouch following irradiation . To rule out the possibility that these results were due to IR-induced cell mixing within the wing disc , we performed a reciprocal experiment in which RFP/GFP markers were expressed using rotund-GAL4 . rn-GAL4 is active in the entire pouch and some inner hinge cells . Under similar experimental conditions as in 30A-GAL4 experiments , with or without irradiation , we saw near complete overlap of RFP and GFP ( Fig 6F–6J ) . In other words , while irradiation resulted in hinge cells moving into the pouch , pouch cells did not move out to the hinge , ruling out general IR-induced cell mixing . We also asked whether cells from outside the hinge are capable of rebuilding the hinge if cell death was directed to this area . We modified a published protocol by which transient expression of hid and an FLP recombinase-mediated lineage marker was directed to the pouch [35] . The authors found that the repaired pouch was made of only a small fraction ( ~25% ) of lineage-marked cells , with the rest without the lineage marker , i . e . , immigrant cells from outside the pouch . Here we directed the expression of rpr to the hinge using 30A-GAL4 while inducing stringer-GFP lineage marker at the same time . Induction of rpr for 18 h produced robust caspase cleavage in the 30A domain ( Fig 6K–6M ) . Examination of wing discs 48 h from the end of rpr/GFP lineage marker induction showed a variable extent to which GFP+ cells occupy the hinge ( Fig 6P , quantified in S as “+rpr” ) . The extent of GFP+ cells in the hinge was similar in control discs that also experienced an 18 h induction of the GFP lineage marker but without rpr ( Fig 6R , “-rpr” in Fig 6S , p-value > 0 . 7; background GFP signal in the notum in R is from the CyO-GFP balancer that is absent in P ) . In other words , the composition of the hinge remained unchanged during recovery from rpr-induced cell death . These data support the idea that regeneration of the hinge relies mostly on GFP-marked hinge cells . Next , we expressed Axin or STATRNAi in the context of 30A-GAL4-driven lineage markers to ask if Wg and STAT are needed for the relocation of cells to the pouch after irradiation . UAS-Axin or UAS-STATRNAi were induced by de-repression of GAL4 at 29°C for 24 or 48 h , with similar results ( Fig 7A shows the experimental protocol ) . In control larvae without the UAS transgenes that were subjected to this protocol , wing discs appeared normal ( Fig 7B–7E ) , and GFP+RFP- cells relocated into the area enclosed by the 30A-GAL4 domain after irradiation ( Fig 7E; quantified in Fig 7P ) . Transient expression of Axin or STATRNAi under the control of 30A-GAL4 did not disrupt with GFP/RFP expression or wing morphology ( S8 Fig ) . Thus , while Wg and STAT are required for wing development , their depletion in late third instar using the protocol shown in Fig 7A was apparently tolerated . The addition of irradiation , however , produced discs that were misshapen to varying degrees ( e . g . , Fig 7J and 7N ) and showed interruption of the 30A domain ( arrow in Fig 7M ) . Nonetheless , we were able to quantify GFP+RFP- cells and found that induction of either Axin or STATRNAi reduced the extent to which such cells populated the pouch ( Fig 7K , 7M , 7O and 7P ) . We conclude that Wg and STAT activities are required to rebuild the wing disc properly after irradiation .
We report four significant findings . First , we identified the frown , an IR-resistant domain of cells in the wing imaginal disc that participates in the rebuilding of the pouch after irradiation . Second , we found that both IR resistance and the ability to participate in regeneration are dependent on STAT and on Wg acting cell-autonomously . Third , we identified transcriptional repression of rpr as a potential mechanism by which Wg confers IR resistance in the frown . The fourth significant finding stems from the interpretation of the data in Fig 4 as follows . IR-induced apoptosis in Drosophila , as in other metazoans , is mediated , to a large extent , by p53 . Dmp53 promotes apoptosis by transcriptional activation of pro-apoptotic genes such as hid and rpr [24 , 25] . This is recapitulated by the induction of the Hid-GFP reporter in a p53-dependent manner throughout the wing disc after exposure to IR ( [31]; reproduced in Fig 4D ) . Based on this data , we infer that p53 is active in the cells of the frown following irradiation . However , it appears to be incapable of inducing in the same cells rpr , a known direct target of p53 [28] , or apoptosis . This surprising result is not without precedent; ectopic and presumably homogeneous elevation of p53 using a heat-shock-inducible transgene induced apoptosis throughout the wing disc except in the frown [15] . In another example , adenovirus E4-ORF3 protein prevents the transcriptional activation of p53 target genes despite the presence of stable , active p53 , by altering the chromatin structure at these loci [36] . We conclude that p53 activity in the frown , although capable of activating the hid-GFP promoter , is incapable of de-repressing rpr . Because ectopic rpr can induce apoptosis ( Fig 4K ) and promote IR-induced apoptosis ( Fig 5 ) , we suggest that the block in apoptosis occurs between p53 activation and rpr induction . Additional studies will be needed to understand how rpr is repressed while hid escapes this repression . A possible mechanism could be of epigenetic nature . In this regard , it is known that expression of hid , rpr , and skl are subject to epigenetic regulation through a cis-element called the irradiation responsive enhancer region ( IRER ) [37] . We note , however , that the IRER regulates both hid and rpr , whereas the mechanism we are interested in represses rpr but not hid . STAT is clearly required to prevent IR-induced apoptosis in the frown , but the effect of STAT depletion on rpr-lacZ reporter expression was minimal . Instead , the published finding that DIAP1 is a direct transcriptional target of STAT [30] suggests another mechanism . This study found that the expression of a DIAP1 transcriptional reporter , especially in the hinge region throughout wing disc development , depended on STAT , especially in the hinge region . Thus , STAT activity may help maintain a threshold of DIAP1 level that Hid , Rpr , and Skl must collectively overcome . Such a threshold model , proposed by others before [26–29] , could also explain why overexpressing rpr alone ( e . g . , Fig 4K ) could induce robust apoptosis in the frown . On the other hand , repressing just one of these while the others respond normally to DNA damage appears to be sufficient to stop the cell from reaching the threshold necessary to overcome DIAP1 ( e . g . , rpr in Fig 5A–5C ) . Similarly , reducing Hid gene dosage by half in heterozygotes is sufficient to reduce and delay the onset of radiation-induced apoptosis [38] . So , in the frown , hid may be induced by IR , but the failure to induce rpr prevents these cells from ever reaching the threshold to undergo apoptosis . With reduced STAT activity , this threshold of DIAP1 may be lower , allowing Hid and Skl to overcome it . We also cannot rule out the possibility that STAT regulates yet another apoptosis-relevant target or acts redundantly with another regulator . The requirement for STAT is as strong as the requirement for Wg in the movement of hinge cells into the pouch after irradiation . We speculate that while Wg makes a greater contribution to protecting the frown cells from death ( Axin had a stronger phenotype than STATRNAi in S4 Fig ) , both contribute equally to regeneration . All the data reported here are from analyses of wing discs up to 3 d after irradiation , by which time the larvae had begun to pupariate . It is possible that the requirement for Wg and STAT are limited to these initial regenerative responses and that different regenerative mechanisms act later in development . If so , transient inhibition of Wg and STAT as we have done in Fig 7 may still be compatible with the production of relatively normal wings with the correct cell number . If , however , the period of regeneration is limited to within 3 d after irradiation , we would expect defective wings to result from inhibition of Wg or STAT during this time . Therefore , it would be informative in future studies to follow the development of wing discs into pupae and possibly into adult stages if they survive . The data reported here indicate that STAT and Wg are necessary for radio-resistance of the frown . Are they sufficient ? We think it unlikely for the following reasons . First , the D/V boundary expresses high Wg protein ( e . g . , Fig 2A ) , yet this region shows robust IR-induced cleaved caspase signal . As for STAT , a published study reported that induction of STAT in the eye disc resulted in cell-autonomous protection from IR-induced apoptosis [30] . Similar protection in wing discs was also reported in this study , but the data was not shown . In our reproduction of this experiment using the same GAL4 driver ( en-GAL4 ) and the same UAS-STAT92E transgene , we saw little to no protection ( S9 Fig ) . We propose that additional factors besides high STAT and Wg activity are required and that these are present in the frown but to a lesser extent elsewhere in the wing disc . Gene expression characteristics in the frown that are relevant to IR resistance ( e . g . , low rpr expression ) are already present before irradiation and maintained after irradiation . This suggests that cells within the frown are intrinsically resistant to apoptosis . This is in contrast to the phenomenon of acquired resistance to IR-induced apoptosis that we reported recently [9] . In what we named the Mahakali effect , dying cells in the wing disc protect nearby survivors from IR-induced apoptosis . The Mahakali effect requires tie , which encodes a receptor tyrosine kinase , and occurs through activation of an anti-apoptotic microRNA , bantam , in the protected cells . The intrinsic resistance ( described here ) and acquired resistance ( published study ) to IR-induced apoptosis are genetically separable; tie mutants that are defective for acquired resistance still show resistance in the frown [9] . The frown , although composed of columnar epithelial cells , is resistant to IR and participates in regeneration , much like stem cells are thought to do . In this regard , it is interesting that Wg and STAT are implicated in maintaining a bona fide stem cell population and promoting their proliferation when the need for regeneration arises . In the adult Drosophila gut , autocrine and paracrine signaling activate JAK/STAT , Wg and EGFR in intestinal stem cells ( ISCs ) to maintain the latter [39–42] . In response to tissue damage by chemicals or by microbial infection , additional signaling from damaged cells further activate the same pathways in ISCs to promote their proliferation and subsequent differentiation to replace lost cells [43–47] . Our results suggest an analogous situation in the wing disc but without the involvement of a dedicated stem cell population or a stem cell niche . The involvement of STAT we found in this mode of regeneration mirrors the results of two previous studies . STAT was activated in regenerating leg discs after surgical damage and in hemocytes in response to localized epidermal damage by ultraviolet radiation [48 , 49] . While the role of JAK/STAT activation in hemocytes was unclear , STAT activation in the leg discs was found to promote regenerative cell proliferation and to delay development . There are other areas of the wing disc with high STAT or high Wg activity , yet resistance to IR-induced apoptosis is not as prominent or consistent in these areas as in the frown . Indeed , STAT must be active even in regions that show little or no reporter activity because STAT mutant clones in the pouch show phenotypes [17 , 30] . So , what distinguishes the frown from other regions of the wing disc ? It lies between Wg inner ring and Wg outer ring . Wg-OR shows lower Wg expression than Wg-IR but is more robust along its dorsal boundary than in any other parts ( e . g . , [50] ) . Therefore , the frown in feeding third instar larvae could benefit not only from high STAT activity but also from two sources of Wg . We speculate that the combination of high STAT and possibly highest Wg signals makes the frown consistently resistant to apoptosis . This is in agreement with published studies that noted different adhesive and growth properties between cells of the hinge and the pouch , differences that appear to be related to Wg activity . For example , cells of the hinge show lower levels of E-Cadherin in adherens junctions , but elevation of Wg activity increased E-Cadherin levels , induced apical constrictions , and caused cells to bulge out , particularly in the hinge [51 , 52] . In another study , activation of Wg slowed proliferation in the pouch but accelerated proliferation in the hinge [53] . The accompanying report by W-M . Deng and colleagues shows that the same area also carries higher tumorigenic potential than the rest of the wing disc , further illustrating the special nature of this region [54] . The regulatory module we identified that operates through Wg , STAT , and Rpr , constitutes a new mechanism for resistance to IR-induced apoptosis . Through variations in gene expression/activity , a domain of resistant cells arises from among cells of identical genotype . A similar phenomenon has been described in mammalian cells in which , among a population of clonal cells , variations in the level or activity-state of proteins regulating receptor-mediated apoptosis can cause some cells to die while others survive exposure to the death-inducing ligand TRAIL [55] . Our work adds to this paradigm by showing that such cells and/or their clonal descendants can migrate , change fate , and participate in reconstruction of other tissue compartments . Deregulation of Wnt and STAT pathways is associated with cancer . Wnt-1 ( ortholog of Wg ) was first identified as int1 , a proto-oncogene targeted during viral carcinogenesis [56] . Likewise , persistent activation of STAT3 and , to a lesser extent , STAT5 is implicated in survival , proliferation , and invasion in cancer [57] . The prominence of STAT3 and STAT5 in cancer is interesting because the sole Drosophila STAT homolog is most like STAT3 and STAT5 in sequence and activity among the seven mammalian STATs [21] . More relevant to our study , deregulation of Wnt and STAT pathways is implicated in resistance to cytotoxic cancer therapy . Gene expression analysis across multiple cell lines found Wnt signaling and STAT3 among genes whose expression correlates with resistance to chemo-radiation or radiation [58 , 59] while combining radiation with a chemical inhibitor for STAT5 reduced survival in head and neck cancer cell lines [60] . Furthermore , inhibition of STAT3 reduced the ability of tumor-initiating colorectal carcinoma cells to form “tumor-spheres” in vitro [61] , suggesting that STAT3 may be important for repopulation of tumors after treatment or for new metastases . Based on these considerations , we suggest that cells of the frown , which are protected from apoptosis by Wg and STAT and participate in tissue rebuilding after radiation damage , may provide a unique opportunity to study how these conserved signaling pathways ensure survival and regeneration in a genetically tractable model organism . For instance , we do not know the mechanism by which hinge cells relocate to the pouch . Directed cell divisions that “send” daughter cells into the pouch , cell migration , or a combination thereof , are some of the possible ways . Likewise , we also do not know the mechanisms by which hinge cells lose their hinge identity . We hope to exploit genetic tools to address these questions in the future .
These stocks are described in Flybase: w1118 , y1w1118 , wgI-12 , wgI-16 ( Df ( 2L ) wg-CX3 , wg[l-16] b[1] pr[1]/CyO ) , STAT92EF , STAT92E06346 , ptc-GAL4 ( on Ch II ) , en-GAL4 ( on Ch II ) , rn-GAL4 ( on Ch III ) , 30A- GAL4 ( on Ch II , Bloomington stock#37534 ) , Ptub-GAL80ts on Ch III , 10XSTAT-GFP , UAS-GFP ( on Ch II ) , UAS-TCFDN ( on Ch III ) , UAS-Axin-GFP ( on Ch III ) , UAS-STATRNAi ( on X ) , rpr-11-lacZ ( on Ch III ) . The stock used for lineage tracing is also described in Flybase; w*; P{UAS-RedStinger}4 , P{UAS-FLP . D}JD1 , P{Ubi-p63E ( FRT . STOP ) Stinger}9F6 /CyO ( Bloomington stock#28280 ) . STAT92EF is a temperature-sensitive EMS-induced point mutation [62] . STAT92E06346 is a strong loss-of-function allele caused by a transposon-insertion that produces no transcript or protein [63 , 64] . wg1-12 is an EMS-induced temperature sensitive allele [22] . Other stocks: DIAP1-lacZ [65] , hid-GFP [31] , ci-GAL4 [66] , UAS-rpr ( on X , [67] ) , UAS-STAT92E [30] , rpr87 [32] , and Df ( 3L ) XR38 [68] . Larvae were raised on Nutri-Fly Bloomington Formula food ( Genesee Scientific ) at 25°C unless otherwise noted . The cultures were monitored daily for signs of crowding , typically seen as “dimples” in the food surface as larvae try to increase the surface area for access to air . Cultures were split at the first sign of crowding . Larval timing , i . e . , that we were working with feeding stage three instar larvae , was through a combination of larval age and size ( late third instar larvae were significantly bigger than second instar in the absence of crowding ) and their location in the food . Larvae in food were irradiated in a Faxitron Cabinet X-ray System Model RX-650 ( Lincolnshire , IL ) at 115 kv and 5 . 33 rad/s . For maytansinol treatment , larvae at 94–98 h after egg deposition were transferred to food containing 2 μM maytansinol or DMSO control . Wing discs were dissected 24 h after the transfer . Cleaved Caspase 3 ( Drice ) ( 1:100 , rabbit polyclonal , Cell Signaling Cat# 3661 ) was used as described before [38] . Antibodies to cleaved Dcp1 ( 1:100 , rabbit polyclonal , Cell Signaling Cat# 9578 ) Phospho-Histone H3 ( 1:1000 , rabbit monoclonal , Upstate Biotech ) , Wingless ( 1:100 , mouse monoclonal , Drosophila Hybridoma Bank Cat# 4D4 ) , β-galactosidase ( 1:100 , Developmental Hybridoma Bank Cat#40-1a ) were used as described before [25] . Anti-γ-H2Av antibody ( 1:2000 , mouse monoclonal , Drosophila Hybridoma Bank Cat# UNC93-5 . 2 . 1 ) was used as described before [69] . Secondary antibodies were used at 1:100 ( Jackson ) . For antibody staining , wing discs were dissected in PBS , fixed in 4% para-formaldehyde in PBS for 30 min , and washed three times in PBS , permeabilized in PBTx ( 0 . 5% Triton X-100 ) for 10’ , and rinsed in PBTx ( 0 . 1% Triton X-100 ) . The discs were blocked in 5% Normal Goal Serum in PBTx ( 0 . 1% Triton X-100 ) for at least 30 min and incubated overnight at 4°C in primary antibody in block . The discs were rinsed thrice in PBTx ( 0 . 1% Triton X-100 ) and incubated in secondary antibody in block for 2 h at room temperature . For TUNEL , wing discs were dissected in PBS , fixed in 4% para-formaldehyde in PBS for 20 min , and washed three times in 0 . 3% Triton X-100/PBS for at least 20 min total . The discs were permeabilized overnight at 4°C in 0 . 3% Triton X-100/PBS , followed by three washes in 0 . 3% Triton X-100/PBS for at least 10 min total . The discs were processed using an Apoptag Red kit ( Millipore ) , according to the manufacturer’s instructions . Stained discs were washed in PBT . The discs were counter-stained with 10 ug/ml Hoechst33258 in PBT or PBTx ( 0 . 1%TritonX-100 ) for 2 min , washed three times , and mounted on glass slides in Fluoromount G ( SouthernBiotech ) . Wing discs were dissected from wild-type ( y1w1118 ) or STAT-GFP reporter larvae , incubated in 10 μM EdU in Schneider’s Insect Medium ( Sigma ) for 1 hr , and processed to detect incorporated EdU according to the manufacturer’s instructions ( Alexa647 Click-iT Edu Imaging Kit , Molecular Probes ) . A cocktail of 48 Quasar 570-labeled custom-made anti-sense probes from Stellaris_LGC Biosearch Technologies were used . Probe sequences in the 5′–3′ direction as generated by Stellaris probe designer version 4 . 1 against hid mRNA variant B ( GenBank Accession NM_001275081 ) were: 1gctccgcggctaaaaatgaa; 2agggcaggaaacacgtctta; 3acgtgtcgcagactcaaaga; 4tttgcttttgctgttgtcaa; 5actgttcacgatggatttcg; 6acccttttcgtgtttagaac; 7gattcttctgcgtttttcat; 8caagtttttgctcggttagt; 9ttatctttcctgatttgtca; 10tttttcgtgcagttttttct; 11gattttgtatttcttgtgca; 12acttttggttagagttcact; 13gctttgttttgctttttatt; 14cttcttgtgattgttcttcg; 15tgcactttgttggcactttg; 16ggcaaataaaagggcacggc; 17gatgaactcgacgctacgtc; 18aggaggagacggacgaggat; 19cgatgcggaggacgaagatg; 20tagagggcgtatagcacttg; 21cggcggatactggaagattt; 22cgtgaaattgcaagaggggc; 23ccgtgcggaaagaacacatc; 24attcgagttcggattcggat; 25ggaagaagttgtactcctcg; 26gatatgacggatgtggttgc; 27gaatggtgtggcatcatgtg; 28tcatgatcgctctggtactc; 29aaagttgtcgtagcgatcgc; 30tccattgaactcctgcagac; 31tattggagctcttcttcttc; 32ggtatggcagactggattat; 33gactgatgtggccatggatg; 34tctgtggtttcttcttctcg; 35acaacagttggccaagtgaa; 36gcccatggccaaaacgaaaa; 37ttcatcgcgccgcaaagaag; 38attcgattacacgtctcctg; 39cttaagggctagctgatttc; 40aactatgtttagatcggca; 41gttgcacttatgtacggttt; 42cgctcctgcagttcaataaa; 43atgttggctgtttgtgtatc; 44ccttcttaatcttaggcaca; 45atatattgttcttgtgtccc; 46tgcagttaccatagacagat; 47gttatctttcgtttcgtttt; 48cttgccagtctaagagtttt . 90–102 h old larvae were dissected in PBS 2 h after irradiation and fixed in 4% paraformaldehyde in PBS for 30 min at room temperature . The discs were dehydrated for 24 h at 4°C in 70% ethanol . The samples were hybridized with 0 . 05 μM probe for 16 h at 37°C in the dark . All other experimental conditions were as published [70] . The discs were counter-stained with 5 μg/ml Hoechst33258 for DNA . The samples were mounted in 50% glycerol , 150 mM NaCl , and 15 mM sodium citrate for imaging . With the exceptions noted below , the discs were imaged on a Perkin Elmers spinning disc confocal attached to a Nikon inverted microscope , using a SDC Andor iXon Ultra ( DU-897 ) EM CCD camera . The NIS- Elements acquisition software’s large image stitching tool was used for the image capture . Twenty to 21 Z-sections 1 um apart were collected per disc and collapsed using “maximum projection” in Image J . The exceptions are: Fig 4A–4G were acquired on a Nikon inverted microscope with a Hamamatsu image EM C9 100–13 EM CCD camera; Fig 4A and 4B show a single optical section each; Fig 4C and 4D were optical sections collapsed using “sum projection” in image J; S1F and S1G Fig were imaged on a Q-Imaging R6 CCD camera using Ocular software .
|
Like other insects , Drosophila larvae have epithelial structures called imaginal discs that will give rise to most of the external adult structures , such as wings , limbs , or antennae; these organ precursors are formed by a single layer of epithelial cells that folds into a sac . Imaginal discs manage to regenerate efficiently if they are damaged . Previous studies have shown that dying cells produce signals that activate cell proliferation of some of their neighbors , allowing them to regenerate the disc and thereby enabling the flies to develop into normal adults . But a dedicated stem cell population that contributes to regeneration , if any , remained to be identified . Here , we report the identification of a subpopulation of cells in wing imaginal discs that is more resistant to the cytotoxic effects of radiation and drugs . We show that the protection of these cells depends on two signaling pathways—Wingless and STAT—that are conserved in humans . Following tissue damage by radiation , we observe that protected cells change their location and their identity , allowing them to fill in for dead cells in other parts of the same organ precursor . In sum , this work identified ways in which a subset of cells in Drosophila imaginal wing discs is preserved through radiation exposure so that they could participate in regeneration of the organ after radiation damage . We also discuss how this situation may resemble human cancers .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"death",
"invertebrates",
"cell",
"processes",
"animals",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"organism",
"development",
"stat",
"proteins",
"embryos",
"morphogenesis",
"drosophila",
"research",
"and",
"analysis",
"methods",
"embryology",
"proteins",
"stat",
"signaling",
"imaginal",
"discs",
"insects",
"arthropoda",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"regeneration",
"apoptosis",
"biology",
"and",
"life",
"sciences",
"metamorphosis",
"cell",
"signaling",
"larvae",
"organisms"
] |
2016
|
Drosophila Wnt and STAT Define Apoptosis-Resistant Epithelial Cells for Tissue Regeneration after Irradiation
|
Receptors of the signalling lymphocyte-activation molecules ( SLAM ) family are involved in the functional regulation of a variety of immune cells upon engagement through homotypic or heterotypic interactions amongst them . Here we show that murine cytomegalovirus ( MCMV ) dampens the surface expression of several SLAM receptors during the course of the infection of macrophages . By screening a panel of MCMV deletion mutants , we identified m154 as an immunoevasin that effectively reduces the cell-surface expression of the SLAM family member CD48 , a high-affinity ligand for natural killer ( NK ) and cytotoxic T cell receptor CD244 . m154 is a mucin-like protein , expressed with early kinetics , which can be found at the cell surface of the infected cell . During infection , m154 leads to proteolytic degradation of CD48 . This viral protein interferes with the NK cell cytotoxicity triggered by MCMV-infected macrophages . In addition , we demonstrate that an MCMV mutant virus lacking m154 expression results in an attenuated phenotype in vivo , which can be substantially restored after NK cell depletion in mice . This is the first description of a viral gene capable of downregulating CD48 . Our novel findings define m154 as an important player in MCMV innate immune regulation .
Pathogens have recourse to innumerable tactics for evading host immune surveillance . Viruses , and in particular large DNA viruses such as herpesviruses , are endowed with the capacity to encode multiple products committed to altering , during all stages of their life cycle , several functions of the innate and adaptive immune system . The homeostatic equilibrium achieved between host immune responses and viral immune escape mechanisms empowers these viruses to successfully establish their characteristic lifelong infections . Human cytomegalovirus ( CMV ) , the prototype β-herpesvirus , usually leads to asymptomatic infections in healthy individuals where it remains in a latent state for life , going through sporadic reactivation and leading to severe diseases in immunocompromised patients [1] , [2] . The generation of an efficient host-elicited immune response against CMV includes the induction of natural killer ( NK ) cells , antibody and T-cell mediated responses [3] . As a consequence , CMV has evolved diverse countermeasures to avoid recognition by T cells , allowing it to interfere with the surface expression of major histocompatibility complex class I ( MHC class I ) and class II and costimulatory molecules , compromising antigen presentation [3]–[6] . Likewise , the virus counteracts NK cell triggering , primarily by suppressing the expression of ligands for activating receptors while preserving engaged inhibitory receptors [7]–[9] . In addition , CMV alters the function of cytokines and their receptors , and interacts with complement factors . While great strides have been made in recent years in identifying CMV inhibitors of immune response mechanisms , current consensus is that among the vast amount of genetic CMV material still requiring a functional assignment , the virus harbours as yet uncovered immunoevasins directed against already known or new immunological targets . Due to the species-specific nature of human CMV ( HCMV ) replication , infection of mice with murine CMV ( MCMV ) has proven to be an invaluable model for studying aspects of the biology underlying CMV infection . In this regard , the MCMV system has been widely used to unveil new immunomodulatory molecules and to explore their roles in infection and viral pathogenesis [10] . The signalling lymphocyte-activation molecules ( SLAM ) family of cell-surface receptors is a distinct structural subgroup of the immunoglobulin ( Ig ) superfamily differentially expressed on hematopoietic cells and found to play pivotal roles in both innate and adaptive immunity [11]–[13] . Among other activities , SLAM immunomodulatory receptors regulate cell adhesion , cytokine production , and cytotoxicity of NK and CD8+ T cells . The SLAM family currently consists of nine members , CD48 , CD84 , CD150 ( SLAM ) , CD229 , CD244 ( 2B4 ) , CD319 ( CRACC ) , CD352 ( SLAMF6 , NTB-A; Ly108 ) , CD353 ( BLAME ) and SLAMF9 . One of the hallmarks of this class of receptors is that they interact with members of the same family via their amino-terminal Ig-V domains . While most of them are typically self-ligands , participating in homophilic interactions , CD48 is a heterophilic receptor for CD244 [14] . The cytoplasmic domain of most SLAM family members carry one or more copies of a distinctive immunoreceptor intracellular tyrosine-based switch motif ( ITSM ) [13] , [15] . Upon receptor engagement , these motifs undergo phosphorylation and recruit with high affinity and specificity adaptor molecules like the SLAM-associated protein ( SAP ) [11] , [16] . In particular , CD48 is a GPI-anchored glycoprotein with expression in a broad range of cells of the hematopoietic lineage , especially on antigen-presenting cells [17] . CD244 , the high-affinity counter receptor of CD48 both in humans and mice , is a transmembrane surface glycoprotein with an intracellular tail containing four ITSMs . It is highly expressed on NK cells , and to a lesser extent on other cytotoxic cells such as CD8+ T cells , basophils , and eosinophils . CD244 is an important activating receptor for the regulation of CD8+ T and mature NK cells , promoting cell-mediated cytotoxicity and cytokine release [18]–[20] . Engagement of CD244 by its ligand leads to the polarization and release of cytolytic granules into the contact zone between NK and target cells [21] . SLAM family receptors have been shown to play specific roles in viral pathogenesis . Various morbilliviruses , including the highly contagious measles virus , employ CD150 as the principal receptor to enter into a subset of immune cells , facilitating their spread and contributing to viral-induced immunosuppression [22] , [23] . In response to Epstein-Barr virus infection , CD48 is strongly induced on the surface of B lymphocytes and may aid viral trafficking [24] . In addition , we have recently shown that HCMV encodes UL7 , a CD229 structural homologue capable of interfering with proinflammatory responses [25] . The role of SLAM receptors in antiviral immunity has been clearly documented in the X-linked lymphoproliferative syndrome , a rare immunodeficiency human disease in which impaired signalling functions of the SLAM receptors , stemming from mutations in the SAP-encoding gene , is associated with an extreme sensitivity to infection with Epstein-Barr virus [26] . Therefore , since SLAM receptors are active components of host immunity , viruses might have evolved immune evasion manoeuvres to specifically ablate triggering of such receptors . Indeed , this is the case for HIV-1 , which utilizes Vpu to elude NK cell recognition through the downregulation of NTB-A expression on the surface of infected CD4+ T cells [27] . Whether this is a more generalized phenomenon and what consequences modulating SLAM receptors may cause in the infected host remain unknown . In this study we show that MCMV infection efficiently decreases the expression of several SLAM family receptors at the cell surface of macrophages , and we pinpoint m154 as the viral downregulator of CD48 . We found that m154 helps to debilitate the effectiveness of anti-MCMV triggered NK cell responses , thereby meliorating viral growth in vivo . Thus , we present here a novel strategy evolved by CMV to subvert detection by NK cells during acute infection , based on the modulation of a SLAM family member .
SLAM family members are differentially expressed among hematopoietic cells . As macrophages play a key role in MCMV infection with regard to viral replication , dissemination , and the establishment of latency [28]–[30] , and constitute one of the principal effectors of innate immunity , we selected this particular cell type to explore potential SLAM perturbations upon MCMV infection . Using flow cytometry , we first assessed whether CD48 , CD84 , CD150 , CD229 , CD244 and Ly108 were present on the surface of thioglycollate-elicited peritoneal macrophages . The lack of a commercially available antibody against CD319 , CD353 , and SLAMF9 prevented the study of these receptors in this cell type . As shown in Figure 1A , the SLAM receptors CD48 , CD84 , CD229 , and Ly108 were expressed on the macrophage surface , whereas CD150 and CD244 could not be detected . We found however , that CD150 was present at the surface of LPS-treated mouse peritoneal macrophages ( data not shown ) , consistent with earlier studies [31] . Thus , peritoneal macrophages represent an MCMV permissive cell type expressing a number of SLAM receptors , allowing us to examine whether members of this family could be targets for modulation during MCMV infection . We then infected peritoneal macrophages with MCMV-GFP at a multiplicity of infection ( moi ) of 2 . The use of MCMV-GFP , based on the bacterial artificial chromosome ( BAC ) -cloned MCMV genome pSM3fr-GFP , which contains a GFP gene inserted within the ie2 locus [32] , allowed us to track and selectively analyze infected cells in the cultures . Under these conditions , infection rates reached approximately 50% . At different times ( 24 h , 48 h , and 72 h ) after infection , cells were stained for the surface expression of CD48 , CD84 , CD229 , and Ly108 . Notably , MCMV infection resulted in the significant progressive downregulation of all the four receptors analyzed over the course of the infection , when compared to both non-infected cells ( GFP negative ) from the same culture ( Figure 1B ) or with mock-infected macrophages ( data not shown ) . Surface reductions in CD84 and Ly108 were already observed at 24 h post-infection ( hpi ) , and at 48 hpi for CD48 and CD229 , becoming for all four receptors more pronounced at 72 hpi . Thus , by 72 hpi macrophages demonstrated a dramatic loss in expression of the four SLAM receptors analyzed . As expected [6] , a significant surface decrease in MHC class I molecules was also detected in infected cells . Similar results were obtained when experiments were performed with wild-type ( wt ) MCMV not expressing GFP ( data not shown ) . We further analyzed the effect of the viral dose on the alteration of SLAM surface expression by infecting peritoneal macrophages at different mois , ranging from 0 . 5 ( ∼5% infected macrophages ) to 5 ( ∼70% infected macrophages ) , with MCMV-GFP . As depicted in Figure 2A , there was a strong dependency on the viral dose for cell-surface reduction of SLAM receptor expression concomitant with the downmodulation of MHC class I , which in turn correlated with the extent of infected peritoneal macrophages . To determine whether viral gene expression was required for SLAM downregulation , macrophages were treated with UV-inactivated MCMV . The results showed no decrease in CD48 , CD84 , CD229 , or Ly108 surface expression after infection of macrophages for 72 h with the UV-inactivated virus ( Figure 2B ) , indicating that SLAM downregulation could be attributed to specific MCMV genome-encoded products . Moreover , for Ly108 , cell-membrane expression levels after infection with UV-inactivated MCMV were even higher than those of uninfected cells , most likely due to the viral-dependent macrophage activation ( data not shown ) . Altogether these results show that MCMV encodes gene products that efficiently diminish the cell-surface levels of SLAM family members . Since CD244 , the high affinity receptor for CD48 , is expressed in NK and cytotoxic CD8+ T cells known to play a prominent role in the clearance of MCMV infection , we decided to further explore the consequences of the cell-surface depletion of CD48 , and sought to identify the viral product ( s ) causing it . The potential modulators of SLAM receptors would most likely be genes dispensable for viral replication in vitro . Thus , to identify the MCMV gene product ( s ) that might mediate the downregulation of CD48 , we systematically screened the viral genome utilizing a panel of mutant viruses bearing deletions of approximately 10–15 kbp each in non-essential regions . These mutant viruses were based on the BAC-cloned MCMV genome containing the GFP . Peritoneal mouse macrophages were infected with wild-type and mutant MCMVs , and at 72 hpi were tested for surface expression of CD48 . After infection with the deletion mutant MCMV-GFPΔm144-m158 ( Figure 3A ) missing genes extending from m144 to m158 , cell-surface CD48 was restored , reaching levels comparable to that of non-infected cells ( Figure 3B ) . As expected , due to the lack of the m147 . 5 gene this deletion mutant was also capable to revert the cell-surface expression of CD86 [33] , whereas it did not significantly affect the downregulation of other SLAM receptors , such as Ly108 . At this point , three additional viral mutants , MCMV-GFPΔm144-m148 , MCMV-GFPΔm149-m153 , and MCMV-GFPΔm154-m157 all containing smaller specific deletions within the m144-m157 region ( from m144 to m148 , from m149 to m153 , and from m154 to m157 , respectively ) ( Figure 3A ) were assessed for their capability to interfere with CD48 . As shown in Figure 3B , only the MCMV mutant in which the genetic region encompassing m154 to m157 was removed , efficiently relieved CD48 downregulation , while levels of CD86 remained similar to those present in wt MCMV-infected macrophages . CD86 , however , was not reduced from the macrophage surface after infection with either MCMV-GFPΔm144-m148 or MCMV-GFPΔm149-m153 , mutants that do lack the m147 . 5 gene . To further narrow down the possible viral CD48 downregulators , we examined two additional viral mutants containing deletions within the m153-m157 genomic region , MCMV-GFPΔm153-m154 and MCMV-GFPΔm155-m157 ( Figure 3A and data not shown ) . Notably , the MCMV mutant lacking m153 and m154 genes , but not the viral mutant missing genes m155 to m157 , reverted CD48 downregulation ( Figure 3B , and data not shown ) . As a role of the m153 gene in CD48 cell-surface alteration had been excluded after analyzing MCMV-GFPΔm149-m153 , we deduced that the m154 gene product was the one leading to reduced macrophage-surface expression of CD48 during MCMV infection . This observation was confirmed with a viral mutant bearing a deletion in m154 , MCMVΔm154 ( Figure 3A ) , which was able to ablate downregulation of CD48 to an extent comparable to that of mock-infected cells , whereas it maintained the downregulation of Ly108 and CD84 ( Figure 3C ) . As Tang and co-workers [34] in a reassessment of global MCMV ORFs using DNA microarray analysis reported two additional small ORFs , m154 . 3 and m154 . 4 , potentially expressed in infected NIH 3T3 cells , that partially overlapped with ORF m154 and which therefore were interrupted in the deletion mutant MCMVΔm154 , a new recombinant MCMV carrying a smaller internal deletion in m154 that preserved intact both m154 . 3 and m154 . 4 ( MCMVΔm154Int ) was generated . In a manner similar to MCMVΔm154 , MCMVΔm154Int did not significantly alter CD48 surface levels ( Figure 3C ) . These data further confirmed that the observed rescue of CD48 surface density in infected macrophages was the result of deleting the m154 gene . Thus , we concluded that m154 abrogates the surface expression of CD48 . The m154 gene belongs to the m145 gene family [35] , comprised of eleven members , some of which encode molecules that adopt an MHC class I fold [36] and which are known to be involved in the modulation of immune responses . In contrast to other members of this family , the m154 gene has no homology with MHC class I genes . It encodes a 368-aa type I transmembrane protein with a 23-aa putative N-terminal signal peptide , a 300-aa ectodomain , a 23-aa transmembrane domain , and a 22-aa C-terminal cytoplasmic tail ( Figure 4A ) . The ectodomain is a mucin-like domain displaying a striking number of serine ( 29 ) and threonine ( 84 ) residues that are potential O-linked glycosylation sites , and contains one putative N-glycosylation site ( at position 161 ) . A search of the available sequence databases using the m154 deduced amino acid sequence revealed no significant degree of sequence identity between m154 and other known viral or host proteins . In order to examine m154 expression during the viral infection , we raised a specific monoclonal antibody ( mAb; m154 . 4 . 113 ) against the protein , using a peptide corresponding to its cytoplasmic tail as an immunogen . Peritoneal macrophages , either mock-infected or infected for 72 h with wt MCMV , were analyzed by Western blot with this mAb . A single protein band with an apparent molecular mass of ∼60 kDa was detected only in the infected cell ( Figure 4B ) . The migration of the detected protein differed from the predicted size of the mature m154 , which is 38 kDa , being highly suggestive of an extensive glycosylation occurring via its copious serine and threonine residues . To identify the expression kinetic class of m154 , we infected macrophages in the presence of either the viral DNA synthesis inhibitor phosphonoacetic acid ( PPA ) , which prevents late viral gene expression , or the protein translation inhibitor cycloheximide ( CHX ) , which selectively limits viral gene expression to immediate early genes . As shown in Figure 4B , m154 was not recovered after release from the CHX block in the presence of actinomycin D , whereas under these conditions , the major immediate early MCMV protein IE1 was abundantly found , as expected . m154 , however , was readily detected after PPA treatment , indicating that this viral protein is expressed with early kinetics . Infected macrophages were also examined by indirect immunofluorescence to determine the subcellular localization of m154 . The protein was strongly expressed on the cell membrane and to a lesser extent in the cytoplasm ( Figure 4C ) . Biotin-labelling of proteins on the surface of wt MCMV-infected macrophages , followed by immunoprecipitation with the anti-m154 mAb , SDS-PAGE , and subsequent Western blot probed with labelled streptavidin , confirmed the presence of m154 at the cell surface ( Figure 4D ) . Localization of m154 on the cell surface was also observed after ectopic expression of m154 . Thus , when 300 . 19 cells stably transfected with an HA-m154 fusion protein containing the influenza hemagglutinin ( HA ) epitope tag inserted at the N-terminal end of m154 were analyzed by flow cytometry using an anti-HA antibody , a cell-surface pattern of HA staining was observed ( Figure 4E ) . It must be noted , however , that when expressed in isolation , m154 exerted no overt effects on the surface levels of CD48 , which was constitutively expressed in 300 . 19 cells ( Figure 4F ) . This result suggests the need of additional MCMV encoded proteins or virally induced cellular molecules for m154 to operate appropriately . To asses the ability of m154 to downregulate CD48 when ectopically expressed in the context of an MCMV infection , we generated a viral mutant ( MCMVm154Ectop , Figure 5A ) in which we inserted the m154 ORF plus 210 nt of its corresponding putative promoter and 60 nt including its putative polyA signal , into the genome of MCMVΔm154 behind the ie2 ORF . As shown in Figure 5B , MCMVm154Ectop largely reduced CD48 macrophage-surface levels when examined at 72 hpi , time at which m154 could be clearly detected by indirect immunofluorescence in the MCMVm154Ectop-infected cell , where it displayed a distribution comparable to that observed during wt MCMV infection ( compare Figure 5C and Figure 4C ) . Based on all these findings , we concluded that m154 , the MCMV downregulator of CD48 , is an early-phase mucin-like protein with a predominant cell-surface localization in the infected cell . Although MCMV-encoded products with immunomodulatory properties are not believed to play a role in the viral replication cycle , we analyzed whether m154 affected MCMV growth in tissue culture . To this end , single-step growth curves of MCMVΔm154 and wt viruses were determined in mouse embryo fibroblasts ( MEFs ) and peritoneal macrophages after infection at a low moi . MCMVΔm154 displayed plaque morphologies in MEFs and growth kinetics in both cell types that were indistinguishable from those of wt virus , confirming the lack of involvement of m154 in the viral replication cycle ( Figure 6A and data not shown ) . In addressing the mechanism by which m154 downregulates CD48 , we first considered the possibility that this viral protein may affect CD48 transcription . However , when we compared CD48 RNA levels in wt MCMV-infected macrophages with mock-infected cells by reverse transcriptase ( RT ) -PCR , we found no substantial change in CD48 mRNA content ( Figure 6B ) . This observation suggested that CD48 expression was being altered through post-transcriptional mechanisms . Therefore , we examined CD48 protein in cell lysates of wt MCMV-infected cells at different time points by using Western blot . As depicted in Figure 6C , CD48 ( ∼40 kDa band ) levels were drastically lower in total cell lysates of infected macrophages , especially after 48 h of infection . This decrease in CD48 occurred concomitantly with the appearance of m154 , which was readily detected 48 h after infection , reaching a maximum at 72 hpi and then continuing to accumulate in the infected cell . Thus , the data pointed towards proteolytic degradation of CD48 during MCMV infection . To assess which protein degradation pathway was involved , we used proteasomal or lysosomal proteolysis inhibitors . MG-132 is considered a specific 26S proteasome inhibitor , while leupeptin is a reversible and competitive intralysosomal proteolysis inhibitor that specifically blocks serine and some cysteine proteases . As revealed by Western blotting , treatment with MG-132 was able to significantly restore CD48 expression ( Figure 6D ) . We also observed however , that the presence of leupeptin partially abrogated CD48 degradation . Consistent with these findings , immunofluorescence microscopy assays evidenced enhanced CD48 signals in the wt MCMV-infected macrophages exposed to the two different proteolysis inhibitors ( Figure 6E , panels j and n ) as compared to that of the untreated-infected cells ( Figure 6E , panel f ) . Moreover , co-localization of m154 and CD48 could be clearly visualized in both MG-132 and leupeptin treated-infected macrophages ( see panels l and p in Figure 6E ) . Altogether , the data indicate that MCMV targets CD48 for degradation , likely using both the proteasome- and the lysosome-mediated mechanisms . As indicated , the natural ligand for CD48 is CD244 , a molecule that is expressed on all NK cells , and to a lesser extent , on other cytotoxic leukocytes . We first sought to explore whether infection of macrophages with MCMV resulted in a reduced recognition by CD244 due to the loss of CD48 on the cell surface . For this purpose , we generated a soluble murine CD244-Fc fusion protein containing the ectodomain of CD244 fused to the Fc portion of the human IgG . As shown in Figure 7A , binding of CD244-Fc fusion protein to macrophages was significantly decreased upon wt MCMV infection . On the other hand , the fusion protein interacted with MCMVΔm154- and MCMVΔm154Int-infected cells in a similar manner to non-infected cells ( Figure 7A ) . The interaction of CD48 with CD244 increases NK cell activation , triggering cytotoxicity . Thus , by suppressing CD48-surface expression , m154 could help MCMV elude NK cell-mediated immune responses . To ascertain whether this was the case , we compared the degranulation capacity of NK cells after exposure to macrophages infected with wt MCMV or MCMVΔm154 . For this purpose , we used a flow cytometric-based assay to measure NK cell-surface expression of LAMP-1 ( CD107a ) . NK cells purified from mouse spleens were incubated with mock- , wt MCMV- or MCMVΔm154-infected cells at a macrophage/NK ratio of 1∶1 . As expected , the percentage of CD107a+ NK cells specifically augmented in response to the viral infection as compared to non-infected cultures . No substantial differences could be detected , however , in any of the experiments performed , when we compared the percentage of CD107a+ NK cells incubated with wt MCMV and those incubated with mutant MCMV-infected macrophages ( e . g . mock: 5 . 5%±1 . 7; wt MCMV: 31 . 6%±2 . 9; MCMVΔm154: 26 . 3%±1 . 3 ) . In contrast , we observed markedly increased CD107a externalization in the NK cell population responding to the m154 defective MCMV , as indicated by a 2-fold increase in the CD107a mean fluorescence intensity ( MFI ) on the CD107a+ NK cells exposed to MCMVΔm154-infected cultures as compared to wt MCMV ( Figure 7B and 7C ) . Thus , the mean number of granules discharged by individual degranulating NK cells during stimulation by MCMV-infected macrophages was lower when m154 was being expressed . To evaluate whether the effects on NK-cell responses observed were caused by m154 acting on the CD48/CD244 axis , we performed degranulation assays on co-cultures of NK cells and MCMVΔm154-infected cells pre-incubated with the CD244-Fc fusion protein . As shown in Figure 7D , the CD244-Fc fusion protein partially blocked CD107a surface expression on NK cells exposed to the MCMVΔm154-infected cells , while an irrelevant control Fc fusion protein did not have a significant impact . Together , the results indicate that m154 contributes to confer protection to MCMV-infected macrophages against NK cell attack , and that these effects are mediated , at least in part , through m154 downregulation of CD48 . We reasoned that reduction of CD48 surface expression on antigen-presenting cells may contribute to the host's impaired ability to control viral growth . We therefore sought to explore the impact that m154 plays in the context of an acute viral infection by inoculating BALB/c mice with MCMVΔm154 or wt MCMV . By day 2 after infection , we could observe that mice intraperitoneally ( i . p . ) inoculated with 2×106 plaque forming units ( PFU ) of MCMVΔm154 had gained a larger percentage of body weight than mice infected with the same dose of wt MCMV ( Figure 8A ) . Moreover , while wt MCMV-infected animals lost a substantial percentage of body weight by days 4 , 6 and 8 after infection , animals infected with MCMVΔm154 did not experienced any weight loss during the course of the assay . Thus , at day 8 after infection the average body weight of mice infected with wt MCMV was 14 . 4 g±4 . 0 , whereas mice infected with MCMVΔm154 had an average body weight of 18 . 3 g±4 . 3 ( data not shown ) . In agreement with the loss of body weight , wt MCMV-infected mice also developed more exacerbated clinical signs of disease , such as ruffled hair , hunched posture and lethargy ( data not shown ) . When we analyzed the frequency of infected peritoneal macrophages , we did not find significant differences between wt MCMV- and MCMVΔm154-infected mice ( wt MCMV: 3 . 0%±0 . 6; MCMVΔm154: 2 . 9%±0 . 2 ) . Neither , the nature of the cellular influx to the peritoneal cavity , as determined by the levels of neutrophils ( CD11b+ Gr-1+ ) , macrophages ( CD11b+ Gr-1− ) , T lymphocytes ( CD3+ ) or B lymphocytes ( IgM+ ) appeared to be distinct amongst the two groups of infected animals ( Figure S1 ) . We subsequently determined the replication levels of the viral deletion mutant in several target organs of the animals at different days post-infection . As depicted in Figure 8B , while at day 2 post-infection , comparable viral titers were observed in the spleens of wt MCMV- and MCMVΔm154-infected mice , at day 4 after infection , viral titers in MCMVΔm154-infected animals were around 32- , 6- , 9- , and 4- fold lower in spleen , liver , kidney , and lung , respectively , than those found in the same organs of wt MCMV-infected mice . Likewise , at day 8 post-infection , viral loads of MCMVΔm154 were considerably lower in the organs analyzed ( kidney , heart , lung , and salivary glands ) compared to those of wt MCMV . Indeed , at this time point , viral titers were below the assay's detection limit in a number of the MCMVΔm154-infected animals ( Figure 8B ) . Comparable results were obtained when mice infected with MCMVΔm154Int were analyzed at day 4 post-infection ( Figure 8C ) . Thus , we can conclude that MCMVs lacking the m154 gene are attenuated in all major organs targeted during MCMV infection . These results , showing the in vivo effects of m154 as early as 4 days post-infection , together with the in vitro data pointing to a contribution of m154 impairing NK degranulation against the infected macrophages , were highly indicative of an MCMV evasion mechanism involving NK cell immune surveillance . Therefore , we decided to examine whether the reduced attenuation of MCMVΔm154 was a consequence of its enhanced susceptibility to NK cells during the in vivo infection . Mice were specifically depleted of NK cells by treatment with rabbit antiserum to asialo GM1 . Four days after infection with 8×105 PFU of wt MCMV or MCMVΔm154 , mice were sacrificed and assayed for viral loads in spleen and liver , the two predominant organs in which NK cells have been reported to intervene in the control of MCMV . As expected , all mice treated with anti-asialo GM1 antibody had significantly higher viral titers in their spleens and livers as compared to the corresponding untreated control mice ( Figure 8C ) . However , the extent to which these viral titers were elevated following NK ablation was considerably superior in the MCMVΔm154-infected animals ( in particular in the spleen , 101-fold ) than in animals infected with wt MCMV ( 9-fold in spleen ) . Thus , this substantial restoration of MCMVΔm154 replication demonstrates that m154 promotes MCMV growth in vivo by subverting NK cell responses .
For an effective immune response against many viral infections , antigen-presenting cells such as dendritic cells and macrophages must expose a concerted repertoire of receptors that alert T and NK cells for their efficient activation . In this context , distortion of the surface receptor content is a maneuver widely adopted by numerous viruses to elude the immune system and secure an optimal milieu for their replication and dissemination . In this study we show that several cell-surface molecules of the SLAM family , which operate as co-signalling molecules triggering distinct signal-transduction networks in T , NK and antigen-presenting cells , are targeted by MCMV . Notably , CD48 , CD84 , CD229 and Ly108 get differentially restricted from the cell surface within the window of time it takes for the virus to complete its life cycle and produce productive progeny . Hence , the fact that CMV might have an active interest in interrupting SLAM interactions through the downregulation of the specific receptors/ligands in the infected cell indicates that , at least for the four SLAM members analyzed in our study , engagement of the corresponding receptors/counter receptors should exert prevailing activating signals in key immune cells during infection . In this study , we decided to further explore in more detail the loss of CD48-surface expression after MCMV infection . CD48 , a GPI-anchored molecule with broad expression in hematopoietic cells , is a SLAM receptor not involved in the homophilic interactions distinctive of this family , its natural ligand being CD244 [17] . Accordingly , reduction of CD48 from the surface of MCMV-infected macrophages leads to a drastic decrease in CD244 binding compared to that observed in mock-infected cells . By screening a battery of MCMV deletion mutants , we identified m154 as the viral downregulator of CD48 . Thus , deletion of both the complete m154 sequence or of an internal part of this viral gene from the MCMV genome is sufficient for restoring the surface levels of CD48 back to those found in non-infected cells . Moreover , m154 ectopically expressed within the MCMV genome leads to a significant decrease of CD48 on the surface of the infected macrophage . The m154 ORF encodes a type I transmembrane protein containing a remarkable mucin-type extracellular region . By generating a specific mAb ( m154 . 4 . 113 ) against the cytoplasmic tail of this protein , we found that m154 expression is initiated in the early phase of infection and continues throughout the infection cycle , a time frame that is concomitant with the progressive downregulation of CD48 in the infected macrophage . In addition , we show that this viral protein preferentially localizes on the surface of the infected cell . m154 belongs to the m145 family of glycoproteins [35] , despite not presenting the MHC class I protein fold characteristic of some family members . Notably , several of the ten members ( m17 , m145 to m158 ) that comprise this family have been reported to perform immunoevasive activities [37] . In particular , the m145 , m152 , and m155 proteins each downregulate one or more ligands of the activating NK cell receptor NKG2D ( H60 , RAE1 , or MULT-1; [38]–[42] ) . Additionally , m152 causes intracellular retention of MHC class I molecules [43] , while m155 reduces cell-surface expression of the costimulatory molecule CD40 [44] . Finally , the m157 protein interacts with Ly49 NK cell receptors and engages both NK activating ( Ly49H ) and inhibitory receptors ( Ly49I ) [36] , [45] , [46] . Thus , m154 can be now added to the group of molecules within the m145 family that operates as an immunoevasin . While it is important to point out that m154 is able to selectively downregulate CD48 , since surface molecules like CD86 , Ly108 , or CD84 , which are used as specificity controls , are not affected by m154 , we can not exclude , however , that this viral protein also has a multifunctional nature and targets additional immune receptors . In terms of delineating the mechanism that leads to the loss of surface CD48 , we determined that it does not occur at transcriptional level . Instead , our data show that m154 appears to be majorly causing CD48 degradation . The viral protein leads to a major reduction in the total cellular amount of CD48 . Through Western blot analysis and immunofluorescence microcopy , we found that treatment of infected cells with either the serine and cysteine protease inhibitor leupeptin or the proteasome inhibitor MG-132 stabilizes CD48 expression in a certain degree , suggesting that both the lysosomal and proteasomal degradation pathways play a role in the downregulation of CD48 . In addition , using these proteolysis inhibitors , co-localization of m154 and CD48 could be appreciated in the wt MCMV-infected macrophages . Interestingly , the m154 cytoplasmic tail displays a motif that has been implicated in lysosomal targeting , and two overlapping recognition sites for the adaptor protein AP-2 , which is involved in clathrin-dependent endocytosis . While the detailed mechanism by which m154 operates remains to be elucidated , it does not seem to involve the overall trafficking of GPI-linked receptors , as cell-surface expression of other GPI-anchored membrane molecules , such as CD55 , are not affected by this viral protein ( data not shown ) . m154 does not have a counterpart in any other of the CMV species whose genome has been sequenced so far . However , akin to MCMV , we have previously reported that CD48 is also downregulated in HCMV-infected macrophages [47] . Therefore , each CMV might have evolved its own CD48-specific inhibitor , as yet to be identified for HCMV , emphasizing the importance of targeting this molecule to evade NK cell recognition during infection . In addition , and similarly to MCMV , we found that other SLAM receptors are markedly reduced from the cell surface of macrophages upon HCMV infection ( Angulo , unpublished observations ) . Whether the overall downmodulation of SLAM receptors in the infected cell is an inherent and unique property of CMVs , reflecting selection pressures faced in their specific niches , or whether this might be used by other viruses as an immune evasion mechanism , remains to be explored . Notably , CD48 and NTB-A have been also reported to be negatively regulated by HIV , leading to impaired NK cell recognition and lysis of the infected CD4+ T cells , and being the viral accessory protein Vpu identified as the NTB-A downregulator [27] , [48] . Increasing evidence indicates that CD244 contributes to the regulation of both NK cell antiviral activity and virus-specific CD8+ T cell functionality in humans and mice [49]–[51] . Engagement of CD244 by CD48 in the NK cell results in the recruitment and clustering of the receptor into lipid rafts , the phosphorylation of the ITSMs within its intracellular tail , and the subsequent association with the adapter molecule SAP [18] , [52]–[54] . This triggers a signalling cascade , leading to the formation of the NK cell synapse , which is characterized by the polarized release of cytolytic granules containing perforin and granzymes . The NK cell synapse is most likely critical for activated NK cells to interact in a productive manner with MHC class I-negative target cells and induce potent cell cytotoxicity [55] . On the other hand , CD244 can inhibit NK cell activation in the absence of functional SAP , such as occurs in cells from patients with X-linked lymphoproliferative syndrome [56] . Taken together , these observations indicate that CD244 and SAP modulate the activity of normal NK cells . Here , we specifically show that disruption of the m154 gene in MCMV leads to an enhanced antiviral NK cell response in vitro . In particular , this viral protein limits NK degranulation capacity against MCMV-infected cultured macrophages . Moreover , we present that the NK cell response to cells infected with the MCMV lacking the m154 gene can be partially inhibited by preincubation of the infected macrophages with the CD244-Fc fusion protein . Hence , we infer from these results that by downregulating CD48 , m154 may help protect MCMV-infected cells from NK killing . We cannot discard the possibility , however , that m154 might be also be capable of exerting other functions that contribute to these effects . As expected , the m154 gene is not required for replication in vitro and an MCMV lacking m154 has not an altered growth phenotype in cultured MEFs or macrophages . It is of note that parental and mutant viruses used throughout the study do all derive from MCMV-BAC pSM3fr , containing an mck-2 frameshift mutation associated with reduced ability to infect macrophages and a diminished capacity to attract leukocytes [57] , [58] . However , the fact that all of these recombinant MCMVs have the same pSM3fr background makes them comparable at the level of the MCK-2 phenotype . In contrast to the in vitro observations , the severity of the infection of viruses that do not express m154 was significantly impaired in vivo , where they exhibited a substantial restricted growth in all organs analyzed . By day 4 post-infection , the differences in splenic and liver growth between wt MCMV and MCMVΔm154 were around 30- and 10-fold , respectively , consistent with m154 counteracting NK cell responses , which are crucial to the early control of MCMV replication . Accordingly , we show that mice depleted of NK cells with an antibody to asialo GM1 , a glycosphingolipid present at high concentrations in this cell population , selectively improved the in vivo replication of the deletion mutant , confirming that the mechanism by which m154 exerts its protective role is NK cell dependent . Because CD244 expression is not restricted to NK cells , the impact of m154 might have implications that extend beyond the regulation of NK cell function . This receptor is also present at lower levels on other cytotoxic cells , such as CD8+ T cells , γδ T cells , basophils , and eosinophils . In particular , upon interaction with CD48 , CD244 helps initiate signalling and cellular cytotoxicity in CD8+ T cells [49] . Hence , one could speculate that other non-NK cell-related mechanisms might be also contributing to the net protective effects of m154 in vivo . However , the fact that NK cell depletion results in the near complete rescue of the m154-deleted MCMV growth phenotype in vivo indicates that , at least under the conditions of early acute infection analyzed here , the potential of this viral protein to influence processes mediated by other immune cell subsets might be relatively minor . It remains to be determined whether additional effects of m154 could be of relevance later during the infection or in other scenarios the virus might encounter . In summary , our study presents the SLAM family of immunoreceptors as a novel target of manipulation by CMV , adding to the diversity of molecular strategies incorporated by this pathogen to escape immune detection . We have identified for the first time a herpesviral gene implicated in the downregulation of the SLAM member CD48 , and documented its protective role in vivo by counteracting NK cell responses . The knowledge gained from the findings reported in this manuscript will contribute to a better understanding of the complex host-CMV interactions and provide additional insights into the functioning of the SLAM receptors in viral immunity . Finally , the identification of novel players that increase the CMV burden early on during infection could prove helpful for the future development of antiviral reagents .
All procedures involving animals and their care were approved ( protocol number CEEA 308/12 ) by the Ethics Committee of the University of Barcelona ( Spain ) and were conducted in compliance with institutional guidelines as well as with national ( Generalitat de Catalunya decree 214/1997 , DOGC 2450 ) and international ( Guide for the Care and Use of Laboratory Animals , National Institutes of Health , 85-23 , 1985 ) laws and policies . The cell lines NS-1 ( mouse myeloma ) and 300 . 19 ( mouse pre-B ) were obtained from the American Type Culture Collection . Cells were grown in RPMI 1640 medium ( GIBCO-BRL , Paisley , UK ) supplemented with 10% fetal bovine serum ( Sigma Aldrich , St . Louis , MO ) , 100 U/ml penicillin , 100 U/ml streptomycin , 1 mM sodium pyruvate , and 2 mM L-glutamine ( GIBCO-BRL ) . 300 . 19 cells were maintained in media supplemented with 0 . 05 mM 2-mercaptoethanol ( GIBCO-BRL ) . Primary mouse embryonic fibroblasts ( MEFs ) were cultured in Dulbecco's modified Eagle's medium ( DMEM; GIBCO-BRL ) supplemented as indicated above . Primary macrophages were elicited from peritoneal exudate cells ( PECs ) following i . p . injection of 1 ml of 3% thioglycollate ( Sigma Aldrich ) into BALB/c mice . PECs were removed by peritoneal lavage . Cells were plated out at 2×105 cells/ml in supplemented RPMI 1640 medium , and incubated for 2 h at 37°C , 5% CO2 , after which nonadherent cells were washed away with phosphate buffered saline ( PBS ) . Macrophage preparations were confirmed by flow cytometry using the markers F4/80 and CD11b ( about 95% were F4/80+CD11b+ ) . NK cells were obtained from mouse spleen using the mouse NK cell isolation kit II ( Miltenyi Biotec , Bergisch Gladbach , Germany ) on an AutoMACS ( Miltenyi Biotec ) . The BAC-derived MCMV , MW97 . 01 , based on the MCMV Smith strain ( ATCC VR-1399 ) and referred to here as wt MCMV [59] , and the MCMV-GFP recombinant virus , a derivative of MW97 . 01 carrying the GFP gene [32] were used as parental viruses throughout the study . Recombinant strains MCMV-GFPΔ6 lacking genes from m144 to m158 ( referred to here as MCMV-GFPΔm144-m158 ) , MCMV-GFPΔ6S1 lacking genes from m144 to m148 ( referred to here as MCMV-GFPΔm144-m148 ) , MCMV-GFPΔ6S2 lacking genes from m149 to m153 ( referred to here as MCMV-GFPΔm149-m153 ) , MCMV-GFPΔ6S3 lacking genes from m154 to m157 ( referred to here as MCMV-GFPΔm154-m157 ) , MCMV-GFPΔm153-m154 lacking genes m153 and m154 ( referred to here as MCMV-GFPΔm153-m154 ) , and MCMV-Dm155Dm157FRT lacking genes from m155 to m157 ( referred to here as MCMV-GFPΔm155-m157 ) have been described previously [33] , [40] . For the generation of recombinant MCMV lacking m154 ORF ( referred to here as MCMVΔm154 ) a kanamycin resistance ( KanR ) cassette was amplified from the plasmid pGP704 with primers Dm154Fw ( 5′- CCC GCC AAT CAC ATT CAC GAG GGG GTG CTC CGA GAT ACG GTC TCG ACC ACA GGA CGA CGA CGA CAA GTA -3′ ) and Dm154Rv ( 5′- CAC ATA AGA CTC GTC ATA ACC TTC CCC GAG TGC CAC CTC CCC ACC CTT ATC GTC TCA GGA ACA CTT AAC -3′ ) ( underlined letters are homologous to the MCMV genome ) . For the generation of recombinant MCMV lacking only ORF m154 and without affecting ORFs m154 . 3 and m154 . 4 described in Tang et al . [34] , referred to here as MCMVΔm154Int , a KanR cassette was amplified from the pGP704 with following primers: Dm154bFw ( 5′- CCG CTG CGG ACG CGA TCT CTT CGG CAA CCC CTA GTG CAG GTG CCG TTA GGA CGA CGA CGA CAA GTA A -3′ ) and Dm154Rv . PCR fragments were inserted into the m154 ORF of the MCMV BAC MW97 . 01 by red-α , -β , -γ-mediated recombineering [60] . Subsequently , the KanR cassette was excised by FLP recombinase . For MCMVm154Ectop , the m154 ORF plus sequences containing the putative promoter and the polyadenylation signal were PCR amplified using primers m154-ek . for ( 5′- CGC GTT AAC CCC GTA TAA ACA CCG CAC CAG A -3′ ) and m154-ek . rev ( 5′- CGC AGA TCT ATG TCC TGA CAG ATT ATC GTG GT -3′ ) with MW97 . 01 DNA as a template . The PCR product was cloned into pOri6K-F5 [61] and then the m154 sequences together with the adjacent KanR cassette ( flanked by mutant [F5] FRT sites ) was amplified using primers m154-ins . for ( 5′- AAC CAC GGG TTC TTT CTC TTG ACC AGA GAC CTG GTG ACC GTC AGG AAG AAG ATT CAG TGA CAG GAA CAC TTA ACG GCT GA -3′ ) and m154-ins . rev ( 5′- GTC CGA TGA ATA AAA CCT CTT TAT TTA TTG ATT AAA AAC CAT GAC ATA CCT CGT GTC CTC CCC GTA TAA ACA CCG CAC CA -3′ ) . The m154 sequences and the KanR cassette were inserted downstream of the ie2 ( m128 ) ORF of the MCMVΔm154 BAC by red-α , -α , -γ-mediated recombineering followed by excision of KanR as described above . The integrity of the MCMVΔm154 , and MCMVΔm154Int , and MCMVm154Ectop genomes was verified by restriction analysis and sequencing . Viral stocks were prepared by infecting MEFs at low moi . Cell supernatants were recovered when maximum cytopathic effect was reached , and cleared of cellular debris by centrifugation at 1 . 750× g for 10 min . Viral titers were determined by standard plaque assays on MEFs . Peritoneal macrophages were infected with parental MCMV , MCMV-GFP or derived deletion mutants at an moi ranging from 0 . 2 to 10 . Adsorption was for 2 h at 37°C , 5% CO2 , including a centrifugal enhancement of infectivity step [62] . Cells were then washed in PBS before fresh medium was added . The percentage of macrophages infected by recombinant viruses not expressing GFP was estimated by indirect immunofluorescence 24 h after infection using the anti-MCMV IE1 mAb Croma 101 followed by goat anti-mouse IgG Alexa Fluor-488 . UV-inactivation of virus was performed using a UV crosslinker ( HL 2000 hybrilinker , UVP [254 nm UV] , Upland , CA ) for 3 min at 360 mJ/cm2 . The anti-m154 mAb ( clone m154 . 4 . 113 , IgG2a ) was generated by fusing an NS-1 myeloma cell line with spleen cells from a BALB/c mouse immunized three times with the synthetic peptide corresponding to the intracellular tail of m154 ( HRWEDDKGGEVALGEGYDESYV ) conjugated to KLH ( Proteogenix , Oberhusbergen , France ) . The hybridoma was subcloned at least three times . The following mAbs were obtained from Biolegend ( San Diego , CA ) : anti-mouse CD48-PE , anti-mouse CD48-Alexa Fluor 488 , CD84-PE , Ly108-PE , H2-PE , CD150-PE , CD86-PE , F4/80-pacific blue , CD49b-PE/Cy7 , CD107a-Alexa Fluor 488 , CD3-Alexa Fluor 647 , and streptavidin-PE . Biotin anti-mouse 2B4 , recognizing mouse CD244 , anti-mouse CD229-APC , CD11b-PE , and Gr-1-APC were purchased from Becton Dickinson Bioscience ( San Diego , CA ) , anti-mouse CD48 ( HM48-1 ) from Santa Cruz Biotechnology ( Santa Cruz , CA ) , and IgM-FITC from Southern Biotech ( Birmingham , AL ) . Anti-mouse IgG Alexa fluor 555 and anti-mouse IgG-Alexa Fluor 488 were purchased from Invitrogen ( Carlsbad , CA ) . Anti-rabbit IgG and anti-Armenian hamster IgG ( H+L ) labelled with horseradish peroxidase ( HRP ) were obtained from Jackson Immuno Research Laboratories ( West Grove , PA ) , and anti-mouse IgG ( H+L ) labelled with HRP from Promega ( Heidelberg , Germany ) . Biotin-conjugated anti-HA and anti-mouse β-actin were purchased from Sigma Aldrich . Isotype control IgGs directly conjugated with the corresponding fluorochromes were obtained from Immunotools ( Friesoythe , Germany ) . The MCMV IE1 specific mAb Croma 101 has been previously described [63] . A murine CD244-Fc fusion protein containing the CD33 leader peptide and the Fc region of human IgG1 was obtained by inserting sequences corresponding to the CD244 ectodomain into the mammalian expression vector signal pIg-Tail ( R&D Systems , Wiesbaden , Germany ) as previously described [64] . The construct corresponding to the Ig fusion protein was subcloned into the expression vector pCI-neo . An NS-1 stable transfectant secreting the CD244-Fc fusion protein was obtained by electroporation and selection using 1 . 2 mg/ml geneticin ( G418 ) ( GIBCO-BRL ) . Eight million cells were electroporated ( 280V , 950 µF ) with 8 µg of linearized DNA using the Gene Pulser II Apparatus ( Bio-Rad , Hercules , CA ) . The transfected cells were plated in flat-bottom , 96-well tissue culture plates ( Costar , Corning , NY ) by limiting dilution and the clone producing the highest amounts of fusion protein was cultured in INTEGRA CL 1000 flasks ( Integra Biosciences AG , Chur , Switzerland ) . The supernatant containing the fusion protein was purified using the Affi-Gel Protein-A MAPS II kit ( Bio-Rad ) . An m154 N-terminal HA fusion protein ( HA-m154 ) was constructed using a PCR product of m154 obtained with primers m154BglII ( 5′- CCA AGA TCT TTG GGT CGT TTA GAG CTT -3′ ) containing a BglII restriction site , and m154PstI ( 5′- CTC CTG CAG TCA CAC ATA AGA CTC GTC -3′ ) containing a PstI restriction site , and DNA extracted from MCMV virions as a template . The PCR product was inserted into the pGEMT vector and a BglII-PstI fragment corresponding to the m154 gene without the signal peptide was then excised and inserted in frame with the HA at the N-terminal end of the m154 protein into the mammalian expression vector pDisplay ( Invitrogen ) treated with BglII and PstI . 300 . 19 stable transfectants were obtained using the same protocol as indicated for NS-1 cells , except that plasmids expressing HA-m154 or the corresponding empty pDisplay vector were transfected using the Amaxa Cell Line Nucleofector Kit V ( Amaxa AG , Koeln , Germany ) according to the manufacturer's protocol , and selection performed with 0 . 8 mg/ml of G418 . Flow cytometry was performed using standard procedures . Fc fusion protein stainings were performed using 2 µg of biotinylated Ig fusion protein followed by incubation with streptavidin-PE . Samples were analyzed using a FACSCanto II ( BD Biosciences ) flow cytometer and processed with the accompanying FlowJo software ( Tree star Inc , Ashland , OR ) . Ten thousand cells were counted for each sample . Cell viability was measured using the LIVE/DEAD Fixable Violet Dead Cell Stain Kit ( Invitrogen ) according to the manufacturer's instructions . The FACSCalibur ( BD Biosciences ) was used for analysis of the cellular influx to the peritoneal cavity . Total RNA was isolated from peritoneal macrophages either uninfected or MCMV-infected for 72 h by the TRIzol method ( Invitrogen ) . RT-PCR was then carried out using the SuperScript III First-strand Synthesis System for RT-PCR ( Invitrogen ) according to the manufacturer's protocol . Briefly , RNA samples were treated with RNase-free DNase I ( Promega ) for 30 min at 37°C , and the DNase was inactivated at 65°C for 10 min . The RNA was reverse transcribed using oligo ( dT ) primers at 50°C for 50 min , and reactions were terminated by heating at 85°C for 5 min . The reverse-transcribed products were treated with RNase H for 20 min at 37°C and amplified using specific primers . Primers m154For ( 5′- CTT GGA TCC ATG CGG GCG ATG TTA CGG -3′ ) and m154Rev ( 5′- CTC GGA TCC CAC ATA AGA CTC GTC ATA -3′ ) were used to amplify a 1116-bp fragment within the MCMV m154 gene , primers mCD48For ( 5′- ATG TGC TTC ATA AAA CAG GG -3′ ) and mCD48Rev ( 5′- TTG TCA GGT TAA CAG GAT CCT GTG -3′ ) were used to amplify a 726-bp fragment within the murine CD48 gene , and primers β-actinFor ( 5′- TAT CCT GAC CCT GAA GTA CC -3′ ) and β-actinRev ( 5′- TCA TCT TTT CAC GGT TGG CC -3′ ) were used to amplify a 170-bp fragment within the murine β-actin gene . PCRs were performed under the following conditions: 1 cycle at 94°C for 5 min; 30 cycles of 1 min at 94°C , 1 min at 58°C , and 1 min at 72°C; and 1 cycle at 72°C for 10 min . Control reactions carried out in the absence of RT were used to assess the specific detection of RNA . Amplified products were separated on a 1% agarose gel and visualized by ethidium bromide staining . For Western blot analysis , peritoneal murine macrophages , either mock-infected or infected with MCMV at an moi of 10 were used . For selective expression of viral immediate-early proteins , cells were incubated from 30 min prior to infection to 4 h post-infection in the presence of CHX ( 100 µg/ml; Sigma Aldrich ) , followed by incubation in the presence of actinomycin D ( 10 µg/ml; Sigma Aldrich ) for another 12 h . Selective expression of early genes was carried out by treatment of the cells with PPA ( 250 µg/ml; Sigma Aldrich ) for 72 h . For proteolysis inhibition experiments , cells were treated with 75 µM MG-132 ( Sigma Aldrich ) or 250 µM leupeptin ( Sigma Aldrich ) for 6 h and 24 h , respectively , before harvesting . Under the conditions used , MG-132 and leupeptin did not generally affect cell viability as assessed by trypan blue cell staining . At the indicated times after infection for each specific case , samples were lysed in protein sample buffer and boiled for 5 min . Cell lysates were subjected to SDS-PAGE in 10% acrylamide gels and subsequently transferred to nitrocellulose membranes ( Protran , Whatman Schleicher & Schuell , Germany ) . Equal quantities of total protein were analyzed per lane . Membranes were probed using the mAb anti-m154 ( clone m154 . 4 . 113 ) , mAb anti-IE1 Croma 101 , and an anti-mouse IgG ( H+L ) HRP as a secondary antibody , and mAb anti-mouse CD48 ( HM48-1 ) followed by a HRP-conjugated goat anti-Armenian hamster IgG ( H+L ) antibody . β-actin was detected using the mAb anti-β-actin and an HRP-conjugated goat anti-rabbit IgG as a secondary antibody . Blots were developed using a SuperSignal West Pico Chemiluminescent Substrate ( Pierce , Rockford , IL ) according to the manufacturer's protocol . Peritoneal murine macrophages , mock-infected or MCMV-infected at different mois , were cultured on glass coverslips in 24-well tissue culture plates . When indicated , cultures were exposed to proteolysis inhibitors as indicated for the Western blot analysis . At specific time points after infection , the cells were washed in PBS and fixed and permeabilized using ice-cold methanol and 0 . 3% Triton X-100 ( for IE1 detection ) , or fixed in ice-cold acetone ( for m154 and CD48 detection ) , and subsequently blocked with 1% bovine serum albumin ( Sigma Aldrich; for IE1 detection ) or with 20% rabbit serum ( Linus ) and 6% fetal bovine serum in PBS ( for m154 and CD48 detection ) . The cells were stained with anti-m154 mAb ( clone m154 . 4 . 113 ) , or with MCMV IE1 mAb Croma 101 , using as secondary antibodies a goat anti-mouse IgG ( H+L ) Alexa Fluor 555 or Alexa Fluor 488 , or directly with anti-mouse CD48-Alexa Fluor 488 . Nuclei were counterstained with DAPI reagent ( Invitrogen ) . The samples were mounted in ProLong Gold antifade reagent ( Invitrogen ) . Fluorescence images were obtained using a Nikon Eclipse E600 microscope ( Nikon , Tokyo , Japan ) or an inverted Leica DMI6000B microscope and the LAS AF software from Leica Microsystems ( Wetzlar , Germany ) . Peritoneal macrophages were surface-labeled with biotin ( Sigma Aldrich ) and lysed in protein sample buffer . Cell lysates were precleared 3 times for 30 min using protein G Sepharose ( GE Healthcare ) and incubated overnight with anti-m154 mAb and protein G Sepharose . Immunoprecipitates were washed , eluted , subjected to SDS-PAGE in 10% acrylamide gels and transferred to nitrocellulose membranes . Membranes were probed with streptavidin-POD conjugate ( Roche Diagnostics GmbH , Mannheim , Germany ) and blots developed as for the Western blot analysis . Multi-step growth in vitro was analyzed by infecting MEFs or peritoneal macrophages in 24-well plates with wt MCMV or MCMVΔm154 at an moi of 0 . 025 and 0 . 2 , respectively . After a 2 h adsorption period , cells were washed with PBS and incubated in the corresponding medium supplemented with 3% fetal bovine serum . At specific time points after infection , the amount of extracellular ( MEFs ) or cell-associated ( macrophages ) infectious virus present in the cultures was determined as previously described [65] by a standard plaque assay on MEFs . NK cell degranulation was evaluated using the CD107a mobilization assay . Cultures of peritoneal macrophages , either mock-infected or infected with 10 PFU/cell of MCMV or MCMVΔm154 for 72 h , were incubated for 5 h at 37°C with purified NK cells in the presence of monensin ( BD Biosciences ) and anti-CD107a-Alexa Fluor 488 , at an effector-to-target cell ratio ( E/T ) of 1∶1 . NK cells treated with 0 . 5 µg/ml ionomycin ( Sigma Aldrich ) and 50 ng/ml PMA ( Sigma Aldrich ) were used as a positive control for degranulation . Cells were then washed in PBS supplemented with 2 mM EDTA , stained for 30 minutes at 4°C with anti-CD49b-PE/Cy7 , recognizing DX5 , and analyzed by flow cytometry . When stated , MCMVΔm154-infected macrophages were pre-incubated with 10 µg/ml of the indicated Fc fusion protein for 30 min at 37°C , cultures washed , and subjected to the CD107a mobilization assay using an E/T ratio of 0 . 5∶1 . Seven-week-old BALB/c . ByJ female mice were obtained from Harlan ( Netherlands ) and housed in the vivarium ( University of Barcelona ) under specific-pathogen-free conditions . Mice were i . p . inoculated with 5×105 or 2×106 PFU of tissue culture-propagated wt MCMV , MCMVΔm154 , or MCMVΔm154Int recombinants . When specified , NK cells were depleted by i . p . injection of rabbit antiserum to asialo GM1 ( Wako Pure Chemical Industries , Osaka , Japan ) at a concentration of 25 µg per mouse , one day before infection and on day 2 after infection . The efficacy of depletion was assessed by cytofluorometric analyses of spleen cells using an anti-mouse pan-NK cell mAb CD49b-PE/Cy7 . At designated times after infection , mice were sacrificed , and specific organs were removed and harvested as a 10% ( weight/volume ) tissue homogenate . Tissue homogenates were sonicated and centrifuged , and viral titers from the supernatants were determined by standard plaque assays , including centrifugal enhancement of infectivity [62] on MEFs . In experiments evaluating the cellular influx to the peritoneal cavity , mice were sacrificed 2 days after infection and cells present in the peritoneal cavity harvested with 5 ml of PBS . Total number of cells were determined with a cell counter , and stained with a combination of mAbs CD11b-PE and Gr-1-APC or mAbs CD3-Alexa Fluor 647 and IgM-FITC to distinguish the different cellular subsets ( macrophages [CD11b+ Gr-1−] , neutrophils [CD11b+ Gr-1+] , T lymphocytes [CD3+] or B lymphocytes [IgM+] ) . The number of peritoneal macrophages infected in vivo was assessed by IE1 staining of peritoneal lavage-derived macrophages of mice infected for 16 h with 2×106 PFU . The signal peptide cleavage site and the transmembrane region were predicted by using the SignalP 4 . 1 ( www . cbs . dtu . dk/services/SignalP/ ) and the TMHMM 2 . 0 ( www . cbs . dtu . dk/services/TMHMM-2 . 0/ ) servers , respectively . The N-glycosylation and O-glycosylation motifs of the protein were identified by using the NetNGlyc 1 . 0 Server , ( www . cbs . dtu . dk/services/NetNGlyc ) , and the NetOGlyc 4 . 0 Server ( www . cbs . dtu . dk/services/NetOGlyc/ ) , respectively . Analyses were performed with GraphPad Prism software ( version 3 . 03 , GraphPad Software , San Diego , CA ) . Statistical significance of viral titers between experimental groups was determined with the Mann-Whitney test ( two-tailed ) . P-values less or equal to 0 . 05 ( * ) , 0 . 01 ( ** ) , and 0 . 001 ( *** ) were considered statistically significant .
|
Cytomegalovirus ( CMV ) has developed diverse tactics to elude the host immune response and guarantee its survival . The signalling lymphocyte-activation molecules ( SLAM ) family of receptors encompasses a number of adhesion molecules expressed on the surface of leukocytes that play critical roles in both innate and adaptive immunity . In this study , we report that murine CMV drastically reduces the expression of several SLAM family receptors at the cell surface of infected macrophages , most likely as part of its immunoevasion mechanisms . We have identified a murine CMV gene product ( m154 ) that downregulates CD48 , a SLAM family member that functions as a ligand of CD244 , a molecule involved in the regulation of natural killer ( NK ) and cytotoxic T cell functions . We show that during infection , m154 targets CD48 for degradation . Moreover , this viral protein contributes to increased MCMV growth during acute infection in the mouse by protecting against NK cell mediated surveillance . These findings are important for better understanding CMV pathogenesis , and provide a novel example of host innate immune subversion by CMV .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"viral",
"immune",
"evasion",
"virology",
"immunology",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"immunomodulation"
] |
2014
|
Cytomegalovirus m154 Hinders CD48 Cell-Surface Expression and Promotes Viral Escape from Host Natural Killer Cell Control
|
Dysregulation of iron metabolism in cancer is well documented and it has been suggested that there is interdependence between excess iron and increased cancer incidence and progression . In an effort to better understand the linkages between iron metabolism and breast cancer , a predictive mathematical model of an expanded iron homeostasis pathway was constructed that includes species involved in iron utilization , oxidative stress response and oncogenic pathways . The model leads to three predictions . The first is that overexpression of iron regulatory protein 2 ( IRP2 ) recapitulates many aspects of the alterations in free iron and iron-related proteins in cancer cells without affecting the oxidative stress response or the oncogenic pathways included in the model . This prediction was validated by experimentation . The second prediction is that iron-related proteins are dramatically affected by mitochondrial ferritin overexpression . This prediction was validated by results in the pertinent literature not used for model construction . The third prediction is that oncogenic Ras pathways contribute to altered iron homeostasis in cancer cells . This prediction was validated by a combination of simulation experiments of Ras overexpression and catalase knockout in conjunction with the literature . The model successfully captures key aspects of iron metabolism in breast cancer cells and provides a framework upon which more detailed models can be built .
This section provides biological background about iron metabolism and its connection to some oncogenic pathways .
We describe a discrete dynamic model of the network in Fig 1 , based on an encoding of the regulatory logic for each node through a “logical” update rule . This type of model is qualitative , in the sense that each species can assume a finite set of states rather than quantitative concentrations of molecular species . For this study , we adopted a ternary logic , an extension of Boolean logic . Our choice of ternary logic was motivated by the fact that iron levels cannot be viewed as either ON = 1 or OFF = 0 . The iron homeostasis pathway is the major focus of our study and both low and high levels of iron are detrimental , so that it is tightly controlled . Additionally , IRP2 at both low and high activity levels does alter the iron pathway [68] . With only two states it would not be clear when IRP2 operates at low activity levels , as it would be represented the same way as normally active protein . For our model to be able to differentiate between iron homeostasis ( normal levels of iron ) and low/high iron levels as well as activities/concentration levels of various proteins , we chose to represent each species by three levels: low , normal and high . In the language of logical models the state of a particular species is described by 0 if the species is low/inactive , by 1 if at normal/intermediate activity , and by 2 if high/active . In analogy to the Boolean formalism , we can compute the future state of a species at time step t + 1 using the states of other species at time step t . Fundamental OR and AND gates for two species X and Y are defined as max{X , Y} and min{X , Y} , respectively , where X , Y ∈ {0 , 1 , 2} . To differentiate from the Boolean OR and AND gates , we denote these gates by Max and Min , respectively . The NOT gate ( denoted here by X ¯ ) is defined by inverting the input , i . e . , leaving 1 unchanged and inverting 0 and 2 . For a concrete example , consider heme in Fig 1 . It is produced through ALA synthase ( ALAS1 ) but inhibited by HO-1 . Then the logical function ( update rule ) that predicts how much heme is present at time t + 1 can be computed as follows: heme ( t+1 ) =Min ( ALAS1 ( t ) , HO−1¯ ( t ) ) . This means that , if HO-1 was 0 ( low ) and ALAS1 was 2 ( high ) at time t , then heme will be 2 ( high ) at time t + 1 . Based on the biological knowledge described in the previous section , we translated the interactions of the normal cell network into logical functions ( see Table 1 ) . One caveat about logical models that is not present for Boolean models is that species can change for example from a low state to a high one in one time step , skipping intermediate concentrations . This is biologically unrealistic . Thus , to address the continuity issue we have also implemented a methodology commonly used for logical models that takes into account the previous state of the regulated species ( see [69] for details ) . For purposes of simulation , we converted the logical rules into polynomial functions to obtain a so-called polynomial dynamical system ( PDS ) . A description of the construction of the PDS and the entire system can be found in the Materials and Methods section and in the supplemental file S1 PDS , respectively . To analyze the dynamic properties of the model we simulated the entire state space and computed the basins of attraction of the system . For this purpose , we used an encoding of the model as a polynomial dynamical system , as described above , and customized scripts written in Perl and Python ( see Materials and Methods section ) . The size of the model’s state space is 324 = 282 , 429 , 536 , 481 , where 24 is the number of species in the network and 3 is the number of states ( low , medium , high ) per species . We employed a synchronous update schedule for the species in the network; all species were updated simultaneously based on the states of their input species at the previous time step . Each state leads to another state , eventually converging to a steady state or a limit-cycle ( a set of recurring states ) , which are called attractors . A collection of initial states that lead to a particular attractor is termed the basin of attraction . Under this scheme , each state belongs to the basin of attraction of only one attractor: a point attractor ( steady state ) or a cycle attractor ( limit-cycle ) . These attractors correspond to different phenotypes in the biological context and can describe various behaviors of the system such as homeostasis . We simulated the normal cell model and also investigated the long-term behavior of this model under different conditions , namely , the effects of knockout ( k/o ) or overexpression ( o/e ) of one or more species . To simulate these experimental conditions , we set the update rule for a particular species to a constant equal to 0 or 2 , respectively . In other words , regardless of the input ( regulators ) , the species of interest will always stay at the chosen level . Our results are summarized in Fig 2 , which shows all the species and their long-term behavior . Simulations were performed by exhaustively enumerating the transitions of the model on all possible 324 states . The normal cell network has no cycle attractors and reaches a unique stable steady state ( point attractor ) indicating that all species are at their respective normal levels regardless of the initial starting state ( Fig 2 top line of the heat map labeled Normal ) .
It is well-known that iron metabolism in breast epithelial cells is differentially regulated as cells transition to malignancy . Determining the causes for this altered phenotype is complicated by the complexity of iron regulation and its connection to several other processes , such as response to oxidative stress and changes in iron consumption [80] , as well as crosstalk with oncogenic pathways . Integrating these different influences on the iron phenotype in normal and malignant cells can benefit greatly from a systematic approach through dynamic mathematical modeling , beyond the network approach taken in [80] . The model presented here is a first step toward a comprehensive understanding of the iron phenotype of cells as it changes in breast cancer . We have chosen to construct a qualitative model of an intracellular iron network ( Fig 1 ) to capture its fundamental dynamic features ( attractors ) . The main reason for our choice of modeling platform is that our current knowledge of the kinetics involved in these different processes as well as mechanisms underlying these complex reactions is very limited , so that a quantitative model , such as a system of ordinary differential equations is more challenging to construct . We have validated our model using both experimental data and information from the literature not used in model construction . In particular , we have experimentally validated the model prediction that IRP2 overexpression in the normal cell network only alters the iron homeostasis pathway , leaving the other model components unchanged . Also , our model agrees with the current literature that overexpression of mitochondrial ferritin ( Ftmt ) increases both IRPs and TfR1 , decreases cytosolic Ft and reduces cytosolic and mitochondrial iron pools [71] . In addition , we have shown that shutting down trafficking of iron into the mitochondria , together with Ras overexpression and Cat reduced bioactivity , does lead to the observed cancer phenotype of the iron homeostasis pathway . However , it might be possible that further refinements of the model can lead to the required phenotype by altering only the oncogenic pathway . Not all known information about the normal and cancer phenotypes can be captured by the model , however . This is likely due to the fact that some key features of this system are not represented completely , such as an iron-sensing regulator in the mitochondria and iron-sulfur cluster ( ISC ) synthesis . It has been suggested that frataxin , a nuclear-encoded mitochondrial protein , may act as an iron-sensing regulator and even function as a switch between heme and ISC synthesis [81–83] . At this stage , one cannot determine whether it is frataxin or some other iron sensor/regulator , but we have suggested the possibility of a mitochondrial iron-sensing node in our current model ( depicted as a question mark in Fig 4 ( a ) ) . This adjusted normal cell model also reaches a unique stable steady state agreeing with our model discussed in the Results section ( Fig 4 ( b ) ) . Additionally , we simulated two more models using the following perturbations: ( i ) knockout of the sensor node and ( ii ) overexpression of Ras and of the sensor node , and low bioactivity of CAT ( 2nd and 3rd rows in Fig 4 ( b ) ) . Interestingly , the sensor node k/o model agrees with experimental data that in frataxin k/o mice heme is decreased , TfR1 is upregulated and iron uptake via Mfrn is increased , leading to cytosolic iron-deficiency and mitochondrial iron overload [84 , 85] . This strongly indicates that there is a sensor/regulator , and thus further refinements of the model can provide insight into mitochondrial iron regulation and utilization , and potentially suggest new experiments that can validate new connections . The latter model produced the same cancer phenotype of the iron homeostasis pathway ( see Eq ( 2 ) ) and also implied that cancer cells have reduced heme biosynthesis . Furthermore , we note that the latter model allows Ftmt and ALAS1 from the iron utilization pathway to have high expression levels ( compare 9th row in Fig 2 to row 3 in in Fig 4 ( b ) ) . While we do not have much evidence about Ftmt in cancer , there are some studies about ALAS1 in lung cancer . It was found that ALAS1 protein levels were substantially increased in non-small-cell lung cancer cells compared to normal cells [86] . This suggests the possibility to expand the cancer phenotype of the iron homeostasis pathway to the iron utilization pathway . Of course , one can simulate a model by setting various proteins to their respective observed levels , but then we gain no information about the drivers that change iron metabolism in cancer . Ideally , we would like to include other pathways implicated in breast cancer to capture different molecular subtypes of breast cancer and iron cancer phenotypes associated with them .
We begin by defining a set of rules that describe various relations between molecular species , from which we then build the entire model . If species X is , inducing species Y ( X → Y ) or species X is inhibiting species Y ( X ⊣ Y ) then we represent these relationships via a transition table as depicted in Table 2 . Notice that inhibition in Table 2 is just a logical NOT gate , denoted here by X ¯ . The other two fundamental gates , OR and AND , for two species X and Y regulating species Z ( X → Z ← Y ) , are defined as max{X , Y} and min{X , Y} respectively , for X , Y ∈ {0 , 1 , 2} , and denoted here by Max and Min . We can express the above gates as polynomials over a finite field on three elements , F 3 . If we limit the exponent of each variable in a polynomial to be less than or equal to 2 , then one can show that any logical rule constructed from these three operations has a unique polynomial representation , using x¯=2+2xMax ( x , y ) =x2y2+x2y+xy2+2xy+x+yMin ( x , y ) =2x2y2+2x2y+2xy2+xy . ( 3 ) One can check that polynomials given by Eq ( 3 ) agree with definitions of fundamental gates as described in the paragraph above , e . g . , max{1 , 2} = 2 and Max ( 1 , 2 ) = ( 1 ) 2 ( 2 ) 2 + ( 1 ) 2 ( 2 ) + ( 1 ) ( 2 ) 2 + 2 ( 1 ) ( 2 ) + 1 + 2 = 2 , where the right-hand side is computed modulo 3 . Various adjustments to the strength of a particular regulation can be made by altering entries in the Table 2 . For example , it has been suggested that IRP1 , when active , contributes less to the regulation of ferritin ( Ft ) than IRP2 [68] ( see Table 3 ) . These tables mean that when IRP2 = 2 ( active ) it will inhibit Ft , whereas when IRP1 = 2 ( active ) it will have a lesser effect on Ft . Thus , we can represent regulation of Ft by IRP2 in Table 3 using Eq 3: IRP2¯=2+2⋅IRP2 . Now , for IRP1 regulating Ft according to this new adjustment , one can also find a polynomial representing Table 3 ( left table ) . For convenience , whenever we use an adjusted regulation we will place an asterisk ( * ) in front of the variable inside the logic gate . *IRP2¯=2+2⋅ ( IRP1 ) 2 . To match current biological knowledge we have adjusted regulation of IRP1 for TfR1 and Fpn as well [68] . The transition table for Fpn is similar to Ft . For IRP1 regulating TfR1: when IRP1 = 2 , then Tfr1 = 1 , while , when IRP1 is 0 or 1 then TfR1 is also 0 or 1 , respectively . Additionally , we modified regulation of Keap1 by Nrf2 to reflect current literature [44] . For Nrf2 regulating Keap1 we have that when Nrf2 = 0 then Keap1 = 1 , while when Nrf2 is 1 or 2 then Keap1 is also 1 or 2 , respectively . To make sure that we preserve continuity ( i . e . , each species changes at most one unit in one time step ) , we are going to employ methodology as described in [69] . The underlying reasoning is that this can be accomplished by taking into account the previous state ( e . g . , concentration or activity ) of the regulated species , in effect adding a self-regulation loop to each network node . The future value of the regulated species under continuity is computed as follows . Let fxi be the update function for xi . To ensure that each variable changes at most 1 unit , define a function h ( xi , fxi ) for the future value of the variable xi: h ( x i , f x i ) = x i + 1 if f x i > x i x i if f x i = x i x i - 1 if f x i < x i ( 4 ) LIP , heme , and ROS do not undergo self-degradation/self-regulation and hence we do not apply continuity to these species . In order to compute final polynomials , we are going to make use of the following property of finite fields: Remark 0 . 1 If h : F p n → F p is any function then there is a polynomial g : F p n → F p so that h ( x ) = g ( x ) for all x ∈ F p n . One can find g by using the following formula , g ( x ) = ∑ c ∈ F p n h ( c ) ∏ j n ( 1 - ( x j - c j ) p - 1 ) , ( 5 ) where h ( c ) is the update function as defined by Eq ( 4 ) , c is a vector of input variables , and the right-hand side is computed modulo p . All of these logic gates , transition tables describing different strength of regulation and continuity , are then appropriately translated into final polynomial functions over a finite field with three elements . These polynomial functions then form what is called a polynomial dynamical system ( PDS ) over a finite field . Below , we fully describe a construction of the update function for ferritin ( Ft ) in our network ( see Fig 1 ) . The entire PDS system can be found in the supplemental file S1 PDS . The attractors of the models were found using 2 algorithms: the attractor finder by random sampling ( Algo . 1 ) that is written in Perl and the attractor finder by iterating over all possible states ( Algo . 2 ) that uses a custom written Python package . The codes can be found at https://github . com/LoLab-VU/LogicalModel . Models that were used for simulations are located in the same directory under NewModels_2015_12_18 and NewModels_2015_8_17 folders . Supporting file S1 Simulations provides additional o/e and k/o simulation results using attractor finder by random sampling . The index for the order of variables is available from row 21 to row 45 . After 3 , 000 random sampling , the basin size of the attractor is specified in the table . The first program requires a model file , the number of states and a sampling size , which is 100 , 000 here . We randomly selected 100 , 000 states and stored the attractor states to have a broad perspective on the possible attractors of a model . It was utilized to test which overexpression and knockout models could be potential cancer models . To ensure that we know all attractors of the models of interest , we ran the second program , which requires a model file and number of states . Optional arguments include start and end states and an option to create images of attractor states . The model file is parsed and compiled into an executable function with Cython [87] . We iterated through all possible states of each model ( 3N ) , storing only the attractor states . Simulations were performed in parallel using mpi4py [88] running on large cluster computers . Algorithm 1 Pseudo code for attractor finder by random sampling 1: procedure For i in 100 , 000 ▹Iterate over 100 , 000 randomly selected states 2: sampled = empty set 3: state = changebase ( random ( 3N ) ) 4: while state ∉ sampled do 5: sampled . add ( state ) 6: state = update ( state ) ▹Update function is the compiled model 7: state . pop ( ) ▹Returns the last state added to sampled Algorithm 2 Pseudo code for attractor finder by iterating over all possible states 1: procedure For i in 3N ▹Iterate over all possible states 2: sampled = empty set 3: state = changebase ( i ) 4: while state ∉ sampled do 5: sampled . add ( state ) 6: state = update ( state ) ▹Update function is the compiled model 7: state . pop ( ) ▹Returns the last state added to sampled MCF10A , non-tumorigenic immortalized human mammary epithelial cells were obtained from the Wake Forest University Comprehensive Cancer Center Tissue Culture Core facility . The cells were maintained in a suggested condition by ATCC . To overexpress IRP2 in MCF10A cells , the lentiviral vector pSL2-IRP2 [68] was applied . Briefly , MCF10A cells were infected with the concentrated viral particles from pSL2-IRP2 and pLS2 empty vector ( as a control ) . The infection efficiencies for both infections were over 90% based on GFP fluorescence in cells . The cell lysates were harvested for subsequent analysis seven days after infection . Western blotting was performed as previously described [68] . Antibodies: GAPDH ( Fitzgerald ) , TfR1 and c-Myc ( Invitrogen ) , IRP2 and EGFR ( Santa Cruz Biotechnology ) , Keap1 ( Cell Signaling Technology ) , HO-1 and IL-6 ( Abcam ) , ferritin H ( [89] ) .
|
Iron is required for cellular metabolism and growth , but can be toxic due to its ability to cause high oxidative stress and consequently DNA damage . To prevent damage , all organisms that require iron have developed mechanisms to tightly control iron levels . Dysregulation of iron metabolism is detrimental and can contribute to a wide range of diseases , including cancer . This paper presents a predictive mathematical model of iron regulation linked to iron utilization , oxidative stress , and the oncogenic response specific to normal breast epithelial cells . The model uses a discrete modeling framework to generate novel biological hypotheses for an investigation of how normal breast cells become malignant cells , capturing a breast cancer phenotype of iron homeostasis through overexpression and knockout simulations . The new biology discovered is ( 1 ) IRP2 overexpression alters the iron homeostasis pathway in breast cells , without affecting the oxidative stress response or oncogenic pathways , ( 2 ) an activated oncogenic pathway disrupts iron regulation in breast cancer cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"heme",
"medicine",
"and",
"health",
"sciences",
"breast",
"tumors",
"oxidative",
"stress",
"cancers",
"and",
"neoplasms",
"oncology",
"physiological",
"processes",
"mathematics",
"homeostasis",
"algebra",
"mitochondria",
"bioenergetics",
"cellular",
"structures",
"and",
"organelles",
"polynomials",
"proteins",
"breast",
"cancer",
"chemistry",
"biochemistry",
"carcinogenesis",
"iron",
"cell",
"biology",
"post-translational",
"modification",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"energy-producing",
"organelles",
"chemical",
"elements"
] |
2017
|
Activated Oncogenic Pathway Modifies Iron Network in Breast Epithelial Cells: A Dynamic Modeling Perspective
|
Schistosomiasis is a chronic neglected tropical disease that is characterized by continued inflammatory challenges to the exposed population and it has been established as a possible risk factor in the aetiology of bladder cancer . Improved diagnosis of schistosomiasis and its associated pathology is possible through mass spectrometry to identify biomarkers among the infected population , which will influence early detection of the disease and its subtle morbidity . A high-throughput proteomic approach was used to analyse human urine samples for 49 volunteers from Eggua , a schistosomiasis endemic community in South-West , Nigeria . The individuals were previously screened for Schistosoma haematobium and structural bladder pathologies via microscopy and ultrasonography respectively . Samples were categorised into schistosomiasis , schistosomiasis with bladder pathology , bladder pathology , and a normal healthy control group . These samples were analysed to identify potential protein biomarkers . A total of 1306 proteins and 9701 unique peptides were observed in this study ( FDR = 0 . 01 ) . Fifty-four human proteins were found to be potential biomarkers for schistosomiasis and bladder pathologies due to schistosomiasis by label-free quantitative comparison between groups . Thirty-six ( 36 ) parasite-derived potential biomarkers were also identified , which include some existing putative schistosomiasis biomarkers that have been previously reported . Some of these proteins include Elongation factor 1 alpha , phosphopyruvate hydratase , histone H4 and heat shock proteins ( HSP 60 , HSP 70 ) . These findings provide an in-depth analysis of potential schistosoma and human host protein biomarkers for diagnosis of chronic schistosomiasis caused by Schistosoma haematobium and its pathogenesis .
Urinary schistosomiasis , caused by the parasite Schistosoma haematobium , is of public health significance in tropical and sub-tropical areas , with an estimated 732 million persons being vulnerable to infection worldwide in well-defined transmission areas [1] . In 2008 , 17 . 5 million people were treated globally for schistosomiasis , 11 . 7 million of those from sub-Saharan Africa [2] . Schistosomiasis is considered to be endemic in Nigeria [1 , 3 , 4] , where about 20 million people are affected by chronic schistosomiasis [1 , 5] . Schistosoma haematobium infection is reported to be more widespread than Schistosoma mansoni infection [6] . Schistosomiasis is characterized by continued health threat and inflammatory challenges in people who are exposed to long-term daily risk of infection [7] . Chronic infection with S . haematobium has been reported as a possible risk factor in the aetiology of bladder cancer [8 , 9] . Several studies have recorded increased urinary tract pathology conditions among populations infected with S . haematobium [3 , 10] . Histopathologists have also associated S . haematobium infection with the development of squamous cell carcinoma of the bladder [11] . S . haematobium has been associated with a two- to ten-fold increase in the risk of bladder squamous cell carcinoma , as well as being a potential cause of kidney damage . Hence , the parasite is considered as a group 1 carcinogen [12] . In some regions where S . haematobium is endemic , bladder cancer is the most common cancer in men and the second most common in women , just behind breast cancer , accounting for as much as 30% of all cancer cases [13] . Early disease detection of bladder cancer would significantly benefit people living in S . haematobium-endemic areas , because bladder cancer is otherwise unlikely to be recognized , as the obvious urinary tract symptoms ( intermittent haematuria , dysuria , increased frequency , urgency and pain with micturition ) are so commonly associated with urinary schistosomiasis that when the cancer manifests the patient is not likely to receive adequate diagnosis and may become severely debilitated with poor disease prognosis [10] . Detecting bladder cancer at the population level is challenging because direct proof requires detailed histopathological study , but invasive examinations are restricted to tertiary hospitals [14] . The detection of cancer-associated biomarkers , preferably isolated from urine and blood , has therefore become important . Such biomarkers are now being developed and will provide tools that could be useful to evaluate the specific effects of long-term exposure to S . haematobium [15] . Demonstration of schistosome-associated bladder damage by ultrasound examination is valuable and useful; however , it cannot be used to construe a diagnosis of cancer . Cancer-specific urine biomarkers may therefore play an important role in people with long-term S . haematobium infections . In addition , considering the fact that treatment of schistosomiasis relies on a single drug , praziquantel , which raises fears of resistance , there is a need to acquire a deeper understanding of the communication between the parasite and the mammalian host , with a view to identifying new methods of controlling schistosomiasis and schistosomiasis-associated bladder cancer . One potential approach to investigating the developing relationship between the parasite and its host is proteomics . Biological fluids are promising sources of diagnostic , prognostic and treatment based biomarkers , due to their easy accessibility [16 , 17] . Biological fluids are associated with tissues that release protein components into them and the disease-altered state could change either the constituents or the amount of such proteins . Biofluid-derived proteins could be parasite or host associated biomarkers . Proteomics has been successfully employed for human-based studies of disease , where it has been a valuable approach for distinguishing diseases and generating candidate biomarkers to determine pathological state [16 , 18] . Mass spectrometry ( MS ) based analysis of a small number of exposed and unexposed subjects has been found to reveal altered expression of proteins that may be identified as intermediate biomarkers of early disease effects [19] . In particular , the potential of the urinary proteome as a non-invasive means to identify biomarkers for carcinogen exposure and metabolism of toxic chemicals has been demonstrated by Moore et al . [19] . Several schistosome-oriented proteomics studies have focused on the parasites [12 , 16] . However , more information on the changes that manifest in the host proteome during active schistosomiasis is required [14] . The goal of the present study is thus to identify candidate biomarkers for the diagnosis of schistosomiasis and schistosomiasis-associated bladder cancer from adults in a rural population in south-west Nigeria , an area which is endemic for urinary schistosomiasis .
Human urine samples were collected from volunteers living in Eggua , Ogun State , a schistosomiasis endemic community in South-west , Nigeria . Eggua lies between latitude 7° 6ʹ4 . 811ʺ N and longitude 2° 52ʹ 43 . 776ʺ E in a derived savanna zone . The area is largely dominated by Yoruba speaking people . The volunteers were screened for the presence of Schistosoma haematobium infection by a combination of microscopy ( Fig 1 ) , detection of macro and microhaematuria ( urinalysis ) and rapid diagnostic test ( RDT ) for schistosomiasis ( Table 1 ) . The urine samples were collected between 10:00 and 14:00 to ensure maximum egg yield and 10 mL of sample were processed for microscopic examination and egg count . The eggs were quantified by counting and classified as light infection if there were ≤50 ( 1–49 ) eggs/10 mL urine and heavy infection if there were >50 eggs/10 mL urine [10] and urinary structural bladder pathology was examined using ultrasonography , which has been published elsewhere [10] . Sample size power calculation was carried out , indicating that 44 individual samples ( N ) were required for a statistical power of 0 . 9 at significance level 0 . 05 . Ethical approval was obtained from the University of Ibadan and University College Hospital ( UI/UCH ) Ethical Committee and Ogun State Ministry of Health . All participants gave informed consent; all participants were adults and were able to decide for themselves . The informed consent document was written in both English and Yoruba languages , the latter being the language of the Nigerian communities . For those participants who could read and write , written informed consent was obtained . For those participants who could not read and write , the informed consent form was read to them in their language . All participants enrolled in the study voluntarily . The informed consent was signed by all participants and those who could not sign provided a thumb print on the informed consent form . This approach to informed consent for those who could not read and write was approved by the UI/UCH Ethical Committee . A total of 49 individual urine samples were placed into four different categories , namely 12 schistosomiasis cases ( SH ) , 12 bladder pathology cases ( PT ) , 15 combined pathology and schistosomiasis ( PS ) cases and 10 controls with no pathology or schistosomiasis ( NPS ) . All samples were processed on the same day using the same batches of all reagents , including trypsin , to minimise batch effects and sample to sample technical variation . An aliquot of 4 ml of urine per individual was subjected to methanol-chloroform protein precipitation followed by in solution tryptic digest prior to MS analysis . Precipitated protein was resuspended in denaturation buffer ( 6 M urea , 2 M thiourea , 10 mM Tris buffer , pH 8 . 0 ) , and then a Bradford assay was carried out to determine protein concentration [17] . For each sample , 100 μg of protein was then further reduced by incubation at room temperature for 1 hour in reduction buffer ( 1 M dithiothreitol ( DTT ) ; 50 mM ammonium bicarbonate ( ABC ) . An alkylating buffer ( 550 mM iodoacetamide ( IAA ) in 50 mM ABC ) was then added and incubated in the dark at room temperature for an hour . The sample was then diluted with 4 volumes of 50 mM ABC and proteolysed overnight for 16 hours at 37°C using Trypsin-Ultra ( mass spectrometry grade; New England BioLabs ) according to the manufacturer’s instructions . An equivalent of 10 μg of the peptide solution was then transferred to in-house prepared stage tips for off-line solid-phase extraction , desalting and clean-up of sample , as described in previous studies [17 , 20] , and the desalted peptides were then dried in a refrigerated speedy vac ( SPD 111v-230 Speed VAC , Thermo Savant , New York , USA ) . Peptide samples were resuspended by diluting the desalted , dried peptides to 200 ng/μL using 2% acetonitrile ( ACN ) in HPLC grade water containing 0 . 1% v/v formic acid ( FA ) before MS analysis . Nanoflow ultra-HPLC was carried out on each sample , without pre-fractionation , using a Dionex UltiMate 3500RSnano UPLC system ( Thermo Fisher , San Jose , CA , USA ) equipped with a reverse phase ( RP ) pre-column trap ( 100 μm × 2 cm; 5 μm; 100 Å; C18 ) and analytical column ( 70 μm × 20 cm; 5 μm; 100 Å; C18 ) . Equal 400 ng injections of each sample were eluted by gradient chromatography at 23°C with a flow rate of 300 nL/min , and a 6–40% gradient of water–ACN from 0 to 120 min . The binary mobile phase system used was as follows: buffer A contained water and 0 . 1% FA , while buffer B contained ACN and 0 . 1% FA . The elution gradient for peptides was 6% B from 10 min to 40% B at 60 min , then increasing to 80% B for 10 min before returning to 2% B for equilibration . The same pre-column trap and analytical column was used for all samples . Discovery proteomic analysis of each sample was carried out on a Q Exactive hybrid quadrupole-Orbitrap mass spectrometer ( Thermo Fisher ) . Analysis of samples introduced from the in-line HPLC system was achieved with the following system settings: Data-dependent automated full scan cycles were performed with automatic switching between MS/MS and MS scans at a scan range of 300–1650 m/z . The top ten most abundant precursor ions selected by the quadrupole during the initial MS scan were subjected to fragmentation using high-energy collision dissociation with normalized collision energy at a pressure of 1 . 2 mTorr and a dynamic exclusion time of 30 s . The abundance threshold for ion selection was 0 . 001 with charge exclusion of z = 1 ions . Acquisition of mass spectra was done at a resolution of 70 , 000 with a maximum injection time of 250 ms or a target automatic gain control value of 3×106 . High-energy collision dissociation and normalized collision energy set at 27 were used for peptide fragmentation . Continuous tandem mass spectra acquisition resolution was set at 17 , 500 at a maximum injection time of 120 ms or target AGC of 2 × 105 . The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD006438 . All raw MS data was processed with MaxQuant software ( version 1 . 5 . 3 . 12 ) and its built-in search engine , Andromeda , for peptide identification and protein inference , using the default settings and the human and S . mansoni databases ( www . uniprot . org ) , as described in detail elsewhere [17] ( S1 Fig ) . FDRs were set at 1% at both peptide and protein level . Peptide identifications were transferred to unidentified features in other LC-MS runs based on matching masses and re-calibrated retention times between runs ( “match between runs” option in MaxQuant ) . Normalisation of data between LC-MS runs was carried out in MaxQuant , based on summation of the total peptide ion signals per sample and using a global Levenberg-Marquardt optimisation procedure that aimed to minimise the overall changes for all peptides across all samples [21] . Label-free quantification ( LFQ ) was then carried out using MaxQuant , based on determination of the median pair-wise common peptide ratios between samples , requiring a minimum of two peptide ratios per identified protein [21] . The LFQ values from MaxQuant were imported into Perseus software ( version 1 . 5 . 3 . 1 ) for differential expression statistical analysis and visualization by hierarchical clustering and Principal Component Analysis ( PCA ) , with a Benjamini-Hochberg multiple testing correction cut off set at FDR 0 . 05 . Three separate independent t-tests were carried out using LFQ data to compare NPS versus SH , NPS versus PT , and NPS versus PS . One-way ANOVA was also carried out to statistically validate the differentially expressed potential biomarkers . Venn diagrams were plotted using VennDIS ( version 1 . 01 ) . Proteins which were determined to be significantly differentially expressed between groups were further subjected to a GO-enrichment analysis using Blast2GO [16 , 22] . Uniquely identified human proteins were subjected to pathway and protein-protein interaction analysis using the Reactome database ( v60; www . reactome . org ) and String DB ( www . string-db . org ) , respectively .
Urine samples from 49 individuals distributed across the four disease groups ( NPS , SH , PT and PS ) were analysed to identify candidate biomarkers for schistosomiasis and its associated pathologies ( Table 1 ) . High levels of correlation between the urinary protein components of these sample groups was demonstrated by scatterplots , hierarchical clustering ( heatmap ) and principal component analysis ( PCA ) ( Figs 2 and 3 ) . The dimensions in Fig 3 account for 76 . 6% and 20% of the observed variation , respectively . Hierarchical clustering of protein groups identified in SH , PT and PS and NPS samples showed clear molecular differences between groups . As expected , differences in the proteomic signatures were observed between the control group ( NPS ) and all disease groups , as all three disease groups clustered distinctly , with the SH and PS group being more closely related to one another as compared to the PT group ( Figs 2 and 3 ) , although a few samples of the pathology group clustered proximally to the control group ( Fig 3 ) . A total of 209 , 923 tandem mass spectra were identified and used to assign peptides and unique protein group identities , leading to the identification of 9 , 701 non-redundant peptides and 1 , 306 protein groups at a false discovery rate of 1% . The majority ( 66 . 3% ) of peptides identified had no missed cleavages , while 27 . 5% had one missed cleavage and 6 . 2% had two missed cleavages ( S4 Fig ) , which is within the expected range for in solution digest of complex protein mixtures . The number of peptide sequences identified and % MS/MS spectra identified per sample are shown in S2 and S3 Figs , respectively . Prior to quantitative analysis of differential abundance , samples from each individual were normalised at the peptide level before mass spectrometric analysis . The intensity data for all identified peptides per sample was then further normalised in MaxQuant , based on the underlying assumption that the majority of the proteome should not change significantly between any two conditions and that the average behaviour can therefore be used as a relative standard [21] . In other words , normalisation between samples was based solely on the distribution of the peptide-level data obtained , without addition of external standards or reliance on any set of housekeeping proteins that are assumed to be stably expressed , as is now commonplace in label-free proteomics . Subsequent label-free differential protein abundance analysis was carried out at the protein level using MaxQuant and volcano plots of pairwise comparisons demonstrated the effectiveness of the normalisation procedures , with the majority of proteins found to be not significantly differentially expressed between the various comparisons after permutation-based FDR truncation ( FDR = 0 . 05; S5 Fig ) . A total of 36 Schistosoma proteins were identified in the host urine when the MS output was searched against a combination of human and Schistosoma databases ( Table 2 ) . These 36 identities were considered to be confident identities due to the relatively small size of the Schistosoma database compared to the human database in the combined database . The GO terms for the molecular function of the identified human and Schistosoma-derived proteins are summarized in Figs 4 and 5 . More ( 124 ) parasite protein groups were identified when the MS output was searched against the Schistosoma DB only , but only 31 Schistosoma proteins were differentially abundant between groups by ANOVA using label-fee quantification ( LFQ ) . Some Schistosoma specific proteins were found in samples from individuals earlier diagnosed and classified as negative for S . haematobium infection by microscopy ( Figs 6 and 7 ) . Venn diagrams were generated to identify proteins unique to each group . Out of the 36 total Schistosoma protein groups confidently identified , 5 ( 15 . 6% ) , 4 ( 12 . 6% ) and 2 ( 6 . 3% ) proteins were unique to SH , PS and PT group respectively while only 8 proteins ( 25% ) were found common to all study groups ( Fig 7A ) . Heat shock protein 70 , elongation factor 1-alpha , camp-response element binding proteins-related , histone H4 and venom allergen-like ( VAL ) 3 proteins were found to be unique to SH group while tubulin alpha chain , calreticulin autoantigen homolog , heat shock protein HSP 60 and putative ADP , ATP carrier protein were found only in PS group . 2 potential biomarkers unique to the PT group include cytoplasmic dynein light chain and putative actin 1 . 13 ( 36 . 1% ) of the predicted Schistosoma protein were membrane-associated , 8 ( 22 . 2% ) nuclear based , 4 ( 11% ) cytoplasmic and 3 ( 8 . 3% ) cytoskeletal and mitochondrial , 1 ( 2 . 8% ) ribosomal and 3 ( 8 . 3% ) unknown . The result of the three independent t-tests performed using LFQ values for “NPS versus SH” , “NPS versus PT” and “NPS versus PS” revealed a total of 54 candidate human protein biomarkers for schistosomiasis and bladder pathology . The proteins are distributed into 43 , 8 and 7 for “NPS versus SH” ( Table 3 ) , “NPS versus PT” and “NPS versus PS” ( Table 4 ) respectively . 37 and 2 proteins were unique to SH and PT groups , respectively , while none were unique to the PS group ( Fig 7B ) . A search for possible marker overlap across study groups showed that cathepsin B ( P07858 ) was shared by all disease groups; arylsulfatase A ( A0A0C4DFZ2 ) and phosphatidylethanolamine-binding protein 4 ( Q96S96 ) were shared by PS and SH group ( Table 5 ) ; and PT and SH groups were found to have 4 proteins in common , namely transthyretin ( P02766 ) , plasma retinol-binding protein ( Q5VY30 ) , phosphatidylcholine-sterol acyltransferase ( P04180 ) and cartilage intermediate layer protein 2 ( K7EPJ4 ) . The majority of the human proteins identified were predicted to be membrane associated and perform “binding” molecular activities . The human proteins that were identified as differentially abundant in the SH , PT and PS groups were subjected to pathway analysis using the reactome tool and string DB . Reactome did not identify any statistically significantly overexpressed human pathways in any of the groups , as the majority of identified proteins did not have mapped identifiers . Using string DB , we found that the cellular component ‘extracellular exosomes’ ( GO:0070062 ) was significantly functionally enriched in the SH group compared to that of the NPS group , with a FDR of 1 . 09e-14 . Furthermore , the Kegg pathway ‘lysosome' ( ID 04142 ) was functionally enriched with an FDR of 0 . 00685 . The number of differentially abundant proteins unique to the PT and PS groups was too few to identify functionally enriched pathways or GO terms .
A total of over 2 , 000 proteins are estimated to be present in normal human urine [23] , and the highest number of human proteins identified in a proteomic study thus far is 1 , 823 [24] . In the present study , we observed 1 , 306 proteins in human urine by unfractionated MS analysis . Sample clustering analyses by PCA and heatmap placed all sample groups into clear-cut strata based on the LFQ intensities of the identified proteins , with little interference between groups , thereby indicating a distinct difference in proteomic signatures between groups . Some of the potential Schistosoma biomarkers identified in this study are clear targets for the generation of new vaccines and drug targets against schistosomiasis . The majority of these proteins appear to be involved in binding activity according to GO-enrichment analysis . This observation is similar to the report of Sotillo et al . [16] . The parasite markers include four heat shock proteins ( HSPs ) which are known as highly conserved stress-induced proteins found in many trematodes and nematodes including Schistosoma specific study [25] . HSP expression in the earliest stages of intra-mammalian schistosomula development has been reported and was suggested to be as a result of thermal changes in the parasite niche/environment i . e . changes between freshwater and the human body [16 , 26] . Venom allergen-like protein ( VAL ) -3 was identified in the present study . Sotillo et al . [16] reported downregulation of VAL-4 and -6 on the maturing schistosomula tergument and the upregulated expression of VAL-6 in cercaria and adults worm has also been reported [27] . The Schistosomas VALs comprise at least 29 members , subdivided into two major groupings , with group 1 including SmVAL1–5 , 7–10 , 12 , 14–15 , & 18–29 which have signal peptides and conserved cysteines positioned for disulphide bond formation . VAL proteins are associated with excretion/secretion products and extracellular environment of the parasite [28] and have been used as a trial vaccine against hookworm infections in humans [29] . Changes in the val gene and the resultant protein expression denotes its functions in different aspects of host-parasite biology , which include snail invasion by miracidium , intra-molluscan sporocyst development , and cercarial development and host penetration [28] . The VAL protein family are abundant in different helminth species including gastrointestinal nematodes , where they are known to carry out several roles in the infective activities of parasites [16 , 30] . Actin 1 protein , as reported in this study , has been identified as a possible drug target for the treatment of schistosomiasis . Strong association between actin and S . mansoni adult worm surface membranes has been confirmed [25 , 31] . Studies have described the role of actin in enhancing the activity of praziquantel ( PQZ ) treatment of schistosomiasis . It is suggested that PQZ intercalates in the surface membrane lipid bilayers , thereby inducing tegumental changes that leads to antigen exposure , including actin [31 , 32] . Elongation factor 1-alpha , phosphopyruvate hydrase and histone-4 were all identified as potential Schistosoma biomarkers in this report , which parallels the results of deWalick et al . [25] , where these proteins were identified in purified eggshell fragments of Schistosoma mansoni . The proteins identified as part of the eggshell protein skeleton are known schistosome antigens and may induce cellular or antibody responses [25] . These eggshell markers may be very useful schistosomiasis diagnostic candidates rather than vaccine candidates , since such a vaccine would be likely to target the eggs and further encourage granuloma formation and pathology rather than priming the immune system against the parasite . The significantly regulated parasite proteins were mostly predicted to be membrane-associated , when classified according to their predicted subcellular location [16 , 25] . The expression of some membrane associated proteins was earlier proposed as possible vaccine antigens in different Schistosoma spp [16 , 33] . Jossic et al . [34] reported membrane proteins as one of the most interesting classes of proteins among disease biomarker candidates due to their localization on the surface cells and organelles . The identification of a large number of membrane and membrane-associated proteins in the present study strongly suggests that these proteins are abundant in the urine of schistosomiasis patients and would therefore be reasonable targets for drug or vaccine development . These proteins are likely present in the urine as a result of their accessibility to the host immune system via the parasitic tegument , which constitutes the host-parasite interface between S haematobium and the human immune system . Despite some recent clarification regarding common transmembrane tegument proteins in S . mansoni [35] , it remains unclear what proteins are abundantly present at this important interface in S . haematobium . An ideal vaccine candidate would present a large extracellular domain , such as the tetraspanin family of proteins , which are a promising candidate for S . mansoni vaccines [36] . The present study therefore offers some insight in to potential membrane protein targets for vaccine development in S . haematobium . Arylsulfatase A and phosphatidylethanolamine-binding protein 4 were both found in the SH and PS sampled group . Arylsulfatases A , B , and C ( arylsulfo-hydrolases ) are a group of hydrolytic enzymes that occur in various tissues and fluids [37] . An increase in the activities of arylsulfatase B ( ASB ) has been reported in bladder tumours [38] . Also , arylsulfatase A ( ASA ) in the livers of Schistosoma infected mice displayed a non- significant decrease in expression vs the control , while the expression of hepatic ASB was significantly increased in Schistosma infected mice in similar study [37] . Aminophospholipids , such as phosphatidylserine and phosphatidylethanolamine are described as specific , accessible and stable markers of the luminal surface of tumour blood vessels [39] . There has already been some development of aminophospholipid-targeted diagnostic and therapeutic constructs for use in tumour intervention . Antibody-therapeutic agent conjugates and constructs that bind to aminophospholipids , including methods that specifically deliver therapeutic agents , such as toxins and coagulants , to the constitutively-expressed aminophospholipids of tumour blood vessels , thereby inducing thrombosis , necrosis and tumour regression , are particularly promising [39] . One of the four proteins shared by PT and SH samples , plasma retinol-binding protein ( RBP ) , is a circulating plasma protein produced in the liver and adipose tissue that transports active natural metabolites of Vitamin A as retinol around the body [40] . Retinol acts pharmacologically to restore differentiation and inhibit growth in some premalignant and malignant cell both in vivo and in vitro ( including bladder cancer cases ) and also modulates cell proliferation , malignant transformation , apoptosis and the immune system [40 , 41] . A recent study revealed that individuals with HIV and S . mansoni coinfection have significantly lower blood RBP levels when compared to participants with HIV and S . haematobium coinfection [40] . Transthyretin has also been identified by Yi-Ting et al . , [42] as a potential urine-derived biomarker for bladder cancer . The programmed cell death 1 ( PD-1 ) surface receptor binds to two ligands , PD-L1 and PD-L2 . PD-1–PD-L interaction is known to control the induction and maintenance of peripheral T cell tolerance . PD-1 and its ligands have been exploited by a variety of microorganisms to reduce the effect of antimicrobial immunity , thereby facilitating chronic infection [43] . The findings of Alvarez et al . [44] on the role played by PD-1 in innate immunity against Mycobacterium tuberculosis also showed that PD-1 signalling might be modulating host innate immunity by inhibiting natural killer ( NK ) cell responses to the pathogen , contributing to avoidance of immune-mediated pathology caused by excessive host response to the infection . Understanding the functions of PD-1 and its ligands in regulating antimicrobial and self-reactive T cell responses and the possibility of manipulating this pathway may eventually reveal its therapeutic potential in chronic schistosomiasis . Human Growth/differentiation factor 15 ( GDF15 ) could be a useful diagnostic marker for chronic urinary schistosomiasis . GDF15 is a divergent member of the transforming growth factor β family found in a broad range of cells [45] . Corre et al . [45] reported that GDF15 could be an integrative signal in pathological conditions and provide may information on the severity of disease . Expression and secretion of GDF15 is heightened in many malignant tissue and cancer cell lines ( prostate , colorectal , pancreatic , gastric and oral squamous carcinoma ) as compared with their normal tissues or cells [46 , 47 , 48 , 49 , 50 , 51 , 52] . Human sialidase protein was identified in the SH group and is known for its immunological role in regulating phagocytosis in macrophages cells [53] . Amith et al . [54] reported Neu1 sialidase as a complex with Toll like receptor ( TLR ) -2 , -3 and -4 , which is induced upon ligand binding to either receptor . Activated Neu1 sialidase hydrolyzes sialyl α-2 , 3-linked β-galactosyl residues distant from ligand binding to remove steric hindrance to TLR-4 dimerization , MyD88/TLR4 complex recruitment , NFkB activation and pro-inflammatory cell responses [54] . We identified differentially abundant human proteins in each clinical group that may contribute further specificity to a panel of biomarkers . We found a total of 54 proteins that were differentially abundant compared to the control group , with 43 that were specific for schistosomiasis , 8 that were specific to bladder pathology , and 7 that were specific to those patients with bladder pathology and schistosomiasis . Of these , only 37 and 2 proteins were uniquely identified in the SH and PT groups , respectively , while none were unique to the PS group . This implies that the combined pathology of schistosomiasis and bladder pathology may not have uniquely identifiable characteristics , but rather has features common to both contributing diseases . On the other hand , the proteins that are unique to Schistosoma infected individuals or those with bladder pathology are of interest in identifying the molecular mechanisms underlying the pathology of these diseases and may serve as differential biomarkers for diagnostic purposes . Due to the relatively small numbers of unique differentially abundant proteins identified in each clinical group , pathway analysis is not especially informative . The Reactome database identified no statistically significantly dysregulated pathways , although String analysis identified that ‘extracellular exosomes’ were enriched as a cellular component , as well as the Kegg pathway relating to ‘lysosomes’ . This suggests that exosomes may have a role in host/pathogen protein trafficking in the urine , which has been of recent interest in other diseases [55] . This study is currently at the discovery phase of identifying schistosome and human based biomarkers for urinary schistosomiasis and its associated pathologies . Actin 1 , elongation factor 1-alpha , phosphopyruvate hydrase , heat shock protein , histone-4 and other schistosome-derived proteins identified in this study could be considered as markers for the diagnosis of urinary schistosomiasis . The presence of venom allergen-like protein-3 in the present study confirms its potential as a promising vaccine biomarker against the parasite . The human programmed cell death 1 and Growth/differentiation factor 15 proteins could also be promising markers for the diagnosis of chronic urinary schistosomiasis , a condition that is difficult to identify by microscopy . The consistency of detection of the human proteins arylsufatase and cathepsin across all disease groups ( SH , PT and PS ) suggests that these markers may be useful in identifying links between schistosomiasis and the development of urinary bladder cancer . We propose that a panel of biomarkers derived from both human and Schistosoma may achieve the best clinical sensitivity and specificity , and this study goes some way to identifying putative candidates for a further quantitative clinical study on a large number of blinded samples . The reduced complexity of the protein content of urine and its non-invasive sample collection renders urine a valuable source for diagnostic biomarkers , especially for urinary tract diseases . In addition , urine is an ideal biofluid for biomarker discovery by mass spectrometry-based proteomics due to the abundant availability of urine samples and the relative stability of urine proteins . With the use of integrated high throughput technologies , we can begin to elucidate how S . haematobium and human host systems interact during infection . The momentous challenge we face is the possibility of parasite resistance to the only known drug , Praziquantel , and the ongoing problem of continual re-infection within at-risk populations . The comparative proteomics approach undertaken in this study has generated promising hypotheses regarding the mechanisms of pathogenesis that can be tested through manipulation of the host and parasite during infection . This study demonstrates that urinary proteomics is a viable approach to discovering candidate biomarkers for schistosomiasis and its associated pathology , but the results presented here require validation in a larger cohort before clinical applications can be considered .
|
Schistosomiasis , caused by S . haematobium , causes inflammation in the bladder and is common in tropical areas such as Nigeria . Undetected schistosomiasis can lead to inflammation in the bladder which may lead to bladder cancer . Diagnosis of bladder cancer in areas with common urinary schistosomiasis is difficult because the two diseases share many common symptoms . It is therefore important to identify biomarkers that could be used for early diagnosis of both schistosomiasis and schistosomiasis-associated bladder cancer . We chose a proteomic approach to identify these candidate biomarkers in a clinical cohort of urine samples obtained in Nigeria . We found several parasite- and host-specific protein biomarkers that could have diagnostic potential in a urine-based test for schistosomiasis , and schistosomiasis-associated bladder cancer . These include proteins that may also have application in drug or vaccine development . This study is one of the first to catalogue the urinary proteomes of people affected by this neglected tropical disease .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"schistosoma",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"helminths",
"tropical",
"diseases",
"dna-binding",
"proteins",
"biomarkers",
"parasitic",
"diseases",
"animals",
"urine",
"membrane",
"receptor",
"signaling",
"neglected",
"tropical",
"diseases",
"cell",
"nucleus",
"cellular",
"structures",
"and",
"organelles",
"nuclear",
"membrane",
"proteins",
"schistosoma",
"haematobium",
"membrane-associated",
"cytoskeletal",
"proteins",
"biochemistry",
"helminth",
"infections",
"schistosomiasis",
"signal",
"transduction",
"cytoskeletal",
"proteins",
"eukaryota",
"cell",
"biology",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"organisms"
] |
2017
|
Quantitative label-free proteomic analysis of human urine to identify novel candidate protein biomarkers for schistosomiasis
|
Adaptation in eukaryotes is generally assumed to be mutation-limited because of small effective population sizes . This view is difficult to reconcile , however , with the observation that adaptation to anthropogenic changes , such as the introduction of pesticides , can occur very rapidly . Here we investigate adaptation at a key insecticide resistance locus ( Ace ) in Drosophila melanogaster and show that multiple simple and complex resistance alleles evolved quickly and repeatedly within individual populations . Our results imply that the current effective population size of modern D . melanogaster populations is likely to be substantially larger ( ≥100-fold ) than commonly believed . This discrepancy arises because estimates of the effective population size are generally derived from levels of standing variation and thus reveal long-term population dynamics dominated by sharp—even if infrequent—bottlenecks . The short-term effective population sizes relevant for strong adaptation , on the other hand , might be much closer to census population sizes . Adaptation in Drosophila may therefore not be limited by waiting for mutations at single sites , and complex adaptive alleles can be generated quickly without fixation of intermediate states . Adaptive events should also commonly involve the simultaneous rise in frequency of independently generated adaptive mutations . These so-called soft sweeps have very distinct effects on the linked neutral polymorphisms compared to the standard hard sweeps in mutation-limited scenarios . Methods for the mapping of adaptive mutations or association mapping of evolutionarily relevant mutations may thus need to be reconsidered .
The speed of adaptation in eukaryotes is commonly assumed to be limited by the waiting-time for an appropriate adaptive mutation . This notion is based on estimates of the population parameter Θ = 4Neμ ( the product of effective population size Ne and per-site mutation rate μ ) derived from levels of standing neutral variation . Θ can be interpreted as the rate at which new mutations arise in the population [1] . In contrast to many prokaryotes or viruses , where Θ can easily be on the order of one or larger - and consequently most single nucleotide mutations exist in the population at every given time – estimated values of Θ in eukaryotes are typically much smaller than one [1] . Adaptation should thus be substantially retarded , especially when adaptive alleles need to carry several independent mutations . However , adaptation to anthropogenic changes such as the evolution of insecticide resistance has been observed to occur very rapidly and often involves complex alleles [2]–[7] . One possible explanation for such cases of rapid adaptation is that complex resistant alleles predate environmental changes [8] , [9] . The other possibility is that adaptive mutations emerge more quickly in eukaryotic populations than commonly believed . The latter would imply that estimates of Θ have to be reconsidered in the context of rapid adaptation . In order to understand the population parameters that allow for rapid adaptation in eukaryotes , we study here a well-documented example: the evolution of pesticide resistance in D . melanogaster . Acetylcholinesterase ( AChE ) , a key neuronal signalling enzyme , is the major target of the most commonly used insecticides , organophosphates ( OPs ) and carbamates ( CMs ) [10] . Introduced in the 1950–1960's , these insecticides have been used pervasively around the world since then . Within a few years of their introduction cases of insecticide-resistant AChE alleles emerged [11] and today insecticide-resistant AChE has been observed and characterized in numerous arthropod species [2]–[7] . In D . melanogaster , four particular point mutations at highly conserved sites ( I161V , G265A , F330Y , G368A ) of Ace ( the gene coding for AChE ) lead to resistance to OPs and CMs [5] , [12] ( Figure S1 ) . Alleles carrying these mutations singly and in combination have been found in natural populations worldwide [12] . In the presence of OPs , these mutations confer semi-additive resistance: single mutations provide moderate levels of resistance to ∼75% of OPs , any two mutations in combination provide higher levels of resistance to ∼80% of OPs , while alleles with three or four mutations lead to strong resistance to practically all OPs [12] . One 3-mutation allele ( I161V , G265A , F330Y ) was found worldwide at particularly high frequencies and is a key determinant of resistance to OPs [12] . In the absence of pesticides all resistant alleles are strongly deleterious with the selective coefficient on the order of negative 5–20% [13] , [14] . Here we collect data and provide quantitative arguments ( both analytical and simulation-based ) that the observed signatures of adaptation at Ace imply a much larger ( ∼100-fold or more ) effective population size than is commonly assumed for D . melanogaster . We discuss the implications of our results for the study of adaptation in Drosophila and other species with large census sizes .
D . melanogaster evolved in sub-Saharan Africa ( AF ) and spread worldwide over the past 10–16 thousand years [15] . The worldwide spread was associated with a severe bottleneck that resulted in sub-sampling of AF diversity by the out-of-Africa strains [15] . Resistant alleles found outside of AF may either have arisen in situ in the derived out-of-Africa populations or were present in the AF population prior to the bottleneck ( similar to [8] , [9] ) . These two hypotheses can be distinguished by studying haplotype backgrounds of the resistant alleles . Resistant mutations that evolved in derived populations in situ , unlike ancient AF resistant alleles , should reside on the background of sensitive haplotypes common in the exposed out-of-Africa populations that passed through the bottleneck . We collected D . melanogaster sequence data ( ∼1 . 5 kb covering the known four sites of resistant mutations in Ace ) from 93 resistant and sensitive strains . We sequenced 9 alleles from the ancestral AF populations , 10 alleles from the derived Eurasian and American populations collected prior to the 1950s ( M strains ) [16] , and 74 alleles from the recently collected ( 1990–2009 ) derived populations in North America ( NA ) and Australia ( AUS ) ( Table S1 and Table S2 ) . We detected resistant mutations at the first three sites ( I161V , G265A , F330Y ) but did not find the resistant mutation at the fourth site ( G368A ) . We estimated that ∼40% of the strains contain resistant mutations in the modern NA and AUS populations of D . melanogaster . Figure 1 shows the most parsimonious haplotype network of the sequenced alleles . Figure 2 shows the segregating sites for sensitive haplotypes , as well as the I161V 1-mutation and the 3-mutation haplotypes ( Table S2 shows segregating sites for all sequenced alleles ) . In all cases the NA and AUS resistant alleles show no signs of having predated the spread of D . melanogaster out-of-Africa . Instead , the resistant alleles appear to have arisen in situ in different populations , as indicated by the observation that locally common resistant alleles are present on the locally common sensitive haplotypes . For instance , AUS alleles with the resistant mutation in the first site ( marked t ) have the haplotype background that is identical to the sensitive haplotype N that is common in AUS but has not been detected by us in NA . In contrast , the NA first site mutation alleles ( marked p through s ) have the haplotype backgrounds that are nearly identical to the sensitive haplotype L that is common in NA . Additionally , the haplotype background of one of the AF alleles ( marked u ) with the resistant mutation I161V is substantially diverged from the NA and AUS resistant strains and is more similar to the sensitive alleles common in AF . This suggests a third independent origin of the mutation I161V in AF . Note that the complex 2- and 3-mutation haplotypes also appear to have arisen in situ in the derived populations as their haplotype backgrounds are most closely related to the common out-of-Africa sensitive haplotype L . In summary , the sequence analysis of the resistant and sensitive alleles reveals two signatures of the adaptive evolution of pesticide resistance at the Ace gene . First , adaptation has been rapid enough such that in the past 50 years ( 1000 to at most 1500 generations [17] ) multiple resistant alleles including a complex allele containing three independent mutations at three different sites evolved and spread to high frequencies worldwide . Second , many resulting resistant alleles are present on distinct haplotypes that differ in the immediate vicinity of the adaptive sites , such as the adaptive change from A to G at the first site ( I161V ) in NA and AUS that is located on the haplotypes p , q , r , s , and t ( Figure 2 ) . Below we consider a simple scenario of a single locus in a panmictic population of effective size Ne . We assume that the resistant alleles were in mutation-selection balance prior to pesticide application with a strongly deleterious selection coefficient of −5% [13] , [14] and that they became advantageous after the application of pesticides . In Box 1 we show that if Θ∼0 . 01 the probability of successful adaptation from standing genetic variation is less than 1% even if positive selection is extremely strong ( s∼100% ) . Thus , if Θ∼0 . 01 , as previously estimated based on analyses of neutral loci in Drosophila [18] , [19] , we only need to consider the case of adaptation from de novo mutations . The probability of successful adaptation from de novo mutations depends on the expected waiting time for an adaptive mutation to emerge and to reach substantial frequencies in the population . This waiting time is the sum of the expected times to complete two distinct phases: ( 1 ) the establishment phase in which an adaptive mutation arises and reaches the frequency at which its escape from stochastic loss is assured and ( 2 ) the sweep phase in which the adaptive allele reaches an intermediate population frequency such that it can be readily observed . In Box 1 we show that the overall waiting time can be estimated as ( 1 ) This equation implies that selection must already be very strong for a single 1-mutation allele to arise and to become prevalent in less than 1500 generations ( s>20% for Θ = 0 . 01 ) . Selection coefficients associated with the 2-mutation and 3-mutation alleles need to be even stronger given that they have to outcompete the 1-mutation and 2-mutation alleles respectively . We have established that under this simple model if Θ is 0 . 01 , the adaptation at Ace likely involved very strong positive selection acting on de novo mutations . Can we then explain the second empirical observation , namely that the same adaptive mutation by state is observed on several haplotypes that differ in the immediate vicinity of the adaptive site ? We can imagine two scenarios that would generate this observation . In the first , the so called hard sweep scenario , a single adaptive mutation arises in frequency in the population and eventually ends up on different haplotypes due to recombination or mutation events that take place in its vicinity during the sweep . In the other , an example of the so-called soft sweep scenario , several independent adaptive mutations take place on different haplotypes and increase in frequency simultaneously . Theoretical investigations under simple scenarios by Pennings and Hermisson [20]–[22] showed that such soft sweeps are extremely uncommon if Θ per site is on the order of 0 . 01 independently of the strength of positive selection . The probability of the hard sweep scenario resulting in the observation of the haplotypic diversity in the vicinity of the adaptive allele is calculated in Box 2 . Specifically , we demonstrate that the probability Pd that at least two haplotypes are observed at the end , where the minor haplotype is present in at least a fraction d of the population , is approximately ( 2 ) Here R is the total rate of mutation or recombination in the locus per individual per generation and s is the strength of positive selection . For our locus of length ∼1500 bp we have R∼6*10−6 when assuming a recent estimate for the single-site mutation rate in D . melanogaster of m∼2 . 5*10−9 [23] and a measured recombination rate of ρ∼0 . 15 cM/Mbp [24] . The probability to observe different haplotypes is therefore still very small ( Pd<1% ) even for a low population frequency of d = 2% and assuming s to be 5% . Note that this calculation is very conservative given that in our data multiple haplotypes are present at much higher frequency than 2% , multiple haplotypes vary at sites extremely close to the adaptive allele ( within 38 bp ) , and positive selection was likely much stronger . In conclusion , under this simple scenario , our empirical observations at Ace are unexpected if Θ is indeed on the order of 0 . 01 . Specifically , considering how strong selection must be , we should not be seeing more than one distinct haplotype containing the same adaptive mutation . Note that if Θ were much higher , for example on the order of one or larger , then all of our observations are expected . Soft sweeps would be commonplace because many more mutations enter the population in every generation and can increase in frequency simultaneously thereby generating multiple haplotypes containing the same adaptive mutations [22] , as observed in the data . The establishment time would become smaller making it easier to observe complex , 3-mutation alleles at Ace in less than 1500 generations . However , selection would still need to be strong because the time it takes for an adaptive allele to reach intermediate frequencies is only weakly ( logarithmically ) dependent on the effective population size and inversely proportional to the selection coefficient . We have shown above that under very simple population scenarios the pattern of adaptive evolution at Ace requires large values of Θ . However , it is unclear whether such large values of Θ are required under more complex and realistic scenarios . Variation in strength of selection , recombination rate , and population structure might affect the probability of evolving complex 3-mutation alleles from the simpler 1- or 2-mutation alleles [25] and the probability of observing multiple haplotypes containing the same adaptive mutations . To investigate quantitatively the potential impact of such effects we conducted extensive simulations of adaptation at Ace under a large number of selective ( s = 2 . 5% to 500% ) and demographic scenarios ( 1 to 100 subpopulations , migration rates M = 0 . 01 to 10 individuals per generation between any two subpopulations ) , and with varying recombination rate ( ρ = 0 to 10 cM/Mbp ) ( Table S3 ) . In Figure 3A and 3B we show the frequency trajectories of adaptive haplotypes for two representative simulation runs in a simple single population scenario together with summary statistics across a large number of runs for the two key Θ regimes ( Θ = 0 . 01 and Θ = 1 ) . We use four statistics: P1m and P3m are the probabilities that a single adaptive mutation ( 1m ) allele or the 3-mutation allele ( 3m ) were ever present in at least 10% of the population during the simulation; Pss is the probability that a single adaptive mutation is present on distinct haplotypes in a sample of reasonable size ( the observation that we will call the soft sweep signature from now on ) ; and Pc is the combined probability of observing both the complex 3-mutation allele and a single-mutation soft sweep signature during the same simulation . Figure 3A and 3B show results consistent with our analytical considerations . When Θ∼0 . 01 and selection is of moderate strength , neither the evolution of complex 3-mutation alleles nor soft sweeps signatures are likely . Only when Θ approaches one do both observations become commonplace . Figure 3C shows the summary of the results for the more complex scenarios ( complete results are shown in Table S3 ) . In these more complex scenarios we assessed Θ by using coalescent simulations to estimate the average heterozygosity per site ( Θπ ) at neutral sites and by summing Θ across all subpopulations ( ΘΣ ) [26] . Our simulations confirm that only when both Θπ and ΘΣ become on the order of one or larger is it likely to observe fast evolution of complex 3-mutation alleles and at the same time soft sweep signatures . Strong selection does indeed improve the probability of seeing complex adaptive alleles but also , as expected , does not generate signatures of soft sweeps when Θ is small . Interestingly our simulations show that if Θ∼1 , then most of the observed signatures of soft sweeps are generated by multiple de novo mutations and are not due to the recombination of the same adaptive mutation onto different haplotypes . This is because signatures of soft sweeps are still commonly observed in simulations even when the recombination level is set at zero . It is also consistent with analytical considerations under simple scenarios ( Text S1 ) .
Our data and analysis strongly suggest that the patterns of adaptation observed at Ace in the last 1000–1500 generations are highly unlikely in a population in which Θ per site is on the order of 0 . 01 as it is commonly assumed . Instead , it appears that Θ per site must have been at least 0 . 1 and more likely on the order of one or larger . It is possible to elevate Θ by increasing the mutation rate or by increasing the effective population size . We assessed whether Ace had an unusually high mutation rate by estimating divergence of Ace in D . melanogaster from its D . simulans ortholog at synonymous sites . We found the divergence to be 7 . 9% , which is similar to the genome average of ∼10% [27] , [28] . In addition , Θπ per site estimated from polymorphisms at synonymous sites in sensitive alleles is 0 . 008 , which is also consistent with the genome average [18] . Thus we conclude that the effective population size in D . melanogaster over the past 1000–1500 generations is likely to be very large ( Ne≥108 ) . Such a large value of Ne might appear puzzling given that levels of standing neutral polymorphism suggest that Ne is much smaller [18] , [19] . To resolve this discrepancy it is necessary to take a closer look at the concept of an effective population size . Effective population size is commonly defined by the inverse magnitude of the frequency-fluctuations of a neutral allele in two consecutive generations [1] . Over a number of generations , effective population size is the harmonic mean of the effective population sizes over individual generations and thus is dominated by the smallest values of Ne . ( Equivalently , frequency fluctuations over many generations are dominated by the largest fluctuations over single generations ) . Estimates of the effective population size using frequent neutral polymorphisms reflect Ne harmonically averaged over long periods of time and are therefore very sensitive to any periods of low population size even far back into the past [29] . In sharp contrast , adaptation at Ace occurred within less than 1500 generations . The Ne relevant to adaptation at Ace is the harmonic mean of Ne values over the past 1500 generations or even fewer . Unlike Ne measured from ancient standing variation , it is not reduced by the bottlenecks and nearby selective sweeps that occurred more than 1500 generations ago . Consider a simple bottleneck scenario outlined in Figure 4 that is similar to the out-of-Africa scenario of Thornton and Andolfatto [18] . It is apparent that even if the current Ne is 100-fold larger than commonly assumed , population behaviour of a frequent neutral allele does not change substantially and the estimates of Θ from standing variation are not altered . To give another example , if D . melanogaster populations were to spend 90% of their time with Ne of 1010 and 10% at Ne of 105 with the shifts occurring about every 1000 generations , the harmonic mean Ne derived from common neutral polymorphisms would be ∼106 and yet the adaptive process would take place primarily in populations of 1010 with Θ>1 per site . In this case , strong adaptation in Drosophila would not be limited by mutation most of the time . The short-term Ne is bounded by the census population size ( N ) and thus if N is much smaller than the reciprocal of the mutation rate per site we can be certain that adaptation would be mutation-limited . In many species N can be much larger than the reciprocal of mutation rate and thus in these species it is possible that adaptation is not limited by mutation at single sites . However , it is Ne measured over time scales relevant for adaptation and not N that needs to be assessed to answer this question . Even short-term Ne might be much smaller than N if populations crash regularly on very fast temporal scales ( such as those induced by winters in temperate climates ) or if the numbers of successfully reproducing adults in each generation is sharply limited by extrinsic factors , for example by available substrates for laying eggs . Thus the studies of strong adaptation , such as the one presented here , are essential to determining whether adaptation in general is mutation-limited in a species . It is reasonable that Drosophila and many other organisms undergo recurrent boom-bust cycles thereby reducing the long-term Ne strongly but allowing adaptation during the boom years to occur in populations of large short-term Ne . In addition , Drosophila appears to undergo pervasive adaptation [30] , [31] with most common neutral polymorphisms estimated to have been affected by several selective sweeps in their genomic vicinity [28] . Such pervasive adaptation generates dynamics similar to recurrent bottlenecks and will also reduce the long-term Ne values even if the short-term Ne might be consistently large . This situation is similar to that found in HIV , where the effective population size estimated from observed diversity underestimates the census size by many orders of magnitude and is likely to underestimate the short-term Ne relevant for adaptation as well [32] . The possibility that adaptation at single sites in D . melanogaster is not limited by mutation has profound implications . The distinction between standing variation and de novo mutations at single sites is blurred since virtually all single-site mutations then exist in the D . melanogaster population at any given time . Strong adaptation should be much more rapid and generally result in soft sweeps . Complex adaptations that require multiple changes can be generated without fixation of interim states and with an enhanced chance of crossing fitness valleys [33] . This raises the question of whether the widespread use of the weak mutation , strong selection ( “WMSS” ) model for the study of adaptation should be broadened to include cases of strong mutation [34] , [35] . The number of sweeps ( hard or soft ) might also in general be lower than the number of adaptive substitutions if complex adaptations requiring multiple substitutions are common . Indeed , in our simulations of evolution at Ace in the strong mutation regime ( Θ per site on the order of 1 ) , the complex 3-mutation alleles generally evolve without fixation of intermediate 1- and 2-mutation alleles ( Figure 3 ) . The number of adaptive substitutions estimated using McDonald-Kreitman approaches should then be larger than the number of independent adaptive fixations and the prediction of the number of selective sweeps derived from the number of adaptive substitutions should be upwardly biased [36] . Note that all of these expectations hold especially well for strong selection because it operates over shorter time scales and is therefore less sensitive to recurrent but infrequent bottlenecks [37] and neighbouring selective sweeps . Most of the current statistical approaches for the study of adaptation rely on the expected signatures of hard sweeps [30] . Such methods should regularly miss or misidentify strong adaptation if it in fact commonly involves soft sweeps as in the case of Ace [20] . For example , if one searches exclusively for hard sweeps , then complete soft sweeps might appear as ongoing hard sweeps and the polymorphisms associated with the most frequent haplotype would appear as the likeliest candidates for the adaptive mutation whereas the true adaptive mutation would be fixed in the population . Methods exist that have high power to detect soft sweeps [20] , but they are used less often because soft sweeps have been considered unlikely a priori . However , a number of cases of adaptation in Drosophila and mosquitoes show clear signatures of soft sweeps [38]–[40] . Soft sweeps might also be common in humans , with the soft sweep associated with lactase persistence providing the strongest signature of adaptation in humans [41] , [42] . Our results suggest that the possibility of pervasive soft sweeps needs to be taken seriously . Recurrent boom-bust cycles are a general feature in population dynamics of most studied organisms . Adaptation and recurrent selective sweeps reducing the long-term but not the short-term Ne might also be common . It follows then that short-term and long-term Ne values are likely to be different as a rule . The shortest term Ne is only bounded by the census population size , which is often very large and can easily be in the billions , particularly for insects or marine organisms . It is thus possible that strong adaptation at single sites may not be limited by mutation in many eukaryotes , similar to the situation found in bacteria and viruses [32] .
We sequenced 1450 bp encompassing exons 2 through 4 of Ace . Resistant mutations I161V and G265A lie in the 3rd exon while F330Y and G368A lie in the 4th exon ( Figure S1 ) . Initially we sequenced this locus in 68 strains from 20 populations chosen to represent the Ace locus in a variety of geographical locations . The list of the populations and the number of lines investigated are given in Table S1 and Table S2 . For some of the strains that appeared heterozygous after sequencing of the PCR product , the DNA was first amplified using a proofreading DNA polymerase ( Platinum Pfx; INVITROGEN ) and cloned using Zero Blunt TOPO PCR cloning kit ( INVITROGEN ) before sequencing . Note that not all heterozygous strains were cloned , only those that contained a resistant mutation and the AF strains . The primers used for PCR amplification of the Ace locus were: PCR products were then sequenced . Of the 68 sequenced strains , 26/68 ( ∼40% ) have a single or multiple resistant mutations . Mutations at I161V , G265A and F330Y were identified in isolation and in combination in multiple populations , while G368A was never observed . We then used PASA [43] to identify strains that contained one or more of the three observed mutations and sequenced the identified strains . The primers used for PASA were: The 161 primer pair amplifies more effectively in the presence of the mutation I161V . The 265 primer pair is specific to G265A and the 330 primer pair is specific to G330Y . The annealing temperatures required for allele specific priming used for 161 , 265 and 330 were 61 . 5°C , 59 . 5°C and 60 . 6°C respectively . As positive and negative controls we performed PASA on strains in which the resistant sites had been previously characterized . We sequenced 37 strains from 8 populations that had amplified with one or more of the allele-specific primers . 31/37 ( 84% ) of these strains contained resistant mutations . The incorrect classification of the 6 strains is likely due to the addition of excess template to these PCR reactions resulting in non-specific priming . In total , we sequenced the Ace locus in 105 strains from 27 populations from five different continents ( Table S1 ) . Twelve of these strains were excluded from the analysis due to poor sequence quality . The most parsimonious haplotype network was constructed using TCS 1 . 21 [44] . All resistant alleles , except those for which we had poor sequence data , and all sensitive alleles observed more than once were used for the construction of the network . All AF strains and M strains were also included in the network to provide information on ancestral and modern variation respectively at the Ace locus . Measures of Θπ and divergence with Drosophila simulans at the Ace locus were obtained using DnaSP [45] . All sensitive strains analyzed in this study were used for the estimation . Our simulation models the population frequency dynamics of haplotypes at the 1 . 5 kb-long sequenced Ace locus and incorporates mutation , recombination , selection , and population substructure . Haplotypes are classified by their particular adaptive allele configuration at the three adaptive sites . We describe this configuration in terms of a vector a1a2a3 , indicating whether at site i the resistance-conferring mutation is present ( ai = 1 ) or not ( ai = 0 ) . A configuration 101 , for example , specifies resistant mutations at sites one and three , but no resistant mutation at site two . We use an infinite alleles model for new haplotypes , i . e . every mutation or recombination event at the locus is assumed to give rise to a new haplotype , which can be distinguished from all other haplotypes in the population . This is implemented in our simulations by assigning a unique ID to every new haplotype . The specific nucleotide sequence of the new haplotype is not relevant for our purposes; only changes in the adaptive-allele configuration are modelled explicitly . We also do not distinguish different sensitive haplotypes as we focus on the population dynamics of adaptive haplotypes . These simplifications substantially increase the performance of our simulations , allowing us to investigate scenarios with population sizes up to 109 in reasonable run-time . Mutations at adaptive sites and recombination events where the recombination breakpoint lies between two adaptive sites can generate new haplotypes with different adaptive-allele configuration ( Table S4 ) . Note that at each site only one specific nucleotide is the resistant allele and thus only one out of three mutations of a sensitive allele will give rise to it . The evolution of haplotype frequencies is simulated in terms of a Wright-Fisher model with directional selection , i . e . we assume panmictic subpopulations of constant size and non-overlapping generations [46] . Every haplotype h has a specific selection coefficient s ( h ) . The mean fitness of a subpopulation at time t is , where xh ( t ) is the frequency of haplotype h in the subpopulation at time t . Haplotype frequencies in generation t+1 are obtained by sampling from a multinomial distribution B ( 2N , {ph} ) with selection-adjusted probabilities . We group resistant haplotypes into three classes according to the number of resistance-conferring mutations they bear: 1m haplotypes have one resistant allele ( 100 , 010 , 001 ) , 2m haplotypes have two ( 011 , 101 , 110 ) , and 3m haplotypes have all three resistant alleles ( 111 ) . For simplicity , we assume that all haplotypes in the same class have equal selection coefficients s1m , s2m , and s3m , respectively . Prior to pesticide application all resistant haplotypes are modelled to be deleterious with selection coefficient −s1m . The key simulation parameters are the selection scenario defined by the selection coefficients s1m , s2m , and s3m , the recombination rate ρ , the number n of subpopulations , the migration rate M between subpopulations , and the value of Θ within subpopulations . We use a constant mutation rate of μ = 2 . 5 * 10−9 per site per generation [23] . Different Θ-values thus correspond to different subpopulation sizes . In particular , Θ = 0 . 01 corresponds to N = 106 , and Θ = 1 . 0 corresponds to N = 108 . We estimated a recombination rate of ρ = 0 . 15 cM/Mbp for our locus [24] , but investigate also other recombination rates in our simulations . Simulation runs start with one single sensitive haplotype present in all subpopulations at 100% frequency . Before pesticide application commences , mutation-selection equilibrium of resistant haplotypes is established within a burn-in period of 1000 generations . This fully suffices to establish equilibrium due to the strong purifying selection against all resistant haplotypes prior to pesticide application ( Box 1 ) . We also verified that longer burn-in times do not change our results . After the burn-in period , pesticide application starts by switching to the corresponding selection scheme . The simulation is then followed for another 1500 generations representing approximately 50 years of pesticide usage . During every generation individual subpopulations evolve according to the following steps: During a simulation run we analyze whether resistant haplotypes emerged and whether soft sweep signatures among 1m haplotypes were observed . We define 1m resistance by at least one of the three 1m adaptive-allele configurations ( 001 , 010 , or 100 ) ever being present in more than 10% of the population during the run . Accordingly , 3m resistance is defined by the complex 3-mutation allele ( 111 ) ever present in at least 10% of the population . A soft sweep signature ( ss ) is ascertained if at any time during the run two independently drawn alleles have greater than 10% probability to bear the same 1m configuration on different haplotypes . The statistics P1m , P3m , and Pss are the respective probabilities averaged over many runs . Pc denotes the combined probability that 3m resistance emerged and a soft sweep signature was observed during the same run . A crucial assumption of our simulation is the applicability of an infinite alleles model , i . e . all mutation or recombination events are assumed to be detectable . This can lead to an overestimation of the probabilities to observe soft sweep signatures in our simulations if independent mutation events frequently occur on the same haplotype , or if newly recombined haplotypes often resemble haplotypes already present in the population . We can estimate the resulting error from the probability that an individual is homozygous for the 1 . 5 kb-long locus . From coalescent simulations using ms [26] we infer it to be on the order of ∼10% when assuming a per site heterozygosity of out-of-Africa D . melanogaster subpopulations of Θπ∼0 . 5% [18] , [19] and the above specified recombination and mutation rates for our locus . Note , however , that in any case the infinite alleles model can only lead to an overestimation of the probability to observe soft sweep signatures . It is therefore always conservative in terms of our analysis . The probabilities P1m and P3m are not affected by the choice of an infinite alleles model . The simulation was implemented in C++ . Runs were performed on the Bio-X2 cluster at Stanford University . All source code is available from the authors upon request .
|
Adaptation in eukaryotes is often assumed to be limited by the waiting time for adaptive mutations . This is because effective population sizes are relatively small , typically on the order of only a few million reproducing individuals or less . It should therefore take hundreds or even thousands of generations until a particular new mutation emerges . However , several striking examples of rapid adaptation appear inconsistent with this view . Here we investigate a showpiece case for rapid adaptation , the evolution of pesticide resistance in the classical genetic organism Drosophila melanogaster . Our analysis reveals distinct population genetic signatures of this adaptation that can only be explained if the number of reproducing flies is , in fact , more than 100-fold larger than commonly believed . We argue that the old estimates , based on standing levels of neutral genetic variation , are misleading in the case of rapid adaptation because levels of standing variation are strongly affected by infrequent population crashes or adaptations taking place in the vicinity of neutral sites . Our results suggest that many standard assumptions about the adaptive process in eukaryotes need to be reconsidered .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2010
|
Evidence that Adaptation in Drosophila Is Not Limited by Mutation at Single Sites
|
Energy parasitism by ATP/ADP transport proteins is an essential , common feature of intracellular bacteria such as chlamydiae and rickettsiae , which are major pathogens of humans . Although several ATP/ADP transport proteins have so far been characterized , some fundamental questions regarding their function remained unaddressed . In this study , we focused on the detailed biochemical analysis of a representative ATP/ADP transporter ( PamNTT1 ) , from the amoeba symbiont Protochlamydia amoebophila ( UWE25 ) to further clarify the principle of energy exploitation . We succeeded in the purification of the first bacterial nucleotide transporter ( NTT ) and its functional reconstitution into artificial lipid vesicles . Reconstituted PamNTT1 revealed high import velocities for ATP and an unexpected and previously unobserved stimulating effect of the luminal ADP on nucleotide import affinities . Latter preference of the nucleotide hetero-exchange is independent of the membrane potential , and therefore , PamNTT1 not only structurally but also functionally differs from the well-characterized mitochondrial ADP/ATP carriers . Reconstituted PamNTT1 exhibits a bidirectional orientation in lipid vesicles , but interestingly , only carriers inserted with the N-terminus directed to the proteoliposomal interior are functional . The data presented here comprehensively explain the functional basis of how the intracellular P . amoebophila manages to exploit the energy pool of its host cell effectively by using the nucleotide transporter PamNTT1 . This membrane protein mediates a preferred import of ATP , which is additionally stimulated by a high internal ( bacterial ) ADP/ATP ratio , and the orientation-dependent functionality of the transporter ensures that it is not working in a mode that is detrimental to P . amoebophila . Heterologous expression and purification of high amounts of PamNTT1 provides the basis for its crystallization and detailed structure/function analyses . Furthermore , functional reconstitution of this essential chlamydial protein paves the way for high-throughput uptake studies in order to screen for specific inhibitors potentially suitable as anti-chlamydial drugs .
Members of the bacterial orders Rickettsiales and Chlamydiales gained attention because they comprise several human pathogenic species causing severe diseases like typhus , pneumonia , trachoma , or sexually transmitted infections [1 , 2] . The obligate intracellular lifestyle of these bacteria depends upon the continuous import of a large number of metabolites from the host cell cytosol [3–8] . Consequently , both metabolism and genome size of intracellular bacteria are typically and substantially reduced compared with those of free-living bacteria [2 , 9–11] . For example , genome sequencing of several rickettsial and chlamydial species revealed that these bacteria show restricted nucleotide metabolism . This restriction is characterized by the inability for de novo synthesis of certain nucleotides and an impaired ability to regenerate the universal energy currency ATP [11–13] . To compensate for these limitations , specific nucleotide transport proteins ( NTTs ) are used , mediating either net uptake of nucleotides , the import of NAD+ , or the counter-exchange of ATP and ADP [14–19] . The latter process has been termed “energy parasitism” and is considered fundamental for the survival of these metabolically impaired intracellular bacteria [4 , 20–22] . In addition to intracellular bacteria , nucleotide transport proteins have also been found in plant chloroplasts where they import cytosolic ATP under certain conditions [20] . This limited occurrence of nucleotide transport proteins in only few , largely unrelated groups of intracellular bacteria and plant plastids suggests an unusual evolutionary history for these proteins . Indeed , phylogenetic analyses showed that nucleotide transport proteins are of ancient origin ( and have evolved at least 700–1 , 000 million years ago ) , and early gene duplications and lateral transfer from rickettsiae to chlamydiae or from chlamydiae to rickettsiae and ( via the cyanobacterial ancestor of chloroplasts ) to plants has been suggested [12 , 18 , 23–25] . The model plant Arabidopsis thaliana possesses two isoforms of plastidic ATP/ADP transporters ( AtNTT1 and AtNTT2 ) . Arabidopsis mutant plants , which lack these non-mitochondrial nucleotide transporters , are retarded in plant development and exhibit a chlorotic phenotype and spontaneous necrotic lesions under short-day conditions [26 , 27] . The physiological and morphological differences between the Arabidopsis mutant plants and the wild type were compensated by extended light conditions ( long day or high light intensity ) [27] . Furthermore , reduction of ntt transcript in potato caused remarkable effects in heterotrophic but not in autotrophic tissues ( ginger shaped tubers with reduced starch contents ) [28] . Latter observations underline the importance of plant NTTs in photosynthetic-inactive heterotrophic plastids and in chloroplasts during periods of reduced or missing photosynthetic activity . Biochemical analyses of the recombinant AtNTT1 and AtNTT2 revealed that these transporters are highly specific for the substrates ATP and ADP . Site-directed mutagenesis of a lysine in a region that is conserved in several NTTs resulted in a markedly reduced capacity of AtNTT1 for ATP but not for ADP import [29] . To fulfil the essential physiological role of energy exploitation , all bacterial ATP/ADP transport proteins have to catalyze the import of host-derived ATP in counter-exchange with endogenous ADP . However , one key question that has not been satisfactorily explained yet is how the required net uptake of energy into intracellular living bacteria is maintained while external ATP and ADP might compete at the binding site . This problem is even more puzzling given that all bacterial ATP/ADP transporters characterized ( by heterologous expression in Escherichia coli ) so far exhibit nearly similar apparent affinities and transport velocities for ATP and ADP ( an overview is given in [18] ) . Thus , in theory , ATP/ADP transporters could also export bacterial ATP if the infected host cell exhibits a reduced energy state . Adenine nucleotide exchange across the mitochondrial membrane is mediated by specific ADP/ATP carriers ( AACs ) . Latter , well-characterized proteins are phylogenetically , structurally , and physiologically completely different from bacterial ( and also plastidic ) NTTs [30–33] . Whereas most NTTs catalyze ATP import into the bacterial cell ( or into the organelle ) , AACs mediate export of ATP synthesized in the mitochondrion to provide cellular metabolism with energy . In this respect , a very interesting feature of mitochondrial AACs is the regulatory influence of the membrane potential on nucleotide exchange , which was analyzed in detail by the help of artificial lipid vesicles containing the purified AAC from beef heart mitochondria [34] . In the absence of a membrane potential ( in “de-energized” mitochondria ) , ADP and ATP transport rates are nearly similar , whereas generation of a membrane potential ( in “energized” mitochondria ) stimulates ADP import in exchange with ATP . Whether bacterial nucleotide exchange is regulated by a similar mechanism is an interesting and still open question . Physiological studies on native bacterial ATP/ADP transporters were hampered by the need to isolate intracellular living bacteria from eukaryotic host cells , and all attempts to cultivate metabolically stable cells outside the host failed . Because existing paralogs and possible contamination by mitochondrial ADP/ATP carriers have to be taken into account [3 , 4 , 17 , 19 , 35] , and because purification of the native protein is still impossible , most NTTs characterized so far were analyzed by heterologous expression in E . coli [14 , 16–19 , 24 , 36 , 37] . However , to determine the catalytic activity of a single isoform ( rather than of a mixture of carriers present in a native bacterial membrane ) , to study the detailed effect of a counter-exchange substrate required to drive an antiport process , and to examine the biochemistry of a carrier uncoupled from metabolic fluxes in the living bacterial cell , it is necessary to incorporate the purified protein into artificial lipid vesicles . In the current study , we established the purification procedure of the heterologously expressed ATP/ADP transporter NTT1 from P . amoebophila ( PamNTT1 ) , a chlamydial endosymbiont of Acanthamoeba sp . [18 , 38] . This protein is an excellent and interesting candidate for further biochemical analyses , because ( i ) it is structurally and functionally similar and phylogenetically related to ATP/ADP transport proteins from important human pathogenic chlamydial species , ( ii ) it is involved in energy exploitation , and ( iii ) it is a recombinantly synthesized protein that is functionally inserted into the E . coli membrane at high amounts . The purification of PamNTT1 provided the basis for the functional integration into liposomes . To further clarify the mechanism of energy exploitation , special focus was laid on uptake kinetics of the reconstituted protein in the presence of internally applied ATP and ADP and also a possible influence of the membrane potential on nucleotide exchange was analyzed . Our results demonstrate that reconstitution of the purified transporter PamNTT1 represents a highly sophisticated tool to gain new and detailed insights into biochemical characteristics and into orientation–function relationships of this transport protein .
To perform a detailed biochemical analysis of an ATP/ADP transporter in artificial lipid vesicles , we expressed PamNTT1 heterologously in E . coli , purified the recombinant protein , and reconstituted the carrier into liposomes . E . coli-synthesized PamNTT1 is a suitable source for recombinant carrier proteins , because high amounts of the heterologous transporters accumulate functionally in the bacterial membrane during synthesis [18] . Moreover , the recombinant protein was extended by an N-terminal His tag , allowing affinity purification and immunodetection . Treatment of the E . coli membrane fraction with the detergent N-dodecyl-β-maltoside ( DDM ) led to efficient solubilization of a majority of the membrane proteins ( Figure 1 , lane 2 ) , including the recombinant PamNTT1 . SDS-PAGE confirmed that PamNTT1 was purified to apparent homogeneity using immobilized metal affinity chromatography ( IMAC ) ( Figure 1 , lanes 3–6 ) . By this method , it was possible to purify about 0 . 75 mg PamNTT1 from 1 l E . coli culture harvested 4 h after induction . To analyze the functionality of the purified transporter and to ascertain the time-linear phase of uptake , proteoliposomes were incubated in buffer medium containing radioactively labelled [α32P] ATP or [α32P] ADP . Reconstituted PamNTT1 mediated nucleotide uptake exclusively into vesicles containing ADP or ATP , whereas no measurable luminal accumulation of radioactivity occurred in unloaded proteoliposomes ( control ) ( Figure 2A and 2B ) or into loaded or unloaded liposomes lacking reconstituted PamNTT1 ( unpublished data ) . PamNTT1-mediated import of ATP , as well as ADP , into loaded vesicles was linear with time for at least 1 min of incubation ( Figure 2A and 2B ) . It appears remarkable that ATP uptake into proteoliposomes was always substantially higher than ADP uptake ( Figure 2A and 2B ) , because this contrasts with previous findings obtained in uptake studies of recombinant PamNTT1 using the heterologous host E . coli [18] . Interestingly , the data revealed an unexpected influence of the type of loaded substrate ( present at the liposomal interior ) on nucleotide import . In general , ATP import ( ATPim ) in exchange with internal ADP ( ADPex ) ( ATPim/ADPex ) was about two times faster than ATP import into ATP-loaded vesicles ( ATPim/ATPex ) , and ADP homo-exchange ( ADPim/ADPex ) was about ten times faster than ADP import in counter-exchange with ATP ( ADPim/ATPex ) ( Figure 2A and 2B ) . ATPim/ADPex was rapid and nearly equilibrated at 30 min ( at about 2 , 000 nmol/mg protein ) . Compared with the ATPim/ADPex hetero-exchange , ATPim/ATPex homo-exchange was slower , and the rate declined progressively until 20–30 min , when no further substantial accumulation of radioactivity was detectable ( at about 700 nmol/mg protein ) ( Figure 2A ) . PamNTT1-mediated ADPim/ADPex homo-exchange was slower than ATPim/ATPex homo-exchange and obviously approached equilibration after about 30 min . Remarkably , ADPim/ATPex hetero-exchange was only slightly stimulated when compared with the corresponding control ( Figure 2B ) . ADPim/ATPex was equilibrated at about 20 min ( at only 60 nmol/mg protein ) and appeared ( in the linear phase of uptake ) at least 20 times slower than in the reciprocal transport mode ( ATPim/ADPex ) ( Figure 2A and 2B ) . To further characterize the function of substrates present at the liposomal interior on nucleotide import , additional uptake studies were performed at different luminal ADP/ATP ratios ( 10 mM final nucleotide concentration; nucleotide import into ATP-loaded liposomes was set to 1 ) . Increasing ADP/ATP ratios in the proteoliposomes resulted in higher ATP uptake rates with a 2 . 34-times increase for the exclusive presence of ADP ( Table 1 ) . As described above , the nature of the luminal adenine nucleotide exhibited an even more noticeable effect on the relative ADP import affinities . Accordingly , a substantial increase of the ADP uptake rate becomes obvious in dependence on rising luminal ADP/ATP ratios , sole luminal ADP led to a 12-times increased PamNTT1-mediated ADP uptake ( Table 1 ) . To analyze whether the lack of the favoured counter exchange substrate ADP or the presence of ATP at the proteoliposomal interior caused the lowered nucleotide import capacity , the accumulation of radioactivity into liposomes loaded with different concentrations of ADP in the absence of any additional nucleotide ( ADP-NTP ) or in the presence of the nonsubstrate GTP ( ADP+GTP ) was determined ( Figure 3 ) . The resulting data were normalized to the import into liposomes loaded with 10 mM ADP ( =100% ) . In general , both loading conditions ( ADP-NTP and ADP+GTP ) led to nearly comparable uptake rates ( Figure 3 ) . Decrease of the internal ADP concentration to about 5 mM caused no significant reduction of nucleotide uptake into ADP-NTP– and ADP+GTP–loaded liposomes when compared to the corresponding import into proteoliposomes loaded with 10 mM ADP ( 100% ) . However , at an interior concentration of 2 . 5 mM ADP , a slight reduction of the nucleotide-import capacity of ADP-NTP– and ADP+GTP–loaded proteoliposomes was observed , probably caused by the limitation of counter-exchange substrate ( Figure 3 ) . Mixed ADP+ATP–loaded liposomes led to lower uptake rates when compared with the corresponding import into solely ADP- or ADP+GTP–loaded vesicles , and decreasing internal ADP/ATP ratios reduced ADP uptake ( Figure 3B ) to a higher extent than did ATP uptake ( Figure 3A ) . Because sole presence of GTP at the proteoliposomal interior caused slight nucleotide uptake , we assume that either GTP contains minor contaminations with adenine nucleotides or that very high concentrations of GTP may also function as substrate of the PamNTT1 . Our results demonstrate that reduction of import into the differently ADP+ATP–loaded liposomes is rather caused by the presence of the additional export substrate ATP than by a limitation of the counter-exchange substrate ADP . In most previous experiments , substrates were internally applied at high concentrations ( 2 . 5 to 10 mM ) , and the externally nucleotides were present at significantly lower concentrations ( 50- to 200-fold ) , which results in a nucleotide gradient across the liposomal membrane . To analyze whether this gradient causes the observed preference of ATPim/ADPex exchange and the low rates of the ADPim/ATPex transport , we analyzed nucleotide uptake into liposomes in the presence of a low ( Figure 4A and 4B ) , and in the absence of a nucleotide gradient ( Figure 4C and 4D ) . In general , nucleotide transport was slower and equilibrium was reached faster in proteoliposomes with a reduced nucleotide gradient ( Figure 4 ) when compared with uptake measurements under saturating interior nucleotide concentrations ( Figure 2 ) . This is due to limiting counter-exchange substrate concentrations at the proteoliposomal interior . For example , 100 μM ATPim/1 mM ADPex transport was equilibrated after 20 to 30 min at about 300 nmol/mg protein ( Figure 4A ) , and 50 μM ATPim/50 μM ADPex transport was equilibrated after 15 min at about 80 nmol/mg protein ( Figure 4C ) . In the linear time phase , PamNTT1 exhibited about 1 . 5 times higher transport rates for ATPim/ADPex , when compared with the ATP homo-exchange , and ADP homo-exchange is at least two times higher than the ADPim/ATPex transport ( Figures 2 and 4 ) . In accordance with the results obtained under saturating interior nucleotide concentrations ( Figure 2 ) , PamNTT1 prefers ATPim/ADPex exchange and suppresses the opposite direction of transport , and therefore , this feature is independent of the nucleotide gradient present at the vesicle membrane . It is well known that transport characteristics of mitochondrial ADP/ATP carriers are regulated by the membrane potential [34] . To analyze whether the potassium gradient in the applied system causes the observed preferred ATPim/ADPex transport , we determined transport activity of reconstituted PamNTT1 in the presence of different potassium concentrations at the liposomal interior or exterior . The ratios of exchange were unaffected by the applied buffer media ( unpublished data ) . We also compared the exchange rates of PamNTT1 and of mitochondrial AAC from yeast in the absence and the presence of a membrane potential . To minimize a possible leakage of potassium across the proteoliposomal membrane , which would result in the generation of a membrane potential , the same buffer media were applied at the interior and exterior side of the proteoliposome ( 30 mM potassium gluconate , 100 mM tricine , pH 7 . 5 ) . Hetero- and homo-exchange of ATP and ADP in the absence of a potassium gradient were measured and set to 100% ( Figure 5 ) . A membrane potential was generated by the application of the ionophore valinomycin . The direction and extent of the potential depends on the difference in the potassium concentrations at the interior and exterior side of the membrane . For the liposomes applied in this study , an external positive ( theoretical ) membrane potential of about +40 mV ( 30 mM K+ external/150 mM K+ internal: potassium efflux ) and an external ( theoretical ) negative membrane potential of −40 mV ( 150 mM K+ external/30 mM K+ internal: potassium influx ) were calculated , respectively . Transport measurements revealed that neither a positive nor a negative membrane potential substantially influenced nucleotide exchange of PamNTT1 , which was purified and reconstituted in presence of DDM ( Figure 5 ) or Triton ( Triton-X-100 ) ( unpublished data ) . However , mitochondrial AAC-mediated ADPim/ATPex transport was stimulated ( to about 160% ) by an external positive membrane potential , and ATPim/ADPex exchange was stimulated ( to about 190% ) by an external negative membrane potential ( Figure 5 ) . Our results demonstrate that in contrast to mitochondrial adenine nucleotide exchange , PamNTT1-mediated transport is independent of a membrane potential . To analyze in more detail the effect of the export substrate ADP on PamNTT1-mediated nucleotide import , we determined the kinetic parameters of this protein in ATP- , mixed ( ATP+ADP ) – , or ADP- ( 10 mM ) loaded proteoliposomes . Rising concentrations of exogenously supplied ATP led to increased rates of nucleotide uptake , reaching saturation at about 23 , 000 nmol ( mg protein ) −1 h−1 in the differently loaded proteoliposomes ( Table 2 ) . In presence of varied internal ADP/ATP ratios , the maximal transport velocities for ADP uptake ( of about 6 , 000 to 8 , 000 nmol ( mg protein ) −1 h−1 ) resembled the maximal velocities of the import into solely ADP- or ATP-loaded liposomes ( Table 2 ) . Interestingly , estimation of the apparent Michaelis-Menten constant ( KM ) values showed that PamNTT1 exhibited different substrate affinities depending on the type of counter-exchange nucleotide ( Table 2 ) . At a substrate concentration of 17 μM , PamNTT1 exhibited a half-maximal velocity of ATPim/ADPex , whereas a six-times-higher ATP concentration was required when solely ATP was internally present . Furthermore , the affinity of PamNTT1 for external ADP was at least ten times higher in the homo-exchange mode of transport ( ADPim/ADPex ) when compared with the hetero-exchange mode ( ADPim/ATPex ) ( ≥918 μM ) . Approximately equivalent KM values were estimated for the ADPim/ADPex and ATPim/ATPex homo-exchange mode of transport ( Table 2 ) . Analyses of kinetic parameters of ATP or ADP import into mixed ( ATP+ADP ) –loaded liposomes revealed intermediate affinities when compared with solely ADP- or ATP- loaded liposomes . Decreasing internal ATP/ADP ratios led to overall enhanced affinities of PamNTT1 for nucleotide import . The determination of kinetic parameters for ADP uptake was problematic in the presence of high internal ATP concentrations , and the resulting KM values are therefore possibly underestimated . Taken together , these data demonstrate that PamNTT1 imports ATP at highest maximal velocity ( Vmax ) independent of the loaded nucleotide and at highest affinity only if ADP , but not ATP , is present at the proteoliposomal interior . Because of the high affinity and maximal velocity , the ATPim/ADPex seems to be the favoured exchange . To analyze whether ADP under certain conditions can compete with ATP as an import substrate of PamNTT1 , the effect of simultaneous presence of different exterior ATP and ADP concentrations on transport into ATP- and ADP-loaded proteoliposomes was determined . Uptake of labelled nucleotide was performed in the absence ( set to 100% ) and presence of a nonlabelled competing substrate ( normalized to the corresponding 100% value ) . Our data demonstrate that the already-lowest concentrations of nonlabelled ATP effectively reduced [α32P] ADP import into ATP-loaded liposomes ( Figure 6A ) . The presence of 50 μM ATP ( 1/3 of the exterior ADP concentration ) led to a significant inhibition of the 150 μM [α32P] ADPim/ATPex to about 40% ( Figure 6A ) and of the 150 μM [α32P] ADPim/ADPex to about 66% ( Figure 6B ) when compared to the corresponding nonaffected ADP import . Twenty-fold excess of nonlabelled ATP reduced the import of 10 μM [α32P] ADP into ATP- and ADP-loaded liposomes to a value of about 8% ( Figure 6A ) and 20% ( Figure 6B ) , respectively . Although the inhibition pattern of nonlabelled ADP on [α32P] ATPim/ADPex ( Figure 6C ) resembles the influence of nonlabelled ATP on [α32P] ADPim/ADPex ( Figure 6B ) the rate of ADP import-inhibition by competing ATP is generally higher ( Figure 6B and 6C ) . The recombinant PamNTT1 exhibits lowest affinities and a low maximal velocity for ADPim/ATPex ( Table 2 ) . Therefore , it is not surprising that ATP can most effectively compete with ADP for import into ATP-loaded liposomes ( Figure 6A ) and that a 20-fold excess of ADP caused only a slight reduction ( to about 75% ) of the second-best [α32P] ATPim/ATPex transport ( Figure 6D ) . These data demonstrate that in accordance with the determined kinetic parameters , lowest concentrations of ATP ( 1/3 of the ADP concentration ) are sufficient to effectively displace the import substrate ADP ( Figure 6A ) and that high concentrations of ADP are unable to significantly suppress [α32P] ATP import in counter-exchange with ATP ( Figure 6D ) . The foregoing data demonstrate that PamNTT1 imports ATP with the highest velocity and exhibits the highest nucleotide ATP import affinity when ADP is present at the proteoliposomal interior . In contrast , PamNTT1 exhibits a low import velocity and a very low import affinity when ADP is present at the exterior and ATP at the opposite side . At first glance , this effect is hard to understand if PamNTT1 is inserted in a bidirectional ( nearly 1:1 ) orientation into the liposome . To examine whether the presence of either ATP or ADP during preparation of proteoliposomes might affect the insertion of PamNTT1 into the vesicle , we analyzed the efficiency of reconstitution and the orientation of PamNTT1 in differently loaded liposomes by protease treatment , SDS-PAGE , and immunostaining . Silver staining and immunodetection with an antiserum raised against purified PamNTT1 ( Figure 7A and 7B; lanes 1 , 4 , 7 ) or with a His tag–specific antibody ( Figure 7C; lanes 1 , 4 , 7 ) revealed that the present single band corresponds to the recombinant PamNTT1 and that the amount of reconstituted protein is comparable in differently loaded liposomes . To ascertain the orientation of PamNTT1 in the vesicles , unloaded or ADP- or ATP-loaded proteoliposomes were incubated with factor Xa . Proteolysis of the recombinant PamNTT1 by factor Xa resulted in the removal of a 2 . 4-kDa peptide containing the His tag ( Figure 7 ) . After factor Xa treatment , the ratio of uncut to cut protein is comparable in the differently loaded liposomes ( Figure 7A–7C; lanes 2 , 5 , 8 ) . Proteolysis of reconstituted PamNTT1 was only partial even over prolonged incubation ( unpublished data ) . However , destruction of the proteoliposomal integrity by the application of a high detergent concentration ( 2% DDM ) resulted in an almost complete cleavage of the N-terminal His tag ( Figure 7A–7C; lanes 3 , 6 , 9 ) . Thus , a ( by eye ) estimated amount of about 50% of the reconstituted PamNTT1 protein was not accessible to factor Xa and therefore obviously exhibited an orientation with the His tag directed to the lumen of the proteoliposome . These results demonstrated that both the efficiency and the direction of PamNTT1 insertion into the vesicle were not influenced by the type of counter-exchange substrate present during preparation of the proteoliposomes . The preferred ATPim/ADPex transport and the marginal transport rates of the opposite exchange ( ADPim/ATPex ) seem to argue against a bidirectional ( nearly 1:1 ) orientation of PamNTT1 in proteoliposomes . The observed ( directed ) transport activity of reconstituted PamNTT1 in liposomes can be explained if we postulate that only one mode of orientation exhibits total catalytic activity . To gain additional insights into orientation/function relationships of the reconstituted PamNTT1 , we analyzed the influence of endoprotease Asp-N treatment on the transport rate of the reconstituted transport protein ( Figures 8 and 9 ) . Proteolysis of ADP-loaded proteoliposomes for 2 h resulted in the cleavage of the majority of reconstituted PamNTT1 ( Figure 8A–8C; 2 h ) . Although , the amino acid sequence of PamNTT1 exhibits 16 potential cleavage sites located in different hydrophilic regions of the protein ( Figure 10 ) , a fraction of PamNTT1 obviously persisted protease treatment for at least 16 h ( Figure 8A–8C ) . To clarify whether the Asp-N–protected or the cleaved portion of reconstituted PamNTT1 corresponds to the functional protein , we determined the import capacity of protease treated ADP-loaded ( and unloaded ) proteoliposomes by transport measurements . Interestingly , no decrease of ATP transport was observed after 2 h of proteolysis , and even an extended incubation with Asp-N resulted only in a slight reduction of ATP uptake ( maximal 7% ) when compared with the control ( ATP import into untreated proteoliposomes ) ( Figure 8D ) . An accumulation of radioactively labelled ATP caused by a possible leakage of the liposomes due to protease treatment was ruled out , because Asp-N–treated as well as untreated unloaded proteoliposomes exhibited comparable , low background values ( less than 8 % of ATPim/ADPex ) . This analysis demonstrates that only a minor fraction of totally reconstituted PamNTT1 mediates nucleotide transport and that this functional protein is protected against Asp-N digestion . Consequently , it was necessary to determine the orientation of the Asp-N–protected , functional PamNTT1 proteins in liposomes . Therefore , we subsequently incubated reconstituted PamNTT1 with the endoproteases factor Xa and Asp-N , and we analyzed proteolysis of the carrier protein by SDS-PAGE and immunostaining . Treatment of the proteoliposomes with factor Xa allowed us to distinguish between the two orientations of the protein , and the subsequent use of Asp-N protease resulted in the identification of the functional protein . Thus , this approach is suitable to determine the orientation of the functional fraction of PamNTT1 . PamNTT1 extended by the His tag exhibited a single band at about 47 . 5 kDa ( Figure 9; lane 1 ) . In the presence of a high detergent concentration , both proteases completely digested PamNTT1 ( Figure 9; lanes 2–4 ) . Incubation of intact proteoliposomes with factor Xa resulted in the cleavage of the N terminus in about 50% of the reconstituted protein ( Figure 9; lane 5 ) . Factor Xa–treated liposomes were subsequently digested with protease Asp-N for 2 and 4 h , ( Figure 9; lanes 6 and7 ) . The resulting single band was still detectable by the His tag–specific antibody ( Figure 9C; lanes 6 and 7 ) ; it exhibited a molecular weight of about 47 . 5 kDa and therefore corresponds to the undigested PamNTT1 , whereas the truncated PamNTT1 ( lacking the His tag ) was completely fragmented ( Figure 9B; lanes 6 and 7 ) . These experiments demonstrate that the functional portion of PamNTT1 is Asp-N resistant ( Figure 8 ) . Moreover , the insensitivity of functional PamNTT1 towards factor Xa ( Figure 9 ) implies that the N terminus of this fraction is directed into the proteoliposomal lumen .
Chlamydial symbionts of free-living amoebae , such as P . amoebophila , exhibit remarkable similarities to human pathogenic Chlamydiaceae , including Chlamydia trachomatis or Chlamydophila pneumoniae , and thus can serve as model systems for the analysis of the interaction between intracellular bacteria and their host cells [39] . An essential common feature of both groups , and also of the parasitic Rickettsiales , is the capacity to perform nucleotide import and energy parasitism by specialized NTTs [14 , 17–19 , 24] . In contrast to all other bacterial ATP/ADP transporters analyzed so far , the recombinant homolog from P . amoebophila ( PamNTT1 ) investigated in this study meets the requirements for a detailed biochemical characterization , especially because purified PamNTT1 ( Figure 1 ) exhibited an extraordinarily high specific activity . The reconstituted carrier showed significant import of ATP or ADP into liposomes loaded with counter-exchange substrates ( ADP or ATP ) , whereas transport was absent in unloaded proteoliposomes ( Figures 2 and 4 ) . This provides evidence for the strict coupling of nucleotide import and export and confirms the previously described counter-exchange mode of PamNTT1 [18] . Although PamNTT1 exhibited a favoured import of ATP in exchange with ADP and moderate transport rates for the ATP and ADP homo-exchange , the import of ADP in exchange with ATP was only slightly in excess of the background , observed in unloaded liposomes . The preference of the ATP import in counter-exchange with ADP and the low activity of the opposite hetero-exchange were independent of the detergent applied during purification and reconstitution ( Figure S1 ) and of the nucleotide gradient across the liposomal membrane ( Figure 4 ) . Additional analyses confirmed that also the position of the His tag ( whether C-terminal or N-terminal ) did not affect the observed substrate preference pattern ( Figure S2 ) . The determination of the kinetic parameters revealed higher affinities of reconstituted PamNTT1 for both adenine nucleotides when ADP , instead of ATP , was present at the proteoliposomal interior ( Table 2 ) . Moreover , PamNTT1 generally exhibited higher maximal velocities for ATP than for ADP import , unaffected by the type of exported substrate ( Table 2 ) . Influence of the exported substrates on nucleotide import was investigated in more detail . Interestingly , vesicles loaded with increased ADP concentrations in the absence of ATP or in the presence of the nonsubstrate GTP led to higher import rates when compared with the corresponding transport into liposomes loaded with both substrates ATP and ADP ( Figure 3 ) . Furthermore , increase of the ADP concentration and simultaneous decrease of the ATP concentration in the proteoliposomes resulted in intermediate kinetic parameters and shifted the low affinities for nucleotide import in counter-exchange with ATP towards the high import affinities of the nucleotide uptake into ADP-loaded vesicles ( Table 2 ) . Thus , decrease of nucleotide import in the presence of rising internal ATP/ADP ratios ( Table 1 ) is a consequence of the reduction of the import affinities by the exported substrate ATP ( Table 2 ) . Because PamNTT1-mediated nucleotide transport into liposomes loaded with decreasing ADP concentrations in absence of ATP was always substantially higher than the corresponding transport in presence of rising concentrations of the additional interior substrate ATP , a possible limitation of the counter-exchange substrate ADP ( from 10 mM up to about 2 . 5 mM ) was ruled out ( Figure 3 ) . The stimulation of nucleotide import affinities by rising ADP/ATP ratios is in agreement with the physiological function of the carrier . A low-energy state of the bacterium ( simulated by a high ADP/ATP ratio in the liposome ) would consequently promote total nucleotide import , and would additionally stimulate ATP import to a higher extent than ADP import . Since not only the bacterial ATP level is subject to fluctuations , the effect of different external ADP/ATP ratios on nucleotide uptake was analyzed to mimic different energy states of the host cell ( Figure 6 ) . The determined competition pattern indicates that not even high concentrations of ADP could markedly affect ATP import into ATP-loaded proteoliposomes ( Figure 6D ) , whereas ADP import into ATP-loaded proteoliposomes was already exceedingly suppressed by low concentrations of ATP ( Figure 6A ) . The biochemical characteristics of the reconstituted PamNTT1 facilitate an effective accumulation of ATP in the liposome—even in the presence of exceeding amounts of exterior ADP ( Figure 6 ) —and in particular when the internal ATP/ADP ratio is low ( Tables 1 and 2 ) . These results confirm and mechanistically explain the postulated physiological role of this carrier , and at the first glance not only argue for only one mode of orientation in the liposomal membrane but also for a directed insertion , which corresponds to the native state . One important step towards the characterization of reconstituted proteins is the identification of their orientation in the vesicle . However , directionality of protein insertion varies with the method of reconstitution used ( for review , see [40–42] ) . Orientation analyses of PamNTT1 showed that this protein in the applied system consistently exhibits a bidirectional , nearly 1:1 orientation in the lipid vesicle ( Figure 7 ) . An extended analysis by the combination of endoproteases Asp-N and factor Xa , accompanied by additional transport measurements , revealed that only the fraction of PamNTT1 orientated with the N-terminally located His tag towards the liposomal lumen is functional ( Figure 8 and 9 ) . In this fraction of PamNTT1 , the exterior Asp-N cleavage sites are obviously not accessible ( Figure 10 ) . On closer inspection , a small fraction of PamNTT1 that was directed with the N terminus towards the interior also seems to have been digested by Asp-N . Since this cleavage has no influence on the transport activity , we assume that some of the carriers are correctly oriented but not functionally inserted . The calculated molecular weight of the largest fragment resulting from Asp-N cleavage of PamNTT1 inserted with the His tag to the exterior is 27 kDa . The complete digestion of PamNTT1 exposing the His tag to the exterior into fragments substantially smaller than 27 kDa argues for the accessibility of additional cleavage sites and leads to the assumption that a significant fraction of PamNTT1 is not correctly ( completely ) inserted . Because the calculated specific activity refers to the total protein applied for reconstitution , the high activity of reconstituted PamNTT1 is even underestimated . Additionally , in the applied liposomal system , the orientation of ADP/ATP carriers from yeast mitochondria was analyzed by the help of the specific inhibitors bongkrekic acid ( BKA ) and carboxyatractyloside ( CAT ) . AAC-mediated transport was strongly inhibited by 10 μM CAT ( to 52% ) or by 10 μM BKA ( to 20% ) when used separately , and was completely inhibited ( to about 0 . 5% ) when both inhibitors where applied simultaneously ( unpublished data ) . This reveals that the reconstituted mitochondrial AACs exhibited a nearly 1:1 orientation and that the right-side-out as well as the inside-out oriented carriers were functional under conditions used for the orientation analyses of the reconstituted PamNTT1 . Until now , the sense of integration of the PamNTT1 in P . amoebophila has been unknown , and therefore , a comparison of the orientation of the reconstituted functional protein with the native state was impossible . However , an analysis of the topology of proteins in membranes revealed that most of them exhibit an accumulation of positively charged amino acids at the cytosolic side [43] . In compliance with the predicted topology ( Figure 10 ) , the hydrophilic loops connecting the transmembrane domains 6 , 7 , 10 , and 11 of PamNTT1 contain a remarkable accumulation of positively charged amino acids , and therefore these positively residues are probably exposed to the cytosol of P . amoebophila . The demonstrated orientation of the functional fraction of PamNTT1 in the lipid vesicles is consistent with the predicted orientation of the native transporter . Whether PamNTT1 exhibits 11 ( as given in the current model ) or 12 transmembrane-spanning helices ( as previously described for structurally related NTTs ) will be analyzed in future [17 , 18 , 29] . Despite the fact that the biochemical properties of the reconstituted PamNTT1 are in line with the physiological function of this carrier in P . amoebophila , the obtained results are surprising . Given that at a pH 7 . 5 , ATP and ADP exist mainly as the species ATP4− and ADP3− , a 1:1 stoichiometry of the hetero-exchange would consequently generate a charge difference across the membrane of the vesicle . The electrogenic ADP/ATP exchange across the mitochondrial membrane was shown to be regulated by the energy state of the organelle . In energized mitochondria , the gradient of the substrate concentration across the inner mitochondrial membrane as well as the membrane potential ( positive outside ) supports the import of ADP3− in exchange with ATP4− , whereas the absence of a membrane potential results in similar transport rates for both adenine nucleotides ( Figure 5 ) [34] . In contrast to the mitochondrial adenine nucleotide exchange , NTT-mediated transport seems to be unaffected by the membrane potential and therefore electroneutral ( Figure 5 ) . The parasitic protist Entamoeba histolytica has reduced mitochondria homologs that lack an electron transport chain . The mitosomal AAC was shown to facilitate an electroneutral ADP/ATP exchange possibly by the uptake of a positive counter ion [44] . Isolated Rickettsia prowazekii cells prefer ATP instead of ADP import in diverse buffer media lacking phosphate [45] . However , rising concentrations of phosphate stimulated ADP import . Because rickettsiae are able to generate ATP through oxidative phosphorylation , the capacity to import ADP in the presence of phosphate is disadvantageous in a host cell that suffers from the infection [45] . The reduced expression of the ATP/ADP translocase gene in heavily infected host cells could prevent the loss of rickettsial ATP in exchange for host-derived ADP [46 , 47] . Until now , it has not been clear whether PamNTT1 compensates the charge difference occurring due to ATP/ADP exchange by the transport of a counter ion . However , since the buffer media used for reconstitution and uptake measurements contained no phosphate ( apart from possible traces by contamination ) , transport of ADP and phosphate in exchange with ATP is unlikely . Our studies revealed that the biochemical characteristics of the reconstituted PamNTT1 differ in some aspects from the results obtained with the E . coli system expressing PamNTT1 and homologous NTT proteins [18] . It is known that E . coli expressing recombinant carrier proteins suffer and exhibit a comparably high cytosolic ADP/ATP ratio [48] . In the proteoliposomal system , such conditions favour ATP import ( Figure 1 , Tables 1 and 2 ) , whereas E . coli exhibits nearly identical apparent affinities for both adenylates in the presence [18] and the absence of phosphate ( unpublished data ) . Thus , in contrast to the ATP/ADP transporter from R . prowazeckii , PamNTT1-mediated ADP import seems to be unaffected by phosphate [45] . Analysis of the substrate spectrum revealed that PamNTT1 is highly specific for ATP and ADP , independent of the applied system ( unpublished data and [18] ) . At this stage , it is not possible to decide which system is closer to the native state , because the E . coli system as well as proteoliposomes are in vitro models . E . coli differs considerably in form , size , and lipid composition from liposomes , and these differences might be the cause of the observed alterations in the biochemical properties of PamNTT1 . It is remarkable that the recombinant PamNTT1 exhibits a “sided-ness , ” which allows an efficient import of ATP in exchange with ADP when reconstituted into liposomes but not when measured in E . coli , in particular since the E . coli system at a first glance seems to be closer to the native environment . In fact , some points could also argue for the liposomal system as a better model for studying PamNTT1 . First of all , both PamNTT1 proteoliposomes and P . amoebophila are smaller than E . coli cells [18 , 38] , and therefore , an influence of the membrane curvature on PamNTT1 function can also be assumed . Second , it was shown that trafficking of host-derived phospholipids to C . trachomatis occurs , and thus the intracellular bacterium in contrast to E . coli harbors ( about 40% ) eukaryotic phosphatidylcholine ( PC ) in the plasma membrane [49–51] . Whether there is a possible correlation between the membrane curvature or the phospholipid PC and the functionality ( in particular the “sided-ness” ) of PamNTT1-mediated transport is the topic of our future studies . Taken together , heterologous expression of nucleotide transport proteins in E . coli is clearly suitable in identifying basic functional properties and mode of transport of these proteins , whereas the complexity of the E . coli cell prevents a more detailed functional analysis , which becomes feasible in the liposomal system . We suggest that the sophisticated , stimulating effect of the exported substrate ADP on import affinities of PamNTT1 , which was observed ( under diverse conditions ) in the less-complex liposomal system ( Figures 2 4 , S1 , and S2 and Tables 1 and 2 ) , is an intrinsic characteristic of the native PamNTT1 . This characteristic allows energy import into a bacterial cell that exhibits a reduced energy state even under conditions of a high ADP/ATP ratio in the host cell . Therefore , it also becomes evident why bacterial ATP/ADP transport proteins fundamentally differ in structure from mitochondrial AACs , which mediate energy export [31–34 , 52] . Interestingly , an influence of the exported substrate on import characteristics of an NTT protein has been previously observed on intact E . coli cells . The recombinant NAD+/ADP transporter from P . amoebophila ( PamNTT4 ) mediated NAD+ import in counter-exchange with ADP and also ADP and NAD+ homo-exchange , respectively , whereas the physiologically unfavourable import of ADP in counter-exchange with NAD+ was nearly unobservable [16] . Whether the observed biochemical characteristics of PamNTT4 are also caused by different kinetic parameters at interior and exterior and by a stimulating influence of the exported substrate ADP on import affinities and whether the trans-stimulating effect of ADP is a common feature of all bacterial and plastidic ATP/ADP transporters remains to be investigated . The establishment of an expression , purification , and reconstitution protocol for a representative NTT from intracellular living bacteria and the generated PamNTT1-specific antibodies are fundamental prerequisites for crystallization; detailed structure–function analyses in the future and will also provide important guidance for NTT inhibitors , which might represent suitable drugs for specific anti-chlamydial therapy .
To generate C-terminal His tag fusion constructs , the PamNTT1 coding sequence was amplified from the pET 16b construct [18] by PCR using Pfu polymerase and the following oligonucleotides: NTTBamHI_sense 5′-GGATCCATGTCGCAAGATGCGAAACAGAC-3′ , NTTNheI_sense 5′-GCTAGCATGTCGCAAGATGCGAAACAAGAC-3′ , and NTTXhoI_antisense 5′-CTCGAGGCTAGTAGCTATTTCCGATGT-3′ . The PCR products containing the generated NheI/XhoI or BamHI/XhoI sites were inserted into a pET 21a plasmid ( Merck Biosciences; http://www . merckbiosciences . co . uk/home . asp ) . Dependent on the position of the insertion in the pET 21a plasmid , the resulting proteins were extended with either a C-terminal His tag ( NheI/XhoI ) or a C-terminal His tag and an N-terminal T7 tag ( BamHI/XhoI ) . DNA manipulations were performed essentially as described in Sambrook et al . [53] . Correctness of the constructs was proven by complete sequencing ( NanoBioCenter , TU Kaiserslautern , Germany ) . The pET16b as well as the pET21a constructs containing Pamntt1 were transformed into BLR ( DE3 ) pLysS cells ( Merck Biosciences ) . The transformed bacterial cells were grown at 37 °C in Terrific Broth medium [48] . At an A600 of 0 . 5–0 . 6 initiation of T7 RNA polymerase expression was induced by addition of isopropyl-β-d-thiogalactopyranoside ( 1 mM ) . The cells were harvested 4 h post-induction by centrifugation ( 5 , 000g , 10 min , 20 °C ) , resuspended in buffer medium ( 1 mM EDTA , 15% glycerol , 10 mM Tris , pH 7 . 0 ) and immediately frozen in liquid N2 . To induce autolysis by the endogenous lysozyme , frozen cells were thawed at 37 °C and treated with a pinch of DNAse and RNAse . To avoid proteolytic activity , phenylmethanesulphonylfluoride ( PMSF ) ( 1 mM ) was added . Cell lysis was supported by sonication . In the first centrifugation step ( 20 , 000g , 15 min , 4 °C ) , cell debris were removed and membranes were collected from the supernatant by centrifugation ( 100 , 000g , 30 min , 4 °C ) . All purification steps were carried out at 6 °C . Membranes ( see above ) were resuspended in buffer medium B ( 10 mM NaCl , 5 mM imidazole , 50 mM Na2HPO4 , pH 7 . 9 ) containing 1% DDM ( Glycon Biomedicals; http://www . glycon . de/ ) , 70 mM NLS ( Sigma; http://www . sigmaaldrich . com ) or 1% Triton ( Sigma ) and stirred for 1 h to allow solubilization of intrinsic proteins . After centrifugation of the solubilized membranes ( 100 , 000g , 30 min , 4 °C ) , the supernatant was incubated with Ni-NTA agarose ( Qiagen; http://www . quiagen . com ) for 2 h . The Ni-NTA suspension was transferred onto a chromatography column , washed with 10 volumes of the Ni-NTA agarose bed volume with buffer medium B containing 0 . 5% DDM , 70 mM NLS , or 0 . 5 % Triton , and afterwards , with 6 column volumes of buffer medium W ( 10 mM NaCl , 60 mM imidazole , 50 mM Na2HPO4 , pH 7 . 9 , 0 . 5% DDM , 70 mM NLS , or 0 . 5% Triton ) . Subsequently , recombinant PamNTT1 was eluted with an appropriate amount of buffer medium E ( 10 mM NaCl , 500 mM imidazole , 50 mM Na2HPO4 , pH 7 . 9 , 0 . 1% DDM , 35 mM NLS . or 0 . 2% Triton ) to obtain a protein concentration of 0 . 4–0 . 5 mg/ml . Protein concentrations of the DDM-purified NTT were quantified according to the method described by Bradford [54] . 500 μl of Bradford solution were mixed with H2O and eluted protein ( 5–10 μl ) to reach a final volume of 1000 μl . The same volumes of elution buffer ( containing detergent ) were used for the blank values . Protein concentrations of the NLS and the Triton-purified proteins were quantified photometrically at 280 nm . To reduce nonprotein absorbance generated by the high imidazol concentration in the elution buffer , the protein was desalted on a NAP-5 column equilibrated with buffer ( 10 mM NaCl , 50 mM Na2HPO4 , pH 7 . 9 , 35 mM NLS , or 0 . 2% Triton ) . The same buffer media were used for the blank values . All measurements were carried out in a BioPhotometer ( Eppendorf; http://www . eppendorf . com ) . Immediately after purification of PamNTT1 ( see above ) , 50 μl of the eluate was mixed with a homogeneous emulsion of 400 μl l-α-phosphatidylcholine ( 125 mg/ml , type IV-S , Sigma ) in buffer medium TG1 ( 100 mM tricine , 30 mM potassium gluconate , pH 7 . 5 ) . For loading , 50 μl of the indicated nucleotides were added to achieve the indicated concentrations . After vigorous vortexing , the mixture ( 500 μl ) was frozen in liquid N2 . Proteoliposomes were thawed on ice and sonified for 20 s ( Sonifier 250 , Branson; http://www . bransonultrasonics . com ) at lowest output level and 50% duty cycle . Proteoliposomes ( 500 μl ) were applied to a NAP-5 column ( GE Healthcare; http://www . gehealthcare . com ) equilibrated with TG2 ( 10 mM Tricine , 150 mM potassium gluconate , pH 7 . 5 ) to remove external nucleotides . Proteoliposomes were eluted from the column with 1000 μl of TG2 . For protease cleavage assays , the NAP-5 column was equilibrated with cleavage buffer medium ( 100 mM NaCl , 20 mM Tris , 2 mM CaCl2 , pH 8 . 0 ) . The endoproteases factor Xa ( New England Biolabs; http://www . neb . com ) and Asp-N ( Sigma ) were added to proteoliposomes to obtain a protein/protease ratio of 20/1 . To allow a combination of Asp-N treatment and uptake experiments , the cleavage was carried out in buffer medium TG2 . Protease activity was stopped with 1 mM PMSF ( factor Xa ) or 10 mM EDTA ( Asp-N ) at the indicated time points . All protease assays were carried out at 25 °C . Radioactively labelled [α32P] ADP was synthesized enzymatically by using hexokinase following a standard protocol [55] . Quality and purity of synthesized [α32P] ADP was analyzed by thin layer chromatography as described elsewhere [55 , 56] . Proteoliposomes were added to 100 μl of TG2 containing [α32P] ATP ( NEN; http://www . perkinelmer . com ) or [α32P] ADP at indicated concentrations . For transport measurements , proteoliposomes were incubated at 30 °C and uptake was terminated at the indicated time periods by removal of residual nucleotides using anion exchange chromatography ( Dowex 1 × 8 Cl , 200–400 mesh , Sigma ) [34] . Liposomes were eluted from the chromatography column by addition of 1500 μl Tricine ( 200 mM , pH 7 . 5 ) . Radioactivity in the eluate was quantified using a scintillation counter ( Canberra-Packard; http://www . canberra . com ) . Before separation by SDS-PAGE , proteoliposomes were treated with SDS ( 4% ) and with an appropriate volume of 6 × concentrated sample buffer medium ( 375 mM Tris/HCl , pH 6 . 8 , 0 . 3% SDS , 60% glycerol , 1 . 5% bromophenol blue ) . Proteins were separated in a discontinuous , denaturing system with a 3% stacking and a 12% separating polyacrylamide gel as described by Laemmli [57] . Following electrophoresis , gels were stained [57 , 58] or proteins were transferred to a nitrocellulose membrane in a wet-blotting apparatus . Immunodetection was performed using a monoclonal anti poly His immunoglobulin G ( IgG ) ( Sigma ) combined with a secondary alkaline phosphatase conjugated anti-mouse IgG ( Sigma ) , or by an anti-PamNTT1–specific serum combined with a secondary alkaline phosphatase conjugated anti-rabbit IgG ( Promega; http://www . promega . com ) . After staining with nitro blue tetrazolium chloride/5-bromo-4-chloro-3′-indolyl phosphate toluidine , salt blots were rinsed in water and air-dried . To estimate molecular masses , a prestained broad range protein marker ( 6–175 kDa , New England Biolabs ) , was used . PamNTT1-specific anti serum was generated at Eurogentec ( http://www . eurogentec . com ) by immunization of rabbits with IMAC-purified recombinant protein ( see above ) .
The National Center for Biotechnology Information ( http://www . ncbi . nlm . neh . gov ) accession number for PamNTT1 is AJ582021 .
|
Diverse members of the bacterial order Chlamydiales cause severe diseases in humans . Chlamydiales cannot survive and reproduce outside of host cells , due to the complete loss of several biosynthetic pathways , but rely on specialized transport systems to import the corresponding metabolites from the host . We performed a detailed biochemical analysis of a purified recombinant ATP/ADP transporter from the Chlamydia-related bacterium Protochlamydia amoebophila ( PamNTT1 ) . Our studies revealed that PamNTT1 favours ATP import into vesicles loaded with ADP , whereas the exchange of ATP in the opposite direction is negligible . Furthermore , we demonstrated that PamNTT1 , in contrast to the mitochondrial ADP/ATP carrier , is independent of a membrane potential . The identified biochemical characteristics of PamNTT1 appear to be a sophisticated adaptation to the requirements of the intracellular lifestyle of P . amoebophila , serving to facilitate effective energy parasitism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"biochemistry",
"infectious",
"diseases",
"microbiology",
"eubacteria"
] |
2007
|
Enlightening Energy Parasitism by Analysis of an ATP/ADP Transporter from Chlamydiae
|
Botulism , characterized by flaccid paralysis , commonly results from botulinum neurotoxin ( BoNT ) absorption across the epithelial barrier from the digestive tract and then dissemination through the blood circulation to target autonomic and motor nerve terminals . The trafficking pathway of BoNT/A passage through the intestinal barrier is not yet fully understood . We report that intralumenal administration of purified BoNT/A into mouse ileum segment impaired spontaneous muscle contractions and abolished the smooth muscle contractions evoked by electric field stimulation . Entry of BoNT/A into the mouse upper small intestine was monitored with fluorescent HcA ( half C-terminal domain of heavy chain ) which interacts with cell surface receptor ( s ) . We show that HcA preferentially recognizes a subset of neuroendocrine intestinal crypt cells , which probably represent the entry site of the toxin through the intestinal barrier , then targets specific neurons in the submucosa and later ( 90–120 min ) in the musculosa . HcA mainly binds to certain cholinergic neurons of both submucosal and myenteric plexuses , but also recognizes , although to a lower extent , other neuronal cells including glutamatergic and serotoninergic neurons in the submucosa . Intestinal cholinergic neuron targeting by HcA could account for the inhibition of intestinal peristaltism and secretion observed in botulism , but the consequences of the targeting to non-cholinergic neurons remains to be determined .
Botulinum neurotoxins ( BoNTs ) are responsible for a severe nervous disease in man and animals known as botulism , characterized by skeletal muscle flaccid paralysis and respiratory arrest , resulting from inhibition of acetylcholine ( ACh ) release in peripheral cholinergic nerve terminals . BoNTs are produced by Clostridium botulinum as single chain proteins ( ap . 150 kDa ) , which are divided into 7 toxinotypes ( A to G ) according to their immunogenic properties . The toxins are exported outside the bacteria and are proteolytically cleaved into a heavy chain ( H; ap . 100 kDa ) and a light chain ( L; ap . 50 kDa ) , which remain linked by a disulfide bridge . The di-chain molecule constitutes the active neurotoxin . The half C-terminal domain of the H-chain ( Hc ) is involved in binding to specific receptors on target neuronal cells and in driving the toxin entry pathway into cells , whereas the N-terminal part permits the translocation of the L chain into the cytosol . The L chain catalyzes a zinc-dependent proteolysis of one or two of the three proteins of the SNARE complex , which play an essential role in evoking neurotransmitter exocytosis . The BoNT/A L-chain cleaves the synaptosomal associated protein SNAP25 at the neuromuscular junction [1]–[4] . The highly specific binding of BoNTs to target nerve endings involves protein and ganglioside receptors that localize at the neuronal plasma membrane [5] . Gangliosides of GD1b and GT1b series are involved in binding and functional entry into cells of BoNT/A and BoNT/B [6]–[9] . The protein receptors on neuronal cells have been identified as synaptotagmin I and II for both BoNT/B and BoNT/G , and synaptic vesicle protein SV2 ( isoforms A , B and C ) for BoNT/A [8] , [10]–[14] , BoNT/E [15] , and BoNT/F [16] , [17] . SV2C is the preferred BoNT/A neuronal receptor [14] , whereas BoNT/E recognizes glycosylated SV2A and SV2B [15] . BoNT/D also uses SV2 proteins as receptor in association with gangliosides for its entry into neuronal cells , but binds to SV2 via a distinct mechanism than BoNT/A and BoNT/E [18] . In addition , SV2A and SV2B have also been evidenced to mediate the entry of tetanus toxin ( TeNT ) into the central target neurons including hippocampal and spinal cord neurons [19] . Botulism usually results from the ingestion of preformed neurotoxin in contaminated food , or ingestion of spores or bacteria , which under certain circumstances , may colonize the gut and produce the neurotoxin in situ [20] . In either case , BoNT escapes the gastro-intestinal tract to reach the target cholinergic nerve endings , possibly through the blood and lymph circulation [21] . Indeed , previous observations have shown that after oral administration of BoNT in experimental animals , the toxin enters the blood and lymph circulation . The upper small intestine was found to be the primary site of absorption [22]–[25] , but BoNT can also be absorbed from the stomach [21] . Penetration of BoNT through an epithelial cell barrier and its subsequent migration to cholinergic nerve endings are the essential first steps of botulinum intoxication . In in vitro models , BoNTs have been found to bind to polarized epithelial cells and to undergo receptor-mediated endocytosis and transcytosis from apical to basolateral sides [24] , [26]–[30] . However , little is known about the precise pathway of BoNT migration from the intestinal lumen to the target nerve endings . The digestive tract contains its own independent nervous system , the enteric nervous system ( ENS ) , which is as complex as the central nervous system , and it is also referred as the “brain of the gut” . ENS controls and coordinates motility , exocrine and endocrine secretions , and blood microcirculation of the gastrointestinal tract . Nerve cell bodies of ENS are clustered into small ganglia which are organized in two major plexuses: the myenteric plexus between the longitudinal and circular muscle layers , and the submucosal plexus associated with the mucosal epithelium between the circular muscles and the muscularis mucosa . Ganglia also contain glial cells and their extensions . ENS neurons can be classified as afferent sensory neurons , interneurons , and motor neurons , which are connected to the central autonomic nervous system through both sensory and motor pathways . More than 20 types of neurotransmitters have been identified in ENS , and most enteric neurons may produce and release several of them . However , neurotransmitter functions have not been fully identified . Secretory and motor neurons are cholinergic , these latter also contain substance P . Myenteric neurons are connected to the cholinergic parasympathetic neurons through nicotinic , and in some areas , muscarinic receptors [31] , [32] . Vasoactive intestinal protein ( VIP ) and serotonin are also major neurotransmitters in the regulation of normal gut function and interconnection with the central nervous system [33] . In this study , we used fluorescent Hc fragment from BoNT/A to monitor the trafficking of the toxin into the mouse intestinal mucosa . It has been previously shown that the Hc domain from TeNT , which shares similar structural organization and catalytic activity with BoNTs , is a useful tool to investigate the intracellular trafficking of the neurotoxin [34]–[36] . Although , it cannot be ruled out that a cross interplay between the BoNT domains may modify the toxin routing driven by Hc [37] , recombinant HcA has been reported to retain the same structure than that of the receptor binding domain of the BoNT/A holotoxin and to enter hippocampal neurons similarly to the whole neurotoxin [16] , [38] . In addition , HcA has been found to bind and transcytose through intestinal cells as well as the holotoxin [26] , [39] , validating its use to investigate the intestinal trafficking of the toxin .
In striated muscle , BoNT/A is known to inhibit ACh release from motor nerve terminals by cleaving the synaptosomal associated protein SNAP-25 , leading to the inability of synaptic vesicles containing ACh to undergo transmitter release [for a review , see [3]] . In the gastrointestinal smooth muscle , BoNT/A also impairs cholinergic transmission by inhibiting ACh release from postganglionic cholinergic nerve endings in vitro and in vivo [40] , [41] . The first aim of this study was to determine whether BoNT/A affected smooth muscle contractility when applied intra-luminally on isolated mouse ileum segments . In preparations that were equilibrated in the standard oxygenated solution for about 30 min , spontaneous contractile responses were usually observed at a frequency rate of about 6–10 min−1 ( n = 4 ) . These spontaneous contractions had a peak force that was variable from preparation to preparation , but comprised between 0 . 5 and 1 . 2 g ( n = 4 ) . BoNT/A reduced their frequency after 2 , 3 and 4 h of the intralumenal injection ( Figure 1A ) . It is worth noting that for control preparations maintained for 4 h in the same conditions as the ones treated with BoNT/A , not only there was no reduction of the spontaneous contractions , but a small increase in their frequency ( 10% after 2 and 3 h ) ( Figure 1A ) . The contraction pattern changed also after exposure to BoNT/A , from very regular oscillations during the first hour to very irregular contractions after more than 2 h ( Figure 1B and 1C ) . Electric field stimulation evoked contractile responses that attained a peak force comprised between 2 and 7 mN ( n = 3 ) under control conditions ( Figure 1E ) , with little rundown of the responses when stimulations were applied every 50–60 min ( Figure 1D ) . As shown in Figure 1D , BoNT/A reduced the electrically-evoked contractile response in a time-dependent manner . The time to decrease to 50% the evoked tension-time integral response was about 110 min , and the toxin reduced to about 90% electrically-evoked contractions within 240 min . Although it is difficult to exclude the possibility of direct muscle stimulation by the applied field-stimuli , several lines of evidence indicate that this would represent no more than 10–15% of the evoked tension-time integral response in our experimental conditions . Most of this evidence comes from: ( i ) Data obtained with BoNT/A showing that blockade of the evoked contraction attains a maximum around 86–90% of control values , and was never complete . The remaining tension can be suspected to be due to direct muscle stimulation unaffected by BoNT/A . ( ii ) If direct stimulation of the muscle would occur , one would expect that tension levels would be sustained and maintained during the field stimulation , which is not the case ( Figure 1E ) . ( iii ) Spontaneous contractions occurring during the falling phase of the evoked-contractile response were not enhanced in amplitude , which is consistent with the low influence of direct muscle stimulation under the experimental conditions used . Interestingly , after BoNT/A has blocked the evoked response by field stimulation , the addition of carbachol ( 20 µM ) or ACh ( data not shown ) to the standard solution evoked a contractile response ( Figure 1F ) . These results suggest that under the conditions used BoNT/A is able to exert an action on cholinergic terminals that leads to a blockade of the contractile responses evoked by electric field stimulation , while spontaneous myogenic contractions were reduced in frequency , but not completely abolished . The fact that carbachol could induce contractile activity after BoNT/A-induced blockade of contraction evoked by electric field stimulation , strongly suggests that the sensitivity of ACh receptors ( muscarinic and nicotinic ) is not affected by the toxin ( Figure 1F ) . In a previous report , we have found that BoNT/A transcytosis through intestinal cell monolayers grown on filters is mediated by SV2C or , at least , an immunologically related protein [30] . To address whether SV2C might be a functional receptor in the ex vivo intestinal tract model , BoNT/A was preincubated with the intravesicular domain segment L4 of SV2C prior to injection into ligated intestinal loop . As shown in Figure 1A , preincubation with SV2C/L4 significantly prevented the BoNT/A inhibitory effects on spontaneous intestinal smooth muscle contractions , suggesting a partially SV2C-dependent BoNT/A uptake through the intestinal mucosa . However , these results do not rule out that SV2C/L4 also passed , independently or associated with BoNT/A , through the intestinal barrier and impaired BoNT/A uptake by nerve terminals . We first investigated the potential binding sites for HcA in mouse intestine mucosa and submucosa , as well as in the musculosa . For that , cryosections of mouse small intestine , fixed on glass slides , were incubated with fluorescent HcA and analyzed by confocal microscopy . Since BoNT has been reported to be preferentially absorbed from the upper small intestine [23] , [24] , sections from ileum were analyzed . Only a faint staining was observed in brush border of enterocytes along intestinal villi ( Figure 2A ) . However , a strong HcA binding was observed on intestinal crypt localized at the bottom of villi . Paneth cells , which are characterized by their numerous secretory-granule content , are spatially restricted to intestinal crypts and were used as a marker of these regions . Staining of Paneth cells with the TRITC-labeled lectin Urex europaeus agglutinin type 1 ( UEA1 ) [42] did not significantly colocalize with HcA ( Figure 2B ) . This may indicate that some cells from intestinal crypts , distinct from Paneth cells , exhibit preferential binding sites for HcA . Moreover , small cells scattered along the villi were stained with HcA ( Figure 2A ) and some of them co-stained with UEA1 ( Figure 2B ) . These UEA1 positive cells likely correspond to goblet cells [43] . It is worth noting that HcA stained neuronal cell bodies and neuronal structures in the submucosa and musculosa . However , only a low proportion of neuronal structures were recognized and labeled by HcA , as revealed by immunolabeling neurofilaments and co-staining with the fluorescent toxin fragment ( Figure 2C ) . Hence , the binding domain of BoNT/A potentially targets epithelial cells in intestinal crypts , and neuronal structures of intestinal plexuses . Note that anti-neurofilament antibodies are not specific of the nerve endings , where BoNT/A is assumed to bind , but recognize neuronal structures all along the neuronal cells . This probably accounts for the irregular co-staining between HcA and anti-neurofilament antibodies . In addition , the irregular pattern of HcA staining might also be related to the variability in orientation and size of the cryosections . Moreover , cryosections might artificially expose certain antigens which are buried in intact tissues . Thus , immunostaining pattern in cryosections has to be considered with caution and confirmed in ex vivo experiments with intact tissues as shown in the following figures . To analyze HcA entry into the intestinal mucosa and submucosa , ex vivo experiments were performed , as previously described with the whole toxin . For this , excised small intestine loops were washed , ligated at both extremities , and incubated in oxygenated Krebs-Ringer solution at 37°C . Fluorescent HcA was inoculated into the intestinal lumen , and at various time intervals , intestinal loops were washed , fixed and processed for dissection , and immunostaining . A competition assay between HcA-Cy3 and native BoNT/A injected into an ileum loop and monitored by fluorescence analysis of the intestinal mucosa , supported that fluorescent HcA follows the same entry pathway than native BoNT/A ( Figure 3A ) . Fluorescent HcA entered similarly ileum , duodenum or jejunum segments , as tested by mucosal fluorescence analysis ( Figure 3A ) . After 30–60 min incubation , labeled HcA was detected inside the lumen of intestinal crypts , and in some crypt cells , but not or with a low intensity in enterocytes or other cells in the villi ( Figure 3 B and C ) . HcA also labeled long cell extensions in the submucosa , which correspond to nerve fibers or neuronal extensions ( Figure 3D; arrow head ) since they were co-stained with antibodies against neurofilaments ( not shown ) . Longer incubation periods ( 90–120 min ) permitted to visualize HcA staining of long filaments in the musculosa , ( Figure 3E ) , which were identified as nerve fibers from the myenteric plexus ( see below ) , but with a weaker intensity . This suggests a progressive entry of HcA from the intestinal lumen through the mucosa , preferentially through intestinal crypts , to certain neuronal cell and extensions in the submucosa , and then in the musculosa . To identify the intestinal crypt cells targeted by HcA , fluorescent HcA was injected into the lumen of an intestinal loop . After an incubation of 15–30 min , the intestinal mucosa was prepared for microscopy observation . Only few numbers ( 1 or 2 ) of small cells from each intestinal crypt were stained with HcA ( Figure 4 ) . Cells stained with HcA were distinct from Paneth cells , which were easily detectable by their numerous granules ( Figure 4 ) , and by their staining with UEA1 ( not shown ) . Chromogranin-A antibodies , a common marker of neuroendocrine cells in the gastrointestinal tract [44] , colocalized with HcA ( Figure 4A ) . All cells stained with HcA were also stained with chromogranin-A antibodies , indicating that HcA specifically entered neuroendocrine cells from intestinal crypts . However , not all chromogranin-A positive cells were stained with HcA , but only about 80% . Serotonin-producing cells , which are abundant in the ENS , were investigated for their colocalization with HcA . In the intestinal crypts , all the cells stained with HcA were also immunolabeled with serotonin antibodies ( Figure 4B and C ) . Interestingly , HcA accumulated in the basal pole of neuroendocrine cells , which is wider than the apical pole exposed to the intestinal crypt lumen . This strongly supports that HcA uses neuroendocrine cells , mostly serotonin-producing cells , from intestinal crypts for its transport through the intestinal mucosa . In addition , we checked whether BoNT/A can be transcytosed through the mouse neuroendocrine intestinal cell line STC-1 [45] . As shown in Figure 5A , the passage of biologically active BoNT/A was monitored from apical to basolateral side of STC-1 cell monolayers . The transcytotic passage of BoNT/A through STC-1 cells was not statistically different from that through Caco-2 enterocytes , but it was lower than through the mouse intestinal crypt cell line m-ICcl2 as shown in Figure 5A ( p<0 . 05 ) and [30] . It is noteworthy that the passage yield through STC-1 cells was more difficult to assess ( high standard deviation values ) , since these cells do not form tight junctions as epithelial cells . Since epithelial cells such as Caco-2 and HT29 cells express at the cell surface and secrete from apical and basolateral sides several types of proteases [46]–[48] , we tested whether these proteases degrade BoNT/A , thus impairing or decreasing the transcytosis level . As shown in Figure 5A , a 2 to 4 fold higher level of BoNT/A transcytosis was observed in Caco-2 and m-ICcl2 cells incubated with a cocktail of anti-proteases . However , even in the presence of anti-proteases , BoNT/A transport was more efficient ( 20-fold ) in m-ICcl2 than in Caco-2 cell monolayers . Thus , the decreased BoNT/A transcytosis through Caco-2 cells is not likely due to a higher protease degradation of BoNT/A before and/or after transport . STC-1 cells possibly also secrete proteases , and a higher level of BoNT/A transcytosis through this cell type might be expected . However , since STC-1 cells do not form tightly organized cell monolayers , the results of experiments with anti-proteases were inconclusive . As we have previously found that m-ICcl2 express SV2C or an imunologically related protein as a putative BoNT/A receptor [30] , we investigated the presence of SV2 proteins and chromogranin A in STC-1 and intestinal cells by Western blotting ( Figure 5B ) . Chromogranin A was strongly expressed in STC-1 confirming its neuroendocrine type , and to a lower extent in m-ICcl2 and even less in Caco-2 cells , but not in Vero cells used as negative control . Antibodies against SV2A and SV2B showed no protein related to the expected size of SV2 ( 82 kDa ) in intestinal and STC-1 cells . In contrast , specific SV2C antibodies directed against the N-terminal part or the intravesicular loop L4 , which is assumed to be the receptor binding domain of BoNT/A [11] , [14] , recognized a protein with the expected size in the intestinal cells and STC-1 , but not in Vero cells ( Figure 5B ) . The specificity of SV2 antibodies is shown in Figure S1 . Next , we investigated whether SV2C was involved in the entry of HcA into m-ICcl2 and STC-1 cells by a competition assay between fluorescent HcA and SV2C/L4 . Cells grown on glass cover slides were exposed to HcA-Cy3 or a combination of fluorescent Hc with a 10-fold higher molar concentration of SV2C/L4 for 10 min at 37°C and then processed for microscopic observation . As shown in Figure 5C , HcA entered into m-ICcl2 and STC-1 cells , and SV2C-L4 greatly impaired the entry of HcA into both cell types by 97 and 94% , respectively , as determined by counting the number of fluorescent HcA patches per µm2 of cell area ( Figure 5D ) , supporting the view that SV2C participates in the entry mechanism of HcA into cells . After 30–60 min incubation of fluorescent HcA into an intestinal loop lumen , HcA was detected in the intestinal submucosa , where it stained certain neuronal structures ( Figure 6A and data not shown ) . Antibodies against neurofilaments allowed visualizing a complex and abundant network of neuronal cell bodies and neuronal extensions in the submucosal plexus immediately underneath intestinal villi and crypts . Only some of these neuronal structures were stained with HcA ( Figure 6A ) . BoNTs are well known to interact with cholinergic neurons and to specifically block spontaneous and evoked quantal acetylcholine release [4] , [49] . However , it has been shown that BoNT/A and BoNT/E are also able to enter other neuronal cell types such as glutamatergic and gamma-aminobutyric acid ( GABA ) -ergic neurons , as well as astrocytes [50] , [51] . Since ENS contains a large variety of neuronal cell types , we investigated the most representative types as putative targets of HcA in the mouse small intestine . Cholinergic neurons were monitored by immunostaining with antibodies for choline acetyltransferase ( ChAT ) . ChAT-immunoreactive neurons are abundant ( about 55% ) in the submucosal plexus , where they are involved in various gut functions including the control of evoked anion secretion by the jejunal and ileal epithelium , and they also interact with Peyer's patch follicles [52] , [53] . Most of ChAT-immunoreactive nerve terminals from the intestinal submucosa were labeled with fluorescent HcA ( Figure 6B and G ) . Vasoactive intestinal peptide ( VIP ) -immunoreactive neurons are known to be located in jejunum and ileum submucosal plexus , as well as in other organs . VIP modulates several basic functions including blood flow , smooth muscle relaxation , and exocrine secretion [54] . VIP-immunoreactive neurons are estimated to represent about 45% of neurons from the submucosal plexus [53] , [55] . Numerous cells were immunostained with anti-VIP antibodies in mouse intestinal submucosa , but a colocalization between VIP immunoreactivity and HcA staining was only observed in a few of them ( less than 3% ) ( Figure 6C and G ) . Note that some of the filaments and cell bodies stained with anti-VIP antibodies did not colocalize with neurofilament staining , and may probably represent glial structures . Glutamatergic neurons , which are the major neurons from the central nervous system involved in excitatory responses , are also present in ENS . Glutamate receptors have been detected in enteric neurons and glutamatergic enteric neurons where they have been found to mediate excitatory synaptic transmission , whereas only a subset of them are involved in sensory responses [56] . In mouse intestinal submucosa , only a low number of glutamatergic neurons ( less than 1% ) , as evidenced by anti-glutamate antibodies , colocalized with HcA ( Figure 6D and G ) . Serotonin is also an important neurotransmitter in ENS , where it is involved in the control of motility , secretion and sensory functions . Serotonin is produced by 2 to 20% of all enteric neurons [33] . In our analysis only a few neuronal cell extensions were stained with anti-serotonin antibodies in the submucosal plexus , and some of them were also labeled with HcA , as shown in Figure 6E . Glial cells were neither immunostained with anti-neurofilament antibodies nor with HcA , but exhibited a clear immunolabelling with GFAP antibodies ( data not shown ) . These results support the view that HcA binding is mostly specific of nerve endings in the intestinal submucosa . Also , we investigated the intestine musculosa , which is the predicted target tissue of BoNT/A for its inhibitory activity on intestinal motility . No significant HcA staining was observed in the musculosa 30 or 60 min after incubation with the fluorescent probe in the intestinal loop , and only a few cell bodies or extensions were stained after a longer incubation period ( 90–120 min ) in our experimental conditions . Cell extensions stained with HcA colocalized with neurofilament staining ( Figure 7 ) . However , only some of the nerve endings were labeled with HcA . Almost all cell bodies labeled with HcA exhibited ChAT-immunoreactivity . Similar results of colocalization with ChAT-immunoreactive neurons were obtained using full length BoNT/A injected into the intestinal loop lumen and detected with anti-HcA antibodies ( Figure S2 ) . This is consistent with the fact that a large majority of neurons in the myenteric plexus are immunoreactive for ChAT , albeit many of them produce additional neurotransmitters [55] . No significant colocalization was observed between HcA staining and glutamate- or serotonin-producing neurons in the myenteric plexus ( data not shown ) . The protein receptor of BoNT/A on neuronal cells has been identified as SV2 . Among the three SV2 isoforms , SV2C shows the highest affinity to BoNT/A in vitro , whereas BoNT/A binds to SV2A and SV2B with a lower strength [11] , [14] , [57] . SV2A is present in almost all neurons whatever their neurotransmitter type is , while SV2B shows a more restricted distribution . SV2C is reported to be present in a subset of neurons [58] . However , SV2 proteins are expressed not only in neuronal cells , but also in other cell types such as neuroendocrine cells , in particular in the gastrointestinal tract [59] . We investigated the distribution of SV2 proteins in mouse intestinal mucosa with our in toto tissue model . Numerous cell extensions in the submucosal plexus were stained with anti-SV2C antibodies ( Figure 6F ) , whereas no specific staining was observed with anti-SV2B antibodies , and only a diffuse staining in certain crypt cells and cell extensions in the submucosa was evidenced with anti-SV2A antibodies ( data not shown ) . The mouse intestinal crypt cell line m-ICcl2 was found to express SV2C at a higher level than in enterocyte-type cell lines ( Figure 5B and [30] ) . However , in our ex vivo conditions intestinal crypt cells were not , or only weakly immunoreactive with anti SV2C antibodies . This does not preclude that a subset of crypt cells express a significant level of SV2C that was not detected in our conditions . Most of SV2C-immunoreative filaments in the submucosa , were also labeled with anti-neurofilament antibodies , indicating that SV2C is widely distributed in neuronal cells and neuronal extensions from the small intestine . However , HcA signal was observed in only some neuronal endings bearing SV2C , but not along the neuronal extensions stained with anti-SV2C antibodies ( Figure 6F ) . Only a low proportion of SV2C-immunoreactive cells were labeled with HcA ( Figure 6G ) . It is worth noting that a distinct network of thin filaments around small blood vessels was stained with anti-SV2C , and only weakly with anti-neurofilament antibodies . No specific binding of HcA was observed on these structures ( Figure S3A and B ) . In the intestinal submucosa and musculosa , large cells with wide cellular bodies and short extensions were stained with anti-SV2C antibodies , but not with anti-neurofilament antibodies . Co-labelling with anti-GFAP indicated that they were glial cells ( Figure S3C ) . However , HcA , as reported above , was not found to label glial cells .
The standard scheme of botulism intoxication includes BoNT transit through the digestive tract , passage across the intestinal epithelial barrier and subsequent delivery to the blood circulation and dissemination to the target motor nerve endings . Indeed , BoNT has been found to be absorbed preferentially from the upper small intestine , but also from the stomach in experimental rodents , and to be delivered in the blood and lymph circulation [21] , [23] , [24] , [60] . The aim of this study was to identify the entry pathway and target cells of BoNT/A in the mouse intestinal wall . For that , we checked the activity of BoNT/A in mouse intestine following intralumenal administration and we used the fluorescent Hc domain , which is the functional binding domain of BoNT , to monitor the trafficking of BoNT/A . First , we tested whether BoNT/A injected into intestinal lumen was able to enter intestinal mucosa and to induce local effects on the intestine . In vitro studies have already shown that BoNT/A reduces cholinergic transmission in gastrointestinal smooth muscles as well as pylori and Oddi sphincter muscles by inhibiting ACh release [40] , [41] , [61]–[63] . In our experimental conditions , BoNT/A passed through the epithelial intestinal barrier and by diffusion through the extracellular space , locally targeted intestinal neurons independently of the blood circulation . BoNT/A reduced the frequency of spontaneous contractions of small intestine and inhibited the contractile response evoked by electric field stimulation within 2–4 h after intralumenal administration . Since carbachol was still able to stimulate muscle contraction after BoNT/A treatment , this supports a toxin-dependent inhibition of ACh release . To the best of our knowledge , this is the first report demonstrating a local intestinal effect of botulism after intralumenal administration of purified BoNT/A . Constipation is often ( about 70% ) , but not always , associated with food-borne botulism and its participation to the progression of the disease is unknown [64] , [65] . However , constipation is a major and early symptom of botulism resulting from an intestinal colonization by C . botulinum such as during infant botulism [66] , [67] . This digestive symptom might result from a local effect of BoNT after crossing the intestinal barrier instead of toxin dissemination through the general circulation . BoNT locally synthesized in the intestine is possibly absorbed in a higher local concentration able to induce an intestinal muscle paralysis , than toxin orally ingested which disseminates more broadly through the digestive tract . Interestingly , BoNT/A-dependent inhibition of evoked smooth muscle contraction was significantly prevented by preincubation with the intravesicular domain of SV2C . This indicates that a protein related to SV2C/L4 might impair the BoNT/A passage through the intestinal barrier and/or toxin uptake by the underlying nerve terminals . We have previously found that SV2C , or a related protein , is part of BoNT/A receptor mediating toxin transcytosis through cultured intestinal cell monolayers [68] . In addition , SV2C/L4 , which is expressed by intestinal and neuroendocrine STC-1 cells ( Figure 5B ) , significantly prevented HcA entry into the intestinal crypt m-ICcl2 and STC-1 cells ( Figure 5C , D ) . Taken together , these data suggest that SV2C , or a related protein , facilitates BoNT/A uptake through the mouse intestinal barrier . BoNT/A trafficking in mouse intestinal wall was investigated with fluorescent HcA as already used [38] . First , we investigated the potential binding sites of HcA by using mouse small intestine cryosections overlaid with fluorescent probes . Thereby , HcA was found to bind only to certain cell types of the intestinal mucosa , preferentially from intestinal crypts , whereas enterocytes showed only a weak staining of the brush border . In contrast , it was reported that the botulinum complexes type A or type C ( BoNT and associated non-toxin proteins , ANTPs ) , strongly bind to the epithelia cell surface and goblet cells of guinea pig small intestine . Moreover , this binding was shown to be mediated by the hemagglutinins HA1 and HA3b , which interact with distinct gangliosides and/or glycoproteins [24] , [29] , [60] , [69] from those recognized by the neurotoxin alone [11] , [14] . This certainly accounts for the differential binding between progenitor toxin and BoNT to intestinal epithelial cells . The functions of ANTPs are still controversial . Ancillary proteins probably participate in BoNT protection from degradation inside the digestive tract in a dose-dependent manner , particularly in the stomach [21] . In addition , it is assumed that HAs are involved in the internalization of progenitor type C toxin into intestinal cells and subsequently in the small intestine [24] , [29] , [70] , and that they disrupt the intestinal epithelial barrier facilitating toxin absorption via the paracellular route [71]–[74] . However , since BoNT/A absorption from mouse stomach or small intestine was found to occur independently of the presence of ANTPs , HAs might have not an essential role but an additional facilitating effect on BoNT/A passage through the intestinal barrier . At pH around neutrality , as found in the intestine , BoNT dissociates from ANTPs [75] , and thus might be absorbed through the epithelial intestinal barrier independently of ancillary proteins . In our model of ligated mouse intestinal loops and recording of muscle contraction , BoNT/A entered the intestinal mucosa in the absence of ANTPs supporting that ANTPs are not absolutely required in the intestinal uptake of BoNT/A . The main finding of the ex vivo intestinal loop experiments was that fluorescent HcA preferentially recognized certain crypt cells . Intestinal crypts contain stem cells which proliferate and differentiate in four main cell types including enterocytes , mucus , endocrine , and Paneth cells , which are the most abundant cell type at the bottom of intestinal crypts [76] . HcA labeling was observed in chromogranin- and serotonin-immunoreactive cells , but not in Paneth cells . It is noteworthy that HcA was detected in only some , but not all , chromogranin-immunreactive cells . This strongly suggests that HcA specifically binds to a subset of neuroendocrine cells from intestinal crypts , which possibly represent the preferential site of toxin uptake . BoNT/A has been found to pass through intestinal cell monolayers grown on filter by a transcytotic mechanism [26] , [27] . We have previously reported that the passage rate of biologically active BoNT/A was higher ( about 10-fold more ) through the mouse crypt cell line m-ICcl2 than through the colonic enterocyte lines Caco-2 and T84 [30] . Here , we found that the neuroendocrine intestinal cell line STC-1 also permits a transcytotic passage of BoNT/A not statistically different from that through Caco-2 cells and to a lower extent than in m-ICcl2 cells . This further supports that a subset of intestinal crypt cells constitute a preferential entry site of BoNT/A within the intestinal mucosa . However , STC-1 cells , which derive from an intestinal endocrine tumor , secrete several hormones and neuropeptides , but not serotonin [45] . Thereby , STC1 cells are not the most representative cells of the serotonin-reactive cells co-labeled with HcA , which have been identified in intestinal loop tests . To the best of our knowledge , no intestinal serotonin-secreting endocrine cell line is known at present . m-ICcl2 cells , which have been isolated from fetal mouse intestine , express various markers of intestinal cells and thus are a totipotent intestinal crypt cell line in a less differentiated state than the tumor cell line STC-1 [77] . m-ICcl2 cells show a neuroendocrine-like phenotype since they synthesize chromogranin A albeit to a lower extent than STC-1 , but more significantly than Caco-2 cells ( Figure 5B ) . In addition , m-ICcl2 cells retain transport pathways similar to those of native crypt cells . Indeed , in contrast to colonocytes such as Caco-2 and HT29 cells , m-ICcl2 cells express polymeric Ig receptors ( pIgRs ) , which mediate the transepithelial transport of polymeric IgA and IgM [77] . It is noteworthy that only a low number of cells ( 1 to 2 ) from each intestinal crypt were labeled by HcA and are likely susceptible to transport the toxin through the intestinal barrier . This might account for the low rate of BoNT absorption from the digestive tract to the general circulation . Indeed , the transport rate of BoNT/B from rat duodenum to the lymphatic circulation has been estimated from 0 . 01 to 0 . 1% [22] . In addition , chromogranin-immunoreactive cells are distributed throughout the gastrointestinal tract , but are more predominant in the pylorus and duodenum [78] . This might support the observation that the upper small intestine is the preferential site of BoNT absorption [22]–[25] . Using the intestinal loop model , we found that neuronal cells in the submucosa and more lately in the musculosa were stained with HcA . However , not all the neuronal cells were recognized by HcA as tested by co-labeling with antibodies against neurofilament , indicating that HcA targeted specific neurons in the intestine . About half of neurons in the submucosa are ChAT-immunoreactive , and most of them produce other neurotransmitter types [53] , [55] . As previously found [79] , HcA stained most of the cholinergic neurons ( more than 90% ) of the submucosa . However , in contrast to previous findings [79] , BoNT/A recognized other neuronal cells in the intestinal mucosa , albeit to a lower extent . Indeed , a low proportion of VIP-immunoreactive neurons , which are also largely present in the submucosa ( about 45% ) [53] as well as a low proportion of glutamatergic and serotoninergic neurons were targeted by HcA . It is noteworthy that the BoNT/A receptor , assigned to ganglioside GD1b/GT1b and SV2 protein [6] , [7] , [10] , [14] , is part of a synaptic vesicle complex , which contains additional membrane proteins including vesicular glutamate transporters ( cGLY-1 and vGLUT-2 ) [80] , supporting that BoNT/A may target glutamatergic neurons . BoNTs are known to block the release of ACh , but also that of other neurotransmitters and neuropeptides the effects of which are still poorly known [61] , [81]–[86] . The BoNT/A-dependent inhibitory activity on non-ChAT neurons in the intestine may contribute to the local effects of the toxin . In the myenteric plexus , HcA stained essentially cholinergic neurons , which are the predominant neuronal cells in this tissue ( about 80% ) [53] . But , only a part of ChAT-immunoreactive neurons were labeled with HcA . This might be due either to a low number of Hc molecules reaching the myenteric plexus , or to the fact that HcA recognized only a subset of ChAT-reactive neurons , which remains to be determined . Targeting of cholinergic neurons of the myenteric plexus by HcA is consistent with constipation which frequently occurs during botulism , and with the BoNT/A-dependent inhibition of evoked intestinal contractions in the ex vivo experiments . SV2C , the protein receptor with the highest affinity to BoNT/A [10] , [14] , was the main isoform detected in the mouse intestine , whereas no , or a weak staining was obtained with anti-SV2B or -SV2A antibodies . Immunostaining with anti-SV2C antibodies was observed in cells with thin arborescent extensions , which were also labeled with anti-neurofilament antibodies . However , SV2C staining did not match all the cells labeled with anti-neurofilament antibodies , indicating a SV2C distribution in a broader range of cell types than neuronal cells in the intestinal mucosa . HcA did not bind all along the neuronal extensions or cell structures which are recognized by anti-SV2C antibodies , but only in discrete zones . One possibility is that HcA staining was too weak to be visualized . Alternatively , the BoNT/A high-affinity receptors , which consist of a ganglioside part and a protein part such as SV2C [11] , [14]–[17] , are distributed in only some restricted areas on the neuronal cell surface . This is supported by the observed colocalization between HcA and SV2C , which is restricted to only some cell areas ( Figure 6F ) . Moreover , SV2 , which is an integral protein of the synaptic vesicle membrane , has to be exposed to the extracellular compartment to be accessible to BoNT/A , as during synaptic vesicle fusion with the presynaptic membrane . Another type of thin cell extensions stained with anti-SV2C antibodies , formed a dense network all around microvessels ( Figure S2 ) . These neuronal extensions , which were only faintly stained with anti-neurofilament antibodies , are probably involved in vessel innervation [87] , but were not observed to be targeted by HcA under our experimental conditions . In addition , glial cells identified by GFAP immunofluorescence were also stained with anti-SV2C antibodies but not with HcA ( Figure S3C ) . Thereby , SV2C seems to have a broad distribution and to form BoNT/A high-affinity receptor only when associated with gangliosides in certain neuronal membrane domains . When passed across the intestinal barrier , BoNT is delivered to the connective tissue of the submucosa , where the toxin can have access to microvessel endothelial cells . BoNT passage into the blood and lymph circulation , supposes a transcytotic transport of the toxin through the endothelial cell barrier of vessels . In ligated intestinal loop experiments , no staining of endothelial cells forming small vessels or microvessel structure was observed in the submucosa and musculosa . Under natural conditions , BoNT binding to endothelial cells is possibly low and/or transient permitting the passage of the toxin into the vessel lumen . In our experimental conditions , the blood circulation was interrupted , thus preventing HcA passage into vessels . But , the diffusion through the matrix of connective tissue can mediate the toxin trafficking until the target cells in the submucosa and musculosa . Indeed , HcA staining of neuronal cells in the submucosa after 30–60 min incubation of the probe into the intestinal lumen and later ( 60–120 min ) in the musculosa , as well as the decrease in fluorescence intensity of HcA from the mucosa to musculosa likely reflects the progressive diffusion of the recombinant protein in the tissue extracellular matrix . It cannot be ruled out that a shorter time period may be required in natural conditions for the passage of low amounts of HcA across the intestinal epithelial cell barrier and subsequent targeting to neuronal cells . Alternatively , another mode of HcA transport might be involved . Indeed , like TeNT , which undergoes a retrograde transport into motor neurons , a transcytotic mechanism might also be supported by neuronal cells from the ENS to disseminate BoNT/A to target neuronal cells both locally and even at distance from the intestine . One possibility is that BoNT/A uses non-cholinergic neurons such as those identified with fluorescent HcA to be transported to other target neurons . Indeed , neurons from ENS are highly interconnected between them and with neurons of the central nervous system [31] , [32] . Moreover , it has been recently found that BoNT/A can use a retrograde transport and transcytosis to migrate from peripheral neurons to central circuits [88] , [89] . This opens novel ways of investigation to unravel the mechanism of BoNT transport from the intestinal barrier to target motoneurons , which is still poorly known . In conclusion , in this work we have shown that BoNT/A enters the intestinal mucosa , possibly via an uptake process involving SV2C or a related protein , and impairs spontaneous and electrically-evoked muscle contractions . Following intralumenal administration , HcA preferentially recognized a subset of neuroendocrine crypt cells in the mouse upper small intestine , which likely represents the main pathway for toxin entry and passage across the epithelial cell barrier . HcA diffused progressively within the submucosa , and later reached the musculosa , where it targeted specifically some cholinergic neurons and to a lower extent glutamatergic and serotoninergic neurons . The implication of these non-cholinergic neurons in the botulinum intoxication process remains to be determined .
All experiments were performed in accordance with French and European Community guidelines for laboratory animal handling . The protocols of experiments were approved by the Pasteur Institute ( Agreement of laboratory animal use n° 75-279 ) . The primary antibodies used recognized neurofilament ( Sigma; mouse , 1∶500 dilution ) , Glial Fibrillary Acidic Protein ( GFAP ) ( Sigma; rabbit , 1∶200 dilution ) , Choline Acetyltransferase ( ChAT ) ( Chemicon International , Temecula , CA , USA; goat , 1∶100 dilution ) , VIP ( Abcam; rabbit , diluted 1∶200 ) , Serotonin ( Sigma; rabbit , diluted 1∶100; and Abcam ; goat , 1/200 dilution ) , Glutamate ( Chemicon; rabbit , diluted 1∶200 ) , SV2A , SV2B ( Synaptic systems; rabbit , diluted 1∶200 ) , SV2C ( Santa Cruz; goat , diluted 1∶200 ) , chromogranin A ( Abcam; rabbit , diluted 1∶200 ) . For Western blotting rabbit anti-SV2C ( Abcam 33892 ) and rabbit anti-SV2C intravesicular domain were used . SV2C/L4 domain ( amino acid 454 to 579 ) was produced as a GST-fusion protein [30] and was used to immunize rabbits . Specific antibodies against SV2C/L4 were purified by immunoaffinity with SV2C/L4 produced as a histidine-tagged protein in pET28 vector ( Novagen ) and immobilized on cyanogen bromide activated Sepharose-4B ( GE Healthcare ) . The secondary antibodies used were: Cy5 coupled donkey anti-mouse IgG ( US Biological ) , Cy3 coupled donkey anti-goat IgG ( US Biological ) , Alexa488 donkey coupled anti-goat IgG ( Invitrogen ) , Alexa488 coupled anti-rabbit IgG ( Invitrogen ) . 4′-6′-Diamidino-2-phenylindole ( DAPI , SIGMA ) was incubated with secondary antibodies to stain cell nuclei . Adult male IOPS mice ( 20–25 g body weight ) purchased from Charles River Laboratories ( L'Arbresle , France ) were anesthetized by inhalation with Isoflurane ( AErrane , Baxter S . A . , Lessines , Belgium ) , and euthanized by dislocation of the cervical vertebrae , as specified by the CNRS Animal Ethics User's Committee . BoNT/A was produced and purified as previously described [90] . Recombinant His-tag Hc fragment of BoNT/A was produced and purified from pET28b vector containing DNA encoding for HcA cloned into BamHI and SalI sites , as previously described [91] . A recombinant derivative plasmid encoding HcA with a N-terminal tag containing 4 cystein residues , as described for TeNT-Hc [92] , was performed inserting the two complementary oligonucleotides P1132 ( 5′-TATGGCAGAGGCAGCAGCACGAGAGGCTTGTTGTCGAGAGTGTTGTGCACGAG-3′ ) and P1131 ( 5′-GATCCCTCGTGCACAACACTCTCGACAACAAGCCTCTCGTGCTGC TGCCTCTGCCA-3′ ) into NdeI and BamHI sites of pET28b vector . HcA with 4 Cys tag was then labeled with either maleimide Alexa 488 reagent ( InVitrogen , Cergy-Pontoise , France ) or maleimide Cy3 ( Amersham , Les Ulis , France ) according to the manufacturer's recommendations . cDNA coding the intraluminal SV2C fragment L4 ( amino acid 454 to 579 ) was PCR-amplified as previously described [68] and cloned into pGEX-2T . The fusion protein , GST-SV2C/L4 , was produced in E . coli BL21 and purified on glutathione agarose matrix ( SIGMA ) equilibrated with 50 mM Tris-HCl , 300 mM NaCl , pH 7 . 5 and eluted with 40 mM reduced glutathione in the same buffer . Purified fusion protein was then dialyzed against phosphate balanced solution ( PBS ) before use . The mouse ileum was removed and placed in an oxygenated standard Krebs-Ringer solution composed of 154 mM NaCl , 5 mM KCl , 2 mM CaCl2 , 1 mM MgCl2 , 11 mM glucose and 5 mM HEPES ( pH 7 . 4 ) . An ileum segment of about 2 cm length was extensively washed to flush out the intestinal content , and mounted in a silicone-lined chamber ( 4 ml volume ) , bathed in the standard medium . For tension measurements , one of the ends of the ileum segment was tied with silk thread , via an adjustable stainless-steel hook , to an FT03 isometric transducer ( Grass Instruments , AstroMed , W . Warwick , RI , USA ) , and the other end was ligatured and pinned onto the silicone-coated bath via stainless-steel micro pins . BoNT/A ( 5 µg/ml ) diluted in the Krebs-Ringer solution was injected into the ileum segment , between the two ligatures , with a micro syringe . Electric field stimulation was performed with an electrode assembly , placed along and on both sides of the length of the ileum segment , and connected to S-48 Grass stimulator . Preparations were usually stimulated with pulses of 0 . 15 ms duration at 30 Hz for 15 or 25 s , every hour . The resting tension was adjusted for each preparation investigated with a mobile micrometer stage ( to allow incremental adjustments of ileum length ) in order to obtain maximal spontaneous or evoked contractile responses , and was monitored during the whole duration of the experiment . Carbachol ( Sigma , 20 µM ) was added to the bath solution of ileal segment exposed to BoNT/A for 3–4 hours to check that ACh receptors were not affected by intoxication . Tension signals from the isometric transducer were amplified , collected , and digitized with the aid of a computer equipped with a DT2821 analogue to digital interface board ( Data Translation , Marlboro , USA ) and expressed in g or N . Data acquisition and analysis were performed with a program kindly provided by Dr . John Dempster ( University of Strathclyde , Scotland ) . All experiments were performed at 22±0 . 5°C . Segments of mouse ileum ( 2 cm length ) were removed and immersed in oxygenated Krebs-Ringer solution . After extensive wash to flush out the luminal contents , segments were ligated on both extremities and injected with a micro syringe with fluorescent HcA ( 0 . 5 µg ) diluted in 150 µl of Krebs-Ringer solution ( 5 . 5 10−8 M ) . The injection site was isolated with another ligature and ileal segments were incubated in oxygenated Krebs-Ringer solution for different times ( 30 to 120 min ) before washing . Tissues were cut along the mesenteric border and pinned out with the mucosal surface facing down in a silicone-lined Petri dish . After fixation with 4% paraformaldehyde ( PFA; 1 h , 22°C ) , tissues were washed in phosphate buffer saline ( PBS ) and autofluorescence quenched with 50 mM NH4Cl ( 30 min , 22°C ) . Specimens were dissected by carefully separating the mucosa and submucosa from the muscle layers under a microscope . Whole-mount preparations of the myenteric and submucosal plexuses of the ileum were permeabilized and blocked with PBS/bovine serum albumin ( BSA ) 2%/Donkey Serum 10%/Triton 2% for 1 h at room temperature ( RT ) and incubated with primary antibody for 16 h at RT . After 3 washes of 10 min in PBS , tissue sections were incubated for 4 h at RT with the appropriate secondary antibodies diluted 1∶500 in PBS/BSA 2%/Triton 2% . After being washed in PBS ( 3×10 min ) , tissues were mounted in Mowiol ( Polysciences Europe , Eppelheim , Germany ) and analyzed using a Zeiss confocal laser scanning microscope and a ×63 oil immersion objective ( N . A . 1 . 4 ) . For quantitative analysis of co-localization of HcA with other markers , images of 512×512 pixels were taken from series of optical sections of 0 . 8 µm thickness . The colocalisation was analyzed with the Zeiss LSM Image Browser software , on 100 HcA-immunoreactive varicosities for each marker using the merged image from the different experiments . For analysis of BoNT/A internalization into intestinal cells , total fluorescence intensity was quantified in serotonin-positive cells in at least 40 optical fields taken from three independent experiments . Data are presented as the mean ± SD . Segments of the ileum were removed and immersed in oxygenated Krebs-Ringer solution . After extensive wash to flush out the luminal contents , the tissues were embedded in OCT embedding medium ( Tissue-Tek , Miles Laboratories , Naperille , IL , USA ) , and stored at −80°C . Sections ( 5 µm ) were cut with a cryostat-microtome and thaw , mounted onto SuperFrost glass slides ( Fisher Scientific , Illkirch , France ) . For HcA binding experiments , sections were removed from the freezer and incubated for 30 min with 10 µg/ml Alexa- or Cy3-labeled HcA in PBS/BSA ( 1% ) . After 3 washes in PBS , tissue sections were fixed with 4% PFA for 20 min at RT ( 22°C ) , rinsed in PBS and autofluorescence was quenched with 50 mM NH4Cl ( 15 min , 22°C ) . After permeabilization with 1% triton X-100 in PBS for 10 min , tissues were stained with TRITC-phalloidin ( Sigma , 0 . 4 µg/ml ) , TRITC-Urex europaeus agglutinin type 1 ( UEA1 , 1 µg/ml ) , immunostained with anti-neurofilament 200 ( Sigma; mouse , diluted 1∶500 ) and Cy3 coupled goat anti-mouse IgG ( Sigma , diluted 1∶250 ) . After 3 washes in PBS , sections were mounted in Mowiol and observed with a Zeiss confocal laser scanning microscope and a ×25 objective . For quantitative analysis of co-localization of HcA with markers , images of 512×512 pixels were taken from serial optical sections of 1 µm thickness . Data are presented as the mean ± SD . The mouse neuroendocrine intestinal cell line STC-1 [45] , m-ICcl2 [77] and Caco-2 ( human colon ) cells were grown on filter ( Transwell , Corning ) in Dulbecco's modified Eagle's medium ( DMEM , Invitrogen ) supplemented with 10% fetal calf serum ( FCS , Invitrogen ) until confluence . Integrity of tight junctions was confirmed by ZO-1 labeling and non-permeability to FITC-labeled dextran ( 4300Da , Sigma-Aldrich ) ( data not shown ) . BoNT/A as prepared previously [30] was added to the apical chamber . After 60 min incubation at 37°C , medium from the basal chamber was collected and BoNT/A was assayed by the mouse bioassay as previously described [30] . In the experiments with anti-proteases , culture medium in the apical and basolateral chambers was replaced with Dulbecco's medium containing 1% bovine serum albumin ( BSA ) and 1× anti-protease coktail without EDTA ( Calbiochem ) 30 min prior addition of BoNT/A into the apical compartment . BoNT/A transcytosis was monitored in basolateral medium samples after 60 and 120 min incubation at 37°C by a biological assay as previously described [30] . m-ICcl2 and STC-1 cells grown on glass coverslips ( coated with poly-ornithine , Invritrogen , for STC-1 cells ) were exposed to HcA-Cy3 ( 2 . 5 µg/ml ) , alone or in combination with a 10-fold more molar concentration of SV2C/L4-GST for 10 min at 37°C . Cells were washed twice with PBS , fixed with 4% PFA and mounted in Mowiol . The number of HcA fluorescent patches per µm2 was evaluated in m-ICcl2 and STC-1 cells by counting in images of 512×512 pixels ( 10 optical fields ) , taken from serial optical sections of 1 µm thickness from 3 experiments . Data was evaluated using ImageJ software ( http://rsbweb . nih . gov/ij/ ) . Cells were lysed with boiling Laemmli buffer and lysates were homogenized by passages through a 26-gauge needle . An equal amount of protein ( 100 µg ) from each boiled sample was loaded on a SDS-polyacrylamide ( 10% ) gel . Samples separated by SDS-PAGE were transferred to nitrocellulose membrane ( Amersham ) and blocked in phosphate saline buffer containing 5% dried milk , washed with Tris-buffered saline containing 0 . 1% Tween20 ( TBST ) , incubated with primary antibodies diluted in TBST overnight at room temperature , then washed 5 times for 10 min , and incubated for 1 h at room temperature with HRP-protein A diluted in TBST . After membrane washing in TBST , the specific signal was detected by enhanced chemiluminescence . Specificity of the anti-SV2A antibodies was tested using lysate from SV2A-transfected cells ( Santa Cruz ) . The lysate ( 10 µg ) was run on a 10% SDS-PAGE , transferred on nitrocellulose , and blotted with anti-SV2A , anti-SV2B , anti-SV2C , and anti-SV2C-L4 antibodies . The specificity of anti-SV2B and anti-SV2C antibodies was tested on rat brain extract by competition with peptides , which had been used for immunization . Rat brain extract ( 5 µg , Santa Cruz ) was run on a 10% SDS-PAGE , transferred on nitrocellulose , and blotted with anti-SV2B or anti-SV2C antibodies preincubated or not with the indicated SV2B or SV2C peptides ( 10 µg , Synaptic System ) for 30 min at room temperature . Values in the text are expressed as the mean ± SD , unless otherwise indicated . Differences between means were tested using Student's t-test , and p-values<0 . 05 were taken to indicate significance .
|
Botulism is a severe and often fatal disease in man and animals characterized by flaccid paralysis . Clostridium botulinum produces a potent neurotoxin ( botulinum neurotoxin ) responsible for all the symptoms of botulism . Botulism is most often acquired by ingesting preformed botulinum neurotoxin in contaminated food or after intestinal colonization by C . botulinum under certain circumstances , such as in infant botulism , and toxin production in the intestine . The first step of the disease consists in the passage of the botulinum neurotoxin through the intestinal barrier , which is still poorly understood . We investigated the trafficking of the botulinum neurotoxin in a mouse intestinal loop model , using fluorescent HcA ( half C-terminal domain of the heavy chain ) . We observed that HcA preferentially recognizes neuroendocrine intestinal crypt cells , which likely represent the entry site of the toxin through the intestinal barrier , then targets specific neurons , mainly cholinergic neurons , in the submucosa , and later ( 90–120 min ) in the musculosa leading to local paralytic effects such as inhibition of intestinal peristaltism . These results represent an important advance in the understanding of the initial steps of botulism intoxication and can be the basis for the development of new specific countermeasures against botulism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"biology",
"microbiology",
"molecular",
"cell",
"biology",
"toxicology"
] |
2012
|
Preferential Entry of Botulinum Neurotoxin A Hc Domain through Intestinal Crypt Cells and Targeting to Cholinergic Neurons of the Mouse Intestine
|
To understand the brain mechanisms of olfaction we must understand the rules that govern the link between odorant structure and odorant perception . Natural odors are in fact mixtures made of many molecules , and there is currently no method to look at the molecular structure of such odorant-mixtures and predict their smell . In three separate experiments , we asked 139 subjects to rate the pairwise perceptual similarity of 64 odorant-mixtures ranging in size from 4 to 43 mono-molecular components . We then tested alternative models to link odorant-mixture structure to odorant-mixture perceptual similarity . Whereas a model that considered each mono-molecular component of a mixture separately provided a poor prediction of mixture similarity , a model that represented the mixture as a single structural vector provided consistent correlations between predicted and actual perceptual similarity ( r≥0 . 49 , p<0 . 001 ) . An optimized version of this model yielded a correlation of r = 0 . 85 ( p<0 . 001 ) between predicted and actual mixture similarity . In other words , we developed an algorithm that can look at the molecular structure of two novel odorant-mixtures , and predict their ensuing perceptual similarity . That this goal was attained using a model that considers the mixtures as a single vector is consistent with a synthetic rather than analytical brain processing mechanism in olfaction .
One hundred years ago , Alexander Graham Bell asked: “Can you measure the difference between one kind of smell and another . It is very obvious that we have very many different kinds of smells , all the way from the odor of violets and roses up to asafetida . But until you can measure their likenesses and differences you can have no science of odor . ” [1] . Although the challenge posed by Bell has been widely recognized in olfaction research [2] , [3] , the field has yet to gravitate to an agreed upon system for odor measurement . Early investigations into quantification of odor revolved around an effort to identify odor primaries , similar to the notion of primary colors in vision [4] . A major tool in this effort was the quantification of specific anosmias [5] . Although specific anosmia remains a powerful tool for linking odor perception to olfactory neurobiology [6] , [7] , this path did not generate a general method to quantify olfactory perception . A conceptually similar approach was an effort to identify specific odorant molecular features that drove specific olfactory perceptual notes . This approach , referred to as structure-odor-relationships or SOR [8] , identified many specific rules linking structure to odor ( e . g . , what structure provides a “woody” note ) , but failed to produce a general framework for measuring smell . An alternative path to measuring smell was to identify general perceptual primaries rather than individual odorant primaries [9]–[12] . This approach , consisting of applying statistical dimensionality reduction to many perceptual descriptors applied to many odorants , repeatedly identified odorant pleasantness , namely an axis ranging from very unpleasant to very pleasant , as the primary dimension in human olfactory perception [13]–[18] . Initial efforts to link such perceptual axes to odorant structural axes saw only limited success because of the limited scope of physicochemical features one could easily obtain for a given molecule [19] . However , the recent advent of software that provides thousands of physicochemical descriptors for any molecule ( e . g . , Dragon software , Talete , Milan , Italy ) now allowed application of similar dimensionality reduction to odorant structure as well . This process revealed odorant structural dimensions that were modestly but significantly predictive of odorant perception [17] and odorant-induced neural activity across species [20]–[24] . Although the above studies combine to generate an initial form of olfactory metrics , they all apply to mono-molecular odorants alone . The real olfactory world , however , is not made of mono-molecules , but rather of complex olfactory multi-molecular mixtures . For example , roasted coffee [25] , red wine [26] , or rose [27] , each contain hundreds of different mono-molecular species , many of them volatile . Thus , a useful metric for smell must apply to such odorant-mixtures . Although an ultimate metric would predict exactly how such mixtures smell in verbal descriptor terms , an initial interim goal is to predict their perceptual similarity . With this in mind , we collected perceptual similarity estimates from a large group of subjects rating a large group of odorant-mixtures of known components . We then tested alternative models linking odorant-mixture structure to odorant-mixture perceptual similarity , and identified a model and algorithm that provided a meaningful predictive framework . Using this algorithm we can now look at two novel mono-molecular odorants , or multi-component odorant-mixtures , and predict a significant portion of their ensuing perceptual similarity .
Odorants can generally be described by a large number of perceptual or structural descriptors . Dravnieks' atlas of odor character profiles includes 138 mono-molecules , each described by 146 verbal descriptors of perception . We call this the ‘perceptual odor space’ . Odorants can also be described by a large set of structural and physicochemical descriptors . We selected 1358 odorants commonly used in olfaction research , and obtained 1433 such descriptors using Dragon software ( v . 5 . 4 , Talete , Milan , Italy ) ( note that Dragon provides 1664 descriptors , but 231 descriptors were without values for the molecules we modeled ) . Since the different descriptors measure properties on differing scales we normalized the Dragon data so that the values of each descriptor ranged between 0 and 1 . That is , for each descriptor d we have a set of 1358 values ld ( barring missing values ) . Each values v in the list ld is normalized to the value vn by the equation ( 1 ) We call this normalized data the ‘physicochemical odor space’ ( Table S1 contains the odorants we modeled and their descriptor values ) . To form odorant-mixtures , we obtained 86 mono-molecular odorants that were well-distributed in both perceptual ( Figure 1A ) and physicochemical ( Figure 1B ) stimulus space ( Dataset #1 in Table S2 ) . We then diluted each of these odorants separately to a point of about equal perceived intensity as estimated by an independent group of 24 subjects , and prepared various odorant mixtures containing different numbers of such equal-intensity odorant components . Importantly , to prevent formation of novel compounds , odorant mixtures were not mixed in the liquid phase , but rather each component was dripped onto a common absorbing pad in a sniff-jar , such that their vapors alone mixed in the jar headspace ( the integrity of this method was later verified in Dataset #2 in Table S2 using gas-chromatography mass-spectrometry ( GCMS ) , see Methods ) . We prepared several different versions for each mixture size containing 1 , 4 , 10 , 15 , 20 , 30 , 40 or 43 components , such that half of the versions were well-spread in perceptual space , and half of the versions were well-spread in physicochemical space . We conducted pairwise similarity tests , using a visual analogue scale ( VAS ) ( see Methods ) , of 191 mixture pairs , in 48 subjects ( 24 women , average of 14 subjects per comparison ) . Each target mixture ( 1 , 4 , 10 , 15 , 20 , 30 , 40 or 43 components ) was compared to all other mixtures ( 1 , 4 , 10 , 15 , 20 , 30 , 40 or 43 components ) , and as a control , to itself . Other than comparisons of a mixture to itself ( 44 comparisons ) , all comparisons were non-overlapping ( 147 comparisons ) , i . e . each pair of mixtures under comparison shared no components in common ( Figure 1C ) ( Table S2 contains all the similarity estimates for the three datasets used in this study ) . One simple model for predicting the perceptual difference between mixtures is to measure all pairwise Euclidean physicochemical distances between all individual mixture components , and then average them . This approach treats each mixture component individually ( Figure 2A ) . To test this model , we obtained the 1433 physicochemical descriptors for each of the 86 mono-molecular components we used . We found that the mean pairwise Euclidean distance over all the descriptors of all mono-molecular components comprising any two mixtures was a poor predictor of perceptual similarity between the two mixtures . The relationship between pairwise-distance and perceived similarity did not fit any simple model , linear or other ( Figure 3A ) . Moreover , the distribution of this relationship was clearly skewed by the similarity ratings given to the comparisons of a mixture to itself , yet eliminating these comparisons revealed a significant correlation in the opposite direction ( r = 0 . 46 , p<0 . 0001 ) ( Figure 3B ) . In other words , this model implied that odor-mixtures identical in structure will be the furthest apart in perceptual similarity . Given this clear failing-point of the model , we set out to investigate an alternative model . An alternative model is to consider the mixture as a whole rather than a set of constituents ( Figure 2B ) . To test this , we used the same 1433 physicochemical descriptors for each mono-molecular mixture component , but this time we created a single vector representing the whole mixture by summing the vectors of its components . To eliminate the effect of the number of components in a mixture on the size of the mixture vector , we divided it by its norm . Thus , each mixture was now represented by a vector made of 1433 descriptors . We then defined the distance between the vector of mixture U and the vector of mixture V , as the angle between the two vectors , given by: ( 2 ) where U·V is the dot product between the vectors , and |U| , |V| are the norms of the vectors . We found that this angle distance was strongly predictive of perceived mixture similarity ( r = −0 . 76 , p<0 . 0001 ) ( Figure 3C ) . Omitting comparisons of mixtures to themselves resulted in a correlation of r = −0 . 49 , p<0 . 0001 ( Figure 3D ) . Unlike the pairwise distance model , this model did not predict that physically identical mixtures would in fact smell dissimilar . Therefore , we set out to optimize this model . In order to optimize the model , we first set out to collect an independent dataset ( Dataset #2 in Table S2 ) . To address the possibility that the performance of our model was somehow influenced by the nature of our mixtures , whose components were selected to span olfactory space , the components for Dataset #2 in Table S2 mixtures were selected randomly . We randomly selected 43 molecules out of the 86 equated-intensity molecules , and made 13 mixtures of 4–10 randomly selected components . Thus , unlike in Dataset #1 in Table S2 , and more like odors in the real world , here there was some overlap in components across mixtures . Twenty-four subjects ( 13 women ) conducted pairwise similarity tests of all 91 possible pairs plus 4 comparisons of identical mixtures for a total of 95 comparisons ( each such comparison was repeated twice ) . Subjects conducted the similarity tests within four sessions on four consecutive days ( ∼48 comparisons per day ) . Comparisons were counter-balanced for order . We set out to extract the most relevant chemical descriptors for predicting perceptual similarity using the angle distance model . In order to do so , we needed to compare the quality of predictions based on different combinations of descriptors . However , because our data includes 1433 different descriptors , it was impossible to compare all possible selections of descriptors in order to pick the best performing selection ( 21433 possibilities ) . With this in mind , we first set out to model the total number of descriptors our model would rely on . The first step in our optimizing method was to decide on the number of features ( descriptors ) we were going to look for . To do this we used a random half of Dataset #2 in Table S2 as a training-set ( 47 comparisons ) and ran a simulation on it . In the simulation we ran through each number of features from 1 to 100 . For each number of features n we selected 20 , 000 random samples of descriptors sized n and calculated the root mean square error ( RMSE ) for the prediction on the training set comparisons based on these descriptors . For each n we then calculated the mean RMSE and the standard deviation and plotted the result ( Figure 4A ) . At n = 20 the value of the mean RMSE minus the standard deviation was the lowest ( Figure 4A , the trend continues to increase for n>100 ) . This told us that at around 20 descriptors , we should expect the selections that would produce the lowest RMSE . Since our feature selection method includes the possibility of selecting a feature twice we searched for slightly larger size sets of features so that at the end of the process we would have about 20 descriptors . Although we could have compared the performance of a selection of descriptors , we wanted to estimate the relevance of individual descriptors . If we selected 25 descriptors at random out of the 1433 and based our predictive model on them , we were likely to obtain a prediction that correlated to an RMSE of about 11 ( Figure 4A ) . However in order to optimize our model we wanted to distinguish those descriptors which give rise to more accurate predictions from those that do not . In order to evaluate a descriptor d in terms of how much it contributes to accurate predictions we ran a simulation for each descriptor . In the simulation for descriptor d we tested the predictive performance of a large number of randomly selected sets of descriptors to which we added descriptor d . We used 2000 random selections of 25 descriptors together with d and tested their predictive performance on the same training and testing set from before . For each selection we calculated the RMSE , and then calculated the mean RMSE across the 2000 selections . This mean is the number assigned to descriptor d ( Figure 4B ) , giving us an indication of how relevant the descriptor d is to making similarity predictions: the lower the mean RMSE , the more relevant d is . Figure 4B is a plot of these averages calculated for each one of the 1433 descriptors . As apparent in the figure , for most descriptors the average performance for random selections that include them is about the same . However , some descriptors stand out . The next step in our descriptor selection process was a second simulation where we selected 4000 samples of 25 descriptor sets based on the performance of the individual descriptors in the second step of the selection process . We gave each of our descriptors a non-negative score based on its mean RMSE calculated in the first part of the process . The score was calculated as ( 3 ) so that only descriptors with an RMSE value lower than the average RMSE value ( i . e . good-performing descriptors ) were associated with a score greater than zero . Then we proceeded to select random samples according to the scores we just calculated . That is , in the third step of the process , those descriptors that performed better in the second step were more likely to be included in the ( semi ) random sample . Using this method we selected 4000 samples of 25 descriptors and picked the ones that performed best , i . e . the selection that produced the lowest RMSE in the training set predictions . We removed repeated descriptors from our best performing selection of 25 descriptors and obtained a selection of 21 descriptors that performed even better ( Table 1 ) . The performance of the descriptors selected according to this two-step training process was tested on the testing set and the resultant correlation between predicted odorant-mixture similarity and actual odorant-mixture similarity was RMSE = 6 . 98 , r = −0 . 85 , p<0 . 001 ( Figure 5 ) . Whereas the above random selection of descriptors may give rise to different descriptor subsets in recurring simulations , a deterministic selection of descriptors did not generate better results ( Text S1 Section 1 ) . One might ask how well our model performs under different conditions . Recall that so far we had optimized our model on Dataset #2 in Table S2 consisting of mixtures ranging in size from 4 to 10 components . We now set out to test the performance of our model and selected descriptors on Dataset #1 in Table S2 . This set not only includes larger mixtures but also includes 43 additional molecules not included in Experiment 2 . Using this set we obtained a correlation of r = −0 . 78 , p<0 . 0001 for all comparisons ( Figure 6A ) , and r = −0 . 52 , p<0 . 0001 for non-overlapping comparisons alone ( Figure 6B ) . To further get a sense of how well this selection of descriptors performs on this data , we compared its performance to that of 4000 randomly selected sets of 21 descriptors . We measured the performance in terms of RMSE on Dataset #1 in Table S2 . The selected set of 21 descriptors predicted similarity with an RMSE of 10 . 66 . Compared to randomly selected sets of descriptors , the optimized set performed better than 95 . 30% of the sets ( Figure 6C ) . Performance was tested using only the 147 comparisons between non-overlapping mixtures . One may ask how the model optimized and tested in odorant-mixtures performs with mono-molecules . To obtain similarity ratings for mono-molecules we pooled three experiments to form Dataset #3 in Table S2 . The first experiment included similarity ratings by 21 subjects ( 11 female ) between 14 pairs of mono-molecules; the second included similarity ratings by 17 subjects ( 9 female ) between 20 pairs of mono-molecules , and the third included 19 subjects ( 6 female ) rating 40 pairs of mono-molecules for similarity . In total , 49 mono-molecules were included in this experiment . The pool of molecules is included in the original pool of 86 molecules in Experiment #1 and includes 42 of the 43 in the pool of Experiment #2 . In total , 74 comparisons were conducted amongst the 49 molecules . Out of these comparisons , 65% ( 48 comparisons ) included at least one molecule that was not used in Experiment #2 . Each comparison was repeated twice . We applied our selected set of descriptors to Dataset #3 in Table S2 . As before , we measured the RMSE of the prediction made based on the descriptors we selected . We obtained an RMSE of 13 . 825 and r = −0 . 5 , p<0 . 0001 ( Figure 6D ) . In comparison , using all descriptors gave r = −0 . 39 , p<0 . 0001 . Thus , the set of descriptors optimized on Dataset #2 in Table S2 improved the predictive performance of our model on Dataset #3 in Table S2 . Notably , Dataset #3 in Table S2 consists of 7 additional molecules that were not included in Dataset #2 in Table S2 which was used to optimize the model . Moreover , as previously noted , 65% of these comparisons include at least one molecule that was not used in Experiment #2 . This renders the test on Dataset #3 in Table S2 fairly unrelated to the set of molecules used to optimize the model . If our model is to be helpful to researchers in the field , it must be applicable to data collected by others . Most published studies on olfactory mixtures looked only at simple mixtures of 2 to 4 components , and moreover , most all did not post their raw similarity matrices . The lack of posted raw data holds true for most studies of mono-molecular perceptual similarity as well , with one notable exception that we are aware of: Wright and Michels ( 1964 ) [28] printed a large table containing the pairwise similarity ratings given by 84 subjects to a matrix of odorants that included 33 odorants not in our experiments or model building . We applied our model to their data . The angle-distance model , whether using the non-optimized or optimized descriptor set , yielded a significant correlation between predicted and actual pairwise odorant similarity ( non-optimized: r = −0 . 60 , p<0 . 0001 ( Figure 6E ) ; optimized: r = −0 . 49 , p<0 . 0001 ( Figure 6F ) ; difference between r values: z = −1 . 34 , p = 0 . 18 ) . Thus , whereas Wright and Michels failed to predict perceptual similarity in their data [28] , our model was a significant predictor of similarity in this data collected half a century ago . The statistically equal performance across the optimized and non-optimized descriptors when applied to this dataset may have resulted from several factors , including that the odorant selection criteria may have reflected the theory they were testing , that the molecules were not first diluted to equated intensity , and that these were indeed mono-molecules whereas our optimization was for the prediction of mixtures . However , the most likely explanation for this relates to their testing procedure: they compared similarity of all odorants to five anchor odorants . The five anchor odorants , by definition , are a skewed representation of olfactory space . Therefore , we take this as a reminder that researchers who set out to use the current model should consider both its optimized and non-optimized versions , especially in cases where the data may be skewed in olfactory space . Based on measures of neural activity and receptor responses , primarily in rodents , but also in humans , two independent studies obtained two alternative sets of optimal physicochemical odor descriptors [20] , [23] . We set out to compare the performance of these sets of descriptors versus the current descriptors in predicting perceptual similarity . Application of the Haddad descriptor set ( containing 32 descriptors ) [20] and the Saito descriptor set ( containing 20 descriptors ) [23] to the testing set of Dataset #2 in Table S2 yielded RMSE = 12 . 4049 , r = −0 . 3608 , p = 0 . 01 and RMSE = 11 . 2255 , r = −0 . 5364 , p<0 . 0001 , respectively . Although significant , these predictions are significantly weaker than those obtained with the optimized angle distance model ( difference between r values , both z>3 . 16 , both p<0 . 005 ) .
One may argue that there are countless potential paths to model the contribution of the various physicochemical descriptors to the perception of similarity , and therefore ask why angle distance model was selected . Here we will describe the evolution of this model in our efforts: The simplest and most naïve initial solution to the problem we addressed was the pairwise distance model , and our initial efforts centered on its optimization . Although the details of this effort are beyond the scope of a single manuscript , we will note that the main weakness of the pairwise distance model is , as previously noted , its implication that the more common molecules two mixtures share , the more different they will smell . This is not a problem in the lab , where one can select non-overlapping mixtures ( e . g . , Dataset #1 in Table S2 ) . In the real world , however , many different mixtures will typically share many common components ( e . g . , Dataset #2 in Table S2 ) . We initially tackled this by adding a parameter that assigned a variable weight to the distance between components of one mixture that were ‘close’ to components of the second mixture . We then added a second parameter that defined the threshold for being considered a ‘close’ point . We optimized the added parameters but the performance of the model did not improve and the inconsistencies remained . In an attempt to further generalize our model we tried replacing the Euclidean distance that defines the pairwise distance with other typical functions . Amongst the functions we tested was dot product . When we did so , the other parameters that were selected in the optimization process pointed to a unified weight for all components in the mixtures . That is equivalent to a dot product of the sum of vectors . That is , the data pointed to the dot product of sums of vectors as a good model . Once we were led to a dot product of a sum of vectors we also normalized by the size of the vectors to eliminate the effect of the sheer number of components in a mixture . At this point we were already very close to an angle distance metric , after all , the cosine of the angle is the normalized dot product . When we finally arrived at an angle distance model the results were consistent with the comparisons of identical mixtures and the correlation was much stronger even without any added parameters . In simple terms , the superior performance of the angle-distance model over the pairwise-distance model suggests a system that does not consider each mixture component alone , but rather a system that , through some configurational process , represents the mixture as a whole . This is in fact highly consistent with olfactory behavior and neural representation . In behavior , humans are very poor at identifying components in a mixture , even when they are highly familiar with the components alone [30] . The typical maximum number of equated-intensity components humans can identify in a mixture is four , this number is independent of odorant type [31] , and does not change even with explicit training [32] . Moreover , perceptual features associated with a mono-molecule will sometimes make their way into a mixture containing that molecule , but sometimes not , and the rules for this remain unknown [33] . In other words , like our algorithm , human perception groups many mono-molecular components into singular unified percepts . This pattern , referred to as either associative , synthetic , or configural , is in contrast to the alternative of retaining individual mixture component identity , referred to as dissociative , analytical , or elemental . Although these patterns are not mutually exclusive , evidence from perception points to a primarily configural process in olfaction . Mixture synthesis may begin with a balance of agonistic and antagonistic interactions between mono-molecules at olfactory receptors in the epithelium [34] , [35] or at glomeruli in the olfactory bulb [36] , [37] . Thus , when components compete for common receptors , they may be harder to pick out of the mixture [38] . The configural mechanisms in epithelium and bulb are further reflected in cortex where patterns of neural activity induced by a mixture are unique , and not a combination of neural activity induced by the mixtures' components alone [39]–[43] . In other words , like our algorithm , also at the neural level , the olfactory system treats odorant-mixtures as unitary synthetic objects , and not as an analytical combination of components [42]–[49] . Although the model performed well , it has three notable limitations . The first is that the mixtures we studied were made of components that were first individually diluted to a point of equal perceived intensity . Intensity influences olfactory perception in complex ways [50]–[53] , and some odorants , such as indole , can sharply shift in percept with changing intensity [54] . Moreover , whereas some odorants can increase the overall intensity of a mixture they are added to , other odorants can reduce overall mixture intensity [55] . Given this complexity , one may assume that when one of two mixtures under comparison contains intensity-sensitive molecules such as indole , the power of our model may diminish . Notably , the independent test of our model ( Figure 6E , 6F ) implied that perceived intensity equation may not be a condition for the model to apply in the case of mono-molecular odorants . That said , the model will likely break down in mixtures whose components were not at all equated for perceived intensity . With this in mind , future iterations of our model should try to incorporate recently developed models for the prediction of odorant detection threshold ( as a proxy for intensity ) [56]–[58] . These models may provide an intensity coefficient that would allow applying our model to mixtures made of components that were not first equated for intensity . A second limitation is related to the odorants used for model building and testing . If the odorants represent only a limited portion of olfactory perceptual space , then our model may apply to this portion of olfactory space alone . To protect against this , we used the largest datasets we could find in order to build the model , and tested our model against subsets of the data not included in model building . Nevertheless , because the full extent of olfactory perceptual space remains poorly defined , this remains a potential limitation . Finally , a similar limitation is in the selection of physicochemical features we modeled . Again , the more features one incorporates into a model , the smaller the risk of not capturing the relevant sources of variance , and we modeled more than a thousand features . That said , due to our dependence on such tools as Dragon software , we model a large set of structural features , but lack in physical features . Specifically , features such as boiling point , vapor pressure , diffusion , etc . , which undoubtedly have a strong relation to olfactory perception , remain unrepresented . In conclusion , despite the above-noted limitations , we provide an algorithm that allows predicting odorant-mixture perceptual similarity from odorant-mixture structure . The synthetic nature of the algorithm is consistent with the synthetic nature of olfactory perception and neural representation . This algorithm can now serve as a framework for theory-based selection of components for odorant-mixtures in studies of olfactory processing .
We tested 139 normosmic and generally healthy subjects ( 63 women , between the ages of 21 and 45 ) who provided written informed consent to procedures approved by the Weizmann Institute Ethics Committee , and the Helsinki Committee . The experiments were conducted in stainless-steel-coated rooms with HEPA and carbon filtration designed to minimize olfactory contamination . All interactions with subjects during experiments were by computer , and subjects provided their responses through a computer keyboard or mouse . Odorant mixtures were sniffed from jars marked arbitrarily , and presentation order was counterbalanced across subjects . In order to minimize olfactory adaptation , a ∼40 second inter-trial interval was maintained between presentations . All odorants were purchased from Aldrich Chemicals ( St . Louis , MO ) in the highest available purity . All odorants were diluted with either mineral oil , 1 , 2-propanediol or deionized distilled water to a point of approximately equally perceived intensity . This perceived-intensity equation was conducted according to previously published methods [29] . In brief , we identified the odorant with lowest perceived intensity , and first diluted all others to equal perceived intensity as estimated by experienced lab members . Next , 24 naïve subjects ( 10 females ) smelled the odorants , and rated their intensity . We then further diluted any odorant that was 2 or more standard deviations away from the mean intensity of the series , and repeated the process until we had no outliers . This process is suboptimal , but considering the natural variability in intensity perception , together with naïve subjects' bias to identify “a difference” , and the iterative nature of this procedure , any stricter criteria would generate an endless process . To verify that our method of odorant-mixture preparation and delivery did not generate novel compounds , we submitted one set of mixtures ( Dataset #2 in Table S2 ) to analysis with GCMS . In brief , we left the samples to sit in closed vials for several hours , then incubated over night at 50°C . This was done to accelerate the kinetics of any potential reactions that may have occurred . All the individual components ( mono-molecules ) of the mixtures were run separately , to ascertain their purity . The single peak retention times and corresponding spectrum identifications were noted and verified using Wiley Registry 9th Edition/NIST 2008 combined mass spectral library ( Wiley , New York , NY ) . The mixture samples were then subjected to the same GCMS method as the single components , and Total Ion Chromatogram peaks were validated to contain only the expected peaks of their constituting single components . Peaks with wide or abnormal shapes were subjected to further spectrum deconvolution using AMDIS software ( NIST , Gaithersburg , MD ) , to assess potentially overlapping peaks . All analyses was made using an Agilent 7890 Gas Chromtograph coupled to Agilent 5975 Mass Spectrometer ( Santa Clara , CA ) , integrated with a Gerstel headspace sampler ( Mülheim an der Ruhr , Germany ) . Prior to injection samples were incubated in the Gerstel agitator for 5 minutes under 35°C and 250 rpm agitation . One ml of vial headspace gas was drawn into a heated syringe and injected to a split/splitless inlet that was kept at 250°C and a Split ratio of 5∶1 . The GC method used a HP-5 MS column ( 30 m×0 . 25 mm×0 . 25 µm ) and Helium as a carrier gas with 1 . 5 ml/min constant flow . Temperature program was 50°C for 3 minutes , 15°C/min ramp up to 250°C for 3 minutes . MS scans were conducted in Electron Impact mode ( 70 eV ) from m/z 40 to 550 , 2 . 86 scans/sec . MS source and Quad temperature were 230°C and 150°C , respectively . In each trial , each subject was presented with two mixtures and was asked to rate their similarity on a VAS . The question at the top of the VAS was “To what extent are these two odors similar” ( in Hebrew ) , and the VAS scale ranged from “not at all” to “highly” . In Dataset #1 the VAS was also numerated from 1 ( “not at all” ) to 9 ( “very” ) , and in the remaining datasets it was not numerated . In both cases , the ratings were normalized within subjects to a scale of 0% to 100% . Each subject repeated the experiment on two different days to assess test-retest reliability . We applied an arbitrary cutoff whereby if the difference between 2 repetitions of the same comparison was greater than 70% , this rating was excluded . This amounted to 109 out of 2070 ratings ( ∼5% ) in Dataset #1 in Table S2 , and no deletions in Datasets #2 and #3 in Table S2 . The ratings by subjects whose similarity ratings for identical mixtures were poorer by at least 2 standard deviations from the mean were discarded . This amounted to 3 subjects . The average rated similarities were calculated across subjects .
|
One hundred years ago , Alexander Graham Bell asked: “Can you measure the difference between one kind of smell and another ? It is very obvious that we have very many different kinds of smells , all the way from the odor of violets and roses up to asafetida . But until you can measure their likenesses and differences you can have no science of odor . ” Here we developed a computational framework and algorithm that looks at the molecular structure of two odors and predicts their ensuing perceptual similarity . Importantly , the algorithm works for odors that are each composed of a mixture containing tens of different molecules , much like natural smells . The algorithm that worked best was one that treats the odor-mixture as a single value , rather than a bunch of values reflecting each of its components . This is consistent with the growing view of how the mammalian brain treats odors: synthesizing a singular odor percept rather than analytically extracting individual odorant features from the odor-mixture . In conclusion , our algorithm's performance may contribute to the practice of the science of odor , and our algorithm's nature may contribute to the understanding of brain mechanisms of smell .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2013
|
Predicting Odor Perceptual Similarity from Odor Structure
|
Both Schistosoma mansoni and Schistosoma haematobium cause schistosomiasis in sub-Saharan Africa . We assessed the diagnostic value of selected Schistosoma antigens for the development of a multiplex serological immunoassay for sero-epidemiological surveillance . Diagnostic ability of recombinant antigens from S . mansoni and S . haematobium was assessed by Luminex multiplex immunoassay using plasma from school children in two areas of Kenya , endemic for different species of schistosomiasis . S . mansoni serine protease inhibitor ( SERPIN ) and Sm-RP26 showed significantly higher reactivity to patient plasma as compared to the control group . Sm-Filamin , Sm-GAPDH , Sm-GST , Sm-LAP1 , Sm-LAP2 , Sm-Sm31 , Sm-Sm32 and Sm-Tropomyosin did not show difference in reactivity between S . mansoni infected and uninfected pupils . Sm-RP26 was cross-reactive to plasma from S . haematobium patients , whereas Sm-SERPIN was species-specific . Sh-SEPRIN was partially cross-reactive to S . mansoni infected patients . ROC analysis for Sm-RP26 , Sm-SERPIN and Sh-SERPIN showed AUC values of 0 . 833 , 0 . 888 and 0 . 947 , respectively . Using Spearman’s rank correlation coefficient analysis , we also found significant positive correlation between the number of excreted eggs and median fluorescence intensity ( MFI ) from the multiplex immunoassays for Sm-SERPIN ( ρ = 0 . 430 , p-value = 0 . 003 ) and Sh-SERPIN ( ρ = 0 . 433 , p-value = 0 . 006 ) . Sm-SERPIN is a promising species-specific diagnostic antigen . Sh-SEPRIN was partially cross-reactive to S . mansoni infected patients . SERPINs showed correlation with the number of excreted eggs . These indicate prospects for inclusion of SERPINs in the multiplex serological immunoassay system .
Globally , more than 240 million people are still infected with schistosomiasis [1] . Over 90% of the infected people are resident in resource-limited settings in sub-Saharan Africa [2] . The next target of the current WHO roadmap for the control and elimination of schistosomiasis is to scale up mass drug administration ( MDA ) with Praziquantel ( PZQ ) [3] . Although PZQ is still efficacious in treating the disease , frequent reinfection necessitates repeated mass chemotherapy [4] . To achieve elimination , there is need for effective diagnostics to guide planning , implementation , monitoring and evaluation of the progress of the control intervention [5] , and for surveillance post-elimination . Conventionally , Kato-Katz stool examination is still the gold standard for the diagnosis . However , this method is now considered relatively less sensitive than the immunological detection of circulating cathodic antigens ( CCA ) or circulating anodic antigens ( CAA ) , for which specificity is still a challenge [6 , 7] . Thus , there is need to continue the search for effective diagnostics with adequate specificity and sensitivity [8] . In addition to its importance in MDA based interventions , better diagnostics are required for proper assessment of the efficacy of new drugs and vaccines [9] . The distribution of schistosomiasis coincides with several other neglected tropical diseases ( NTDs ) and other infectious diseases , including the big three: HIV , malaria and tuberculosis [10] . Integrating the control activity of these diseases presents a unique opportunity for optimum utilization of the meagre resources for research and health care delivery , especially for the NTDs whose distributions overlap with poverty [10] . Thus , the need for the development of novel strategies to simultaneously diagnose these pathogens has been recognized . Such strategy will be potentially cost effective and more feasible given the dearth of human resources , in addition to the requirement for minimal volume of human samples [11] . Our group have been exploring strategies for reliable epidemiological surveillance for infectious diseases , especially the NTDs . In one such approach , we developed a microsphere based multiplex immunoassay system to simultaneously detect multiple infectious diseases from a single minimal volume of human sample [12] . This method is ideal in a resource-limited context and is amenable to specific epidemiological settings; depending on the prevalent etiological agents and epidemiological situations . This strategy is already deployed for screening serotypes of a single pathogen [13–16] , and recently utilized by our team and others for simultaneous detection of several diseases , including NTDs [12 , 17] . For an efficient multiplex detection system , careful selection of antigens to include in the multiplex immunoassay system is crucial to the efficacy of the product , and indeed is the most important aspect of the development . For application of this multiplex system in schistosomiasis endemic areas , we utilized a literature-guided approach to select a panel of 10 Schistosoma mansoni antigens for assessment and inclusion in the multiplex immunoassay . We identified serine protease inhibitor ( SERPIN ) and S . mansoni recombinant protein RP26 as promising candidates . Also , based on assessment of samples from areas of single-species and non-overlapping transmission of S . mansoni and S . haematobium , we show that SERPIN can be applied to differentially detect the Schistosoma species . The biological role of SERPINs and RP26 , and the rationale and significance of their use for differential diagnosis of Schistosoma species is discussed .
This study was approved by the ethical review committee of Kenya Medical Research Institute , Kenya ( KEMRI , SSC No . 2084 ) and the ethical review board of Institute of Tropical Medicine , Nagasaki University , Japan ( No . 10121666 and 131219116 ) . Written informed consent was obtained from parents/guardians and school children prior to the study . The samples utilized in this study were collected in a cross-sectional study conducted in Mbita and Kwale areas in Kenya between September 2011 and March 2012 . The characteristics of the population and some initial analyses were detailed in a separate report [18] . Sera from amoebic liver abscess were obtained from the International Centre for Diarrheal Disease Research , Bangladesh . Diagnoses were confirmed by detection of Entamoeba histolytica-specific DNA in liver abscess pus specimens . Twenty Leishmania donovani patients’ sera were obtained from the Rajshahi Medical College in Bangladesh . Again , diagnoses were parasitologically confirmed by microscopic examination of spleen aspirates . As negative controls , sera from 32 Japanese individuals were included . S . mansoni adult worms were stored in RNAlater ( QIAGEN ) on collection . The samples were then crushed using Tissue-Ruptor ( QIAGEN ) . Total RNA was isolated from crushed adult worms using TRIzol plus RNA Purification kit ( Ambion ) , according to the manufacturer’s instructions . The cDNAs of target regions were amplified by using PrimeScript High Fidelity RT-PCR Kit ( Takara ) , according to the manufacturer’s instructions . Briefly , the first strand cDNA was synthesized from RNA with reverse transcriptase using oligo dT primer , and PCR was performed by using the reverse transcription reaction mixture as the template with a pair of specific primers for each candidate antigen . PCR products were purified using QIAEX II Gel Extraction Kit ( QIAGEN ) . Sh-SERPIN cDNA was chemically synthesized by Integrated DNA Technologies , Inc . Introduction of G444A mutation in Sm-SERPIN was performed by overlap extension PCR . The cDNAs were then sub-cloned into pET-52b ( + ) vector ( Novagen ) . Structure of the antigens and the fusion tags are summarized in Table 1 . Recombinant antigens were expressed and purified as previously described [12] . GAPDH and Tropomyosin was located in inclusion body , whereas the other antigens were located in soluble fraction . S . mansoni soluble egg antigen ( SEA ) was prepared using standard methods and as previously detailed [19] . Two antigen-coupling protocols were used in this study . For the initial experiments on S . mansoni antigens , protocol previously reported by our group was used ( panel #1 ) [12] . Briefly , for 1 . 25 million MagPlex microspheres , 100μg of each antigen ( except Tropomyosin ) were added . Because microspheres aggregation was observed for Tropomyosin at 100μg antigen , 0 . 3μg of Tropomyosin was added as determined by titration experiment . Because new protocol was released from Luminex before the subsequent experiment , the new protocol was used for the second experiment with RP26 , Sm-SERPIN and Sh-SERPIN ( panel #2 ) . Binding reaction of antigen on microspheres was performed in coupling buffer ( 50mM MES pH 5 . 0 ) instead of PBS ( - ) . The amount of antigens in the reaction was reduced to 25μg for 1 . 25 million microspheres . Data analyses were performed on GraphPad Prism version 6 . 0 . Receiver operating characteristics ( ROC ) analysis was performed to assess diagnostic value of antigens . Mann-Whitney tests were performed for comparison between two groups . Spearman’s rank correlation coefficient analysis was performed to assess correlation between two measured variables . Data was presented as dot plots , with lines and error bars showing the median and interquartile ranges . Statistical significance was set at p < 0 . 05 .
Previously , we developed a microsphere based multiplex immunoassay system to simultaneously detect multiple infectious diseases from a single human sample [12] . To include schistosomiasis in the range of antigens in our multiplex system , we used literature-guided approach to select 10 Schistosoma mansoni antigens for evaluation and potential inclusion in the multiplex system . Cathepsin B ( Sm31 ) [20] , Filamin [21] , Glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) [22] , Glutathione S-transferase ( GST ) [23] , Legumain ( Sm32 , Haemoglobinase ) [20] , RP26 ( Sm22 . 3 , LGG ) [24] , Tropomyosin-2 ( TM-2 ) [25] were selected from reports on S . mansoni diagnosis . Serine protease inhibitor ( Sm-SERPIN ) was selected from report on S . haematobium diagnosis [26] . Leucine aminopeptidase-1 ( LAP-1 ) [27] and Leucine aminopeptidase-2 ( LAP-2 ) were selected from reports on S . japonicum diagnosis [27] . S . haematobium SERPIN was also included for comparison with Sm-SERPIN . Soluble egg antigen ( SEA ) from S . mansoni was used as control . Table 1 shows the selected antigens , their GenBank accession numbers , the N- and C-terminus fusion tags , and predicted molecular weights . The purified proteins were subjected to SDS-PAGE and protein bands of expected molecular weight were identified ( Fig 1A ) . Although protein bands of expected molecular weight was identified for both Sm-SERPIN and Sh-SERPIN , we observed additional shorter fragment of about 30 kD on Sm-SERPIN lane ( Fig 1A ) . Sequence analysis showed that this shorter fragment was a product of transcription from prokaryotic start codon ( GUG ) located at nucleotide 442 in the reference sequence ( GenBank: CCD60071 ) . Therefore , to compare diagnostic ability with similar quality of Sm-SERPIN and Sh-SERPIN , we introduced a point mutation at nucleotide 444 of the cDNA sequence ( G444A ) , resulting in codon change from the prokaryotic start codon GUG to GUA . This is a silent mutation encoding for Valine . This strategy effectively reduced this extra 30 kD fragment to the minimum in Sm-SERPIN with G444A mutation ( Fig 1B ) . The Sm-SERPIN G444A variant was therefore used in the subsequent experiments . To confirm success of antigen coupling to the MagPlex microsphere beads , phycoerythrin-conjugated antibody against the poly-histidine tag was used to determine the number of antigens on the microspheres . In the initial coupling beads set ( Panel #1 ) , all the antigens showed significantly high signals , indicating success of antigen coupling ( Fig 2A ) . Relatively , SERPIN showed the highest level of success in antigen coupling ( MFI = 8 , 805 ) , while LAP-1 and RP26 showed the lowest level with MFIs of 1 , 523 and 1 , 629 , respectively ( Fig 2A ) . In the second coupling beads set ( Panel #2 ) , the antigens showed similar MFI values using different coupling protocol from that used for Panel #1 ( Fig 2B ) . The choice of antigens is crucial for the success of the multiplex assay development . To assess the 10 selected antigens for suitability for diagnosis of schistosomiasis in the multiplex assay , we compared the reactivity of the S . mansoni candidate antigens in multiplex format against plasma from pupils resident in two epidemiologically distinct endemic areas in Kenya . While Mbita area is endemic for S . mansoni , Kwale area is an S . haematobium endemic area . For this initial screening , twenty-six ( 26 ) plasma samples each from well characterized healthy and schistosomiasis patients from each endemic area were included in the assays and analyses . Sera from healthy Japanese volunteers were also included as negative controls from non-endemic area . Only RP26 and Sm-SERPIN were specifically reactive to plasma from S . mansoni patients ( Fig 3 ) . The other antigens did not show disease-specific reactivity , although the reactivity to plasma from endemic areas is higher than reactivity to Japanese control as would be expected . The two antigens with disease-specific reactivity ( Sm-RP26 and Sm-SERPIN ) showed fluorescence signals ( MFI ) that are similar or higher than signals recorded from evaluation of coupling with anti-polyhistidine-tag antibody ( Fig 2A ) . This further shows that the absence of reactivity from the other 8 antigens was not a function of number of coupled antigens . For instance , Sm-RP26 antigen with relatively low microsphere-coupled antigens based on evaluation with anti-polyhistidine-tag antibody showed significant disease-specific reactivity . Interestingly , Sm-RP26 was also reactive to plasma from S . haematobium patients ( Fig 3 ) . Our data also showed that Sm-SERPIN was efficient in species-specific differential detection of infections with S . mansoni . Also , S . haematobium SERPIN was previously reported to be species-specific [26] . The crude egg antigen ( SEA ) from S . mansoni showed similar MFI value between egg negative and positive in Mbita area , where S . mansoni is endemic; suggesting that some egg negative individuals may have history of previous infections . Surprisingly , SEA from S . mansoni showed disease-specific detection only for S . haematobium patients ( Fig 3 ) . Taken together , our data showed that Sm-SERPIN and Sm-RP26 are promising diagnostic candidates to be potentially included in our multiplex system . In addition , Sm-SERPIN demonstrated species-specific reactivity . To further confirm the diagnostic efficacy of Sm-SERPIN and Sm-RP26 from the preceding section , we assessed a larger number of plasma for reactivity with the antigens , Sm-RP26 and Sm-SERPIN . In addition , Sh-SERPIN was included in the analysis to corroborate the species-specific diagnostic efficacy of SERPINs . Sequence homology of amino acid for the two SERPINs is presented in Fig 4 . Two sequences are very similar having 76% identical amino acids . Equally , to preclude the possibility of cross reactivity with other helminths and infectious diseases , plasma from hookworm patients in Kwale area of Kenya , amoebic liver abscess ( ALA ) and visceral leishmaniasis ( VL ) patients in Bangladesh were included in the assays and analyses . Again , similar to data presented in Fig 3 , Sm-RP26 was only reactive to plasma from patients with either S . mansoni or S . haematobium ( Fig 5A ) , while Sm-SERPIN was species-specific ( Fig 5B ) . Sh-SERPIN was partially cross-reactive to plasma from patients with S . mansoni; with median of MFI value higher in S . haematobium than in S . mansoni ( Fig 5C ) . It was also noteworthy that cross-reaction was not observed with plasma from ALA , VL and hookworm patients ( Fig 5 ) . Our observations further provide evidence for the utility of both Sm-RP26 and SERPINs as good diagnostic antigens for schistosomiasis , in addition to the species-specific diagnostic advantage of the Sm-SERPIN . We performed ROC analysis for further evaluation of the diagnostic performance of the SERPINs and Sm-RP26 in the multiplex assay . Sensitivity and specificity were calculated at MFI value giving maximum sensitivity plus specificity . For the ROC analysis for Sm-RP26 , egg negative and positive individuals in both Mbita and Kwale were included in the analysis . For SERPINs however , data from plasma samples from the endemic area for each species were used . The ROC curve and the corresponding statistics are summarized in Fig 6 . The AUC values for Sm-SERPIN ( 0 . 888 ) and Sh-SERPIN ( 0 . 947 ) were higher than AUC value for Sm-RP26 ( 0 . 833 ) . The two SERPINs recorded sensitivity of approximately or over 90% ( 92 . 5% for Sh-SERPIN and 89 . 4% for Sm-SERPIN ) , while Sm-RP26 showed lower sensitivity of 67 . 8% . Sh-SERPIN was the most specific antigen with 90% specificity , followed by Sm-RP26 and Sm-SERPIN with specificities of 89 . 5% and 81 . 5% , respectively ( Fig 6 ) . Succinctly , our data indicate that the SERPINs recorded acceptable levels of sensitivity and specificity and are suitable for inclusion as diagnostic antigens in the multiplex system . The additional advantage of species-specific detection also provides further utility for these antigens . The three promising diagnostic antigens from this study will be more valuable if they can also , in addition to qualitative diagnosis , provide an estimate of the number of excreted eggs . Such information will be very useful in planning interventions and for prioritization of intervention to the high-risk populations . Spearman’s rank correlation coefficient was applied for identification of correlation between MFI signals from multiplex immunoassays and the number of excreted eggs from subjects ( Fig 7 ) . The Spearman’s rank correlation coefficient showed that the multiplex assay for Sm-SERPIN and Sh-SERPIN had a statistically significant correlation to average number of observed eggs ( for Sm-SERPIN , ρ ( rho ) = 0 . 430 , p-value = 0 . 003; for Sh-SERPIN , ρ = 0 . 433 , p-value = 0 . 006 ) . Although Sm-RP26 could detect both species in the earlier described immunoassays , it only showed slight correlation with only S . mansoni egg number ( ρ = 0 . 307 , p-value = 0 . 036 ) , but not for S . haematobium ( ρ = 0 . 060 , p-value = 0 . 713 ) . Taken together , these data indicate the diagnostic potentials of Sm-SERPIN and Sh-SERPIN for respective species , and strongly support their inclusion in the multiplex immunoassay system .
GeneBank accession number for each antigen were as follows; Cathepsin B ( Sm31 ) : AAA29865 , Filamin: CCD59036 , GAPDH: P20287 , GST ( Sm28 ) : P09792 , LAP-1: ACQ77148 , LAP-2: ACQ77149 , Legumain ( Sm32 ) : P09841 , RP26 ( Sm22 . 3 ) : AAB81008 , Sm-Serpin: CCD60071 , Tropomyosin-2: P42638 , Sh-Serpin: AAA19730 .
|
More attention is now shifting towards elimination of some of the neglected tropical diseases , including schistosomiasis . Efficient diagnostics and surveillance tools are the bedrock of planning , implementation , monitoring and evaluation of such disease interventions . We had developed a multiplex immunoassay system for simultaneous detection of several pathogens in a single limited volume of human sample . To include Schistosoma antigen among the panel of pathogen antigens , we assessed the diagnostic value and suitability of selected Schistosoma antigens in multiplex format . S . mansoni serine protease inhibitor ( SERPIN ) and Sm-RP26 showed good diagnostic value with significant reactivity to patient plasma as compared to the control group . However , S . mansoni Filamin , GAPDH , GST , LAP1 , LAP2 , Sm31 , Sm32 and Tropomyosin did not show disease-specific reactivity to plasma from S . mansoni infected patients . While Sm-RP26 was cross-reactive to plasma from S . haematobium patients , Sm-SERPIN showed species-specific reactivity . There was also significant positive correlation between the number of excreted eggs and fluorescence signals from the multiplex immunoassays for the SERPINs . These findings indicate potentials for utilization of SERPINs in the multiplex system .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Species-Specific Serological Detection for Schistosomiasis by Serine Protease Inhibitor (SERPIN) in Multiplex Assay
|
In order to complete their life cycle , papillomaviruses have evolved to manipulate a plethora of cellular pathways . The products of the human Alphapapillomavirus E6 proteins specifically interact with and target PDZ containing proteins for degradation . This viral phenotype has been suggested to play a role in viral oncogenesis . To analyze the association of HPV E6 mediated PDZ-protein degradation with cervical oncogenesis , a high-throughput cell culture assay was developed . Degradation of an epitope tagged human MAGI1 isoform was visualized by immunoblot . The correlation between HPV E6-induced degradation of hMAGI1 and epidemiologically determined HPV oncogenicity was evaluated using a Bayesian approach within a phylogenetic context . All tested oncogenic types degraded the PDZ-containing protein hMAGI1d; however , E6 proteins isolated from several related albeit non-oncogenic viral types were equally efficient at degrading hMAGI1 . The relationship between both traits ( oncogenicity and PDZ degradation potential ) is best explained by a model in which the potential to degrade PDZ proteins was acquired prior to the oncogenic phenotype . This analysis provides evidence that the ancestor of both oncogenic and non-oncogenic HPVs acquired the potential to degrade human PDZ-containing proteins . This suggests that HPV E6 directed degradation of PDZ-proteins represents an ancient ecological niche adaptation . Phylogenetic modeling indicates that this phenotype is not specifically correlated with oncogenic risk , but may act as an enabling phenotype . The role of PDZ protein degradation in HPV fitness and oncogenesis needs to be interpreted in the context of Alphapapillomavirus evolution .
Papillomaviruses ( PVs ) are a diverse family of dsDNA viruses infecting most , if not all , amniotes . Based on nucleotide similarities , PVs are classified into genera identified by Greek letters . A genus is further divided into numbered species [1 , 2] . Persistent infection with specific human papillomaviruses ( HPVs ) has been shown to be necessary for the induction of cervical carcinoma [3 , 4] . All established oncogenic HPV types ( OTs ) belong to the genus Alphapapillomavirus [5] . Of note , phylogenetically , these oncogenic HPV types cluster into a so-called high-risk ( HR ) clade , indicating an evolutionary relationship between these viruses [5 , 6] . Importantly , available epidemiological data suggests that while some HPV types ( e . g . HPV16 ) within this HR clade are strongly associated with cancer , others ( e . g . HPV68 ) are only oncogenic in rare cases [7] . Throughout papillomavirus evolution PVs continuously adapted to new ecological niches on the host . This process selected for PVs with specific phenotypes needed to interact with changing cellular environments . It is highly unlikely that the ability to cause malignancy provided certain PVs with an evolutionary advantage . This suggests that the Alphapapillomaviruses acquired a particular combination of phenotypes while adapting to a specific ecological niche . In some viruses , the resulting cellular insult may inadvertently drive the infected host cell towards transformation . We previously postulated that use of phylogenetic , epidemiological and biochemical analyses would be essential for the identification of viral phenotypes specifically associated with oncogenicity [5 , 8 , 9 , 10] . Most papillomaviruses express at least 7 proteins , two of which—E6 and E7—have been demonstrated to be sufficient for oncogenesis . To date , the exact mechanisms by which these viral proteins cause cellular transformation is unknown . The viral E6 and E7 proteins interact with a diverse set of cellular pathways . Some of these interactions have been proposed to be unique to oncogenic viruses , while others appear to be shared by all investigated types [reviewed in 10 , 11 , 12 , 13 , 14 , 15 , 16] . It is well established that the E6 protein from specific HPVs targets PDZ containing proteins for degradation [17] . PDZ domains represent an abundant class of protein interaction modules that target specific motifs on partner proteins . PDZ containing proteins regulate multiple biological processes including differentiation and the maintenance of cellular polarity [17 , 18 , 19 , 20 , 21] . The interaction with PDZ containing proteins is dependent on a canonical class I PDZ binding motif ( PBM ) at the extreme C-terminal of the E6 protein [22] . Several E6 proteins interact with the PDZ-protein MAGI1 ( a member of the membrane-associated guanylate kinase ( MAGUK ) protein family ) [23] , and MAGI1 is highly sensitive to degradation by E6 proteins [24 , 25] . However , only HPVs containing a type I PBM are able to degrade MAGI1 in vitro , and it has been stated that only oncogenic types contained such a motif [26] , implying that interactions with PDZ containing proteins are critical for HPV-induced oncogenesis [27] . In the present paper , we provide evidence that all members of the HR-HPV clade ( i . e . , α-5 , -6 , -7 , -9 and -11 species groups ) indeed contain a class I PBM . Nevertheless , there does not appear to be a correlation between the presence of this motif and oncogenic classification . We also present data that all tested HR types , regardless of their oncogenic potential , degrade a human MAGI protein ( hMAGI1d ) . Importantly , the design of our study ( in which we tested E6 proteins from viruses representing all known species within the genus Alphapapillomavirus ) allowed us to evaluate this phenotype from an evolutionary perspective . The derived evolutionary model demonstrates that the capability to degrade hMAGI1 entered the viral-host relationship prior to the oncogenic phenotype . In addition , while PDZ protein degradation is not sufficient for cancer development , the model suggests that it enabled the evolution of a phenotype associated with cell transformation . These data further support the notion that biological processes are best understood once their evolutionary origin is taken into account [28] .
The patient samples used in this study have been IRBB approved . ( IRB number: 2009–274; Approval date: 07/08/2009; Title: Epigenetic Profiling of Cervical Cancer: A Pilot Study . ) The C-33A cell line is an HPV negative cervical cancer cell line expressing a mutant p53 protein [29 , 30] and was obtained from the ATCC ( Manassas , VA , USA ) . The C-33A cells were maintained in Dulbecco’s modified Eagle’s medium ( DMEM ) with 10% fetal bovine serum ( FBS ) . Western blot analysis C-33A cells were plated in 6-well plates ( 800 , 000 cells/well ) and transfected with 1400 ng pQCXIN-HA-hMAGI1d , 600 ng of each pQCXIN based E6 expression vector and 100 ng pSUPER-GFP ( transfection control ) using lipofectamine LTX ( Invitrogen ) according to the manufacturer’s instructions . The cells were lysed in RIPA buffer ( Millipore , Billerica , MA , USA ) , containing appropriate protease inhibitors . Equal amounts of protein lysates ( 10 ug ) were separated by SDS-PAGE . Western blot membranes were probed with antibodies recognizing the HA-tag ( Cell Signaling , Danvers , MA , USA ) , eGFP ( Cell Signaling ) or actin ( Sigma ) . Epidemiological classification of HPVs is based on a recent review of available world-wide reported data as conducted by an expert panel [7] . E6 proteins were selected to include at least one representative for each species within the genus Alphapapillomavirus . In addition , closely related viruses with differences in epidemiological classification were selected . The isolation and cloning of E6 ORFs has been described previously [8 , 9] . E6 sequences representing all viral types within the genus Alphapapillomavirus were downloaded from the PaVE database ( http://pave . niaid . nih . gov/#home ) [31] . The DNA sequences were translated to amino acids and aligned using MAFFT ( the L-INS-I algorithm was used ) [32] , and the corresponding nucleotide coding regions were aligned within the Seaview program [33] . Bayesian tree reconstruction was performed using MrBayes [34 , 35] while implementing the GTR+I+G model ( as selected by Jmodeltest2 [36 , 37]; S1 Table ) . To ensure the most efficient Bayesain analysis the convergence of the Markov chain Monte Carlo ( MCMC ) chains was explored graphically using AWTY [38] . The final posterior sample of trees is summarized as a majority rule consensus tree in Fig 1 . The viral traits under study ( MAGI1 degradation and oncogenic potential ) were coded as discrete presence/absence variables . Clades for which no epidemiological and/or biochemical data were available were coded as missing , and were considered ambiguous by the program . The computer program BayesDiscrete [39] ( available from http://www . evolution . rdg . ac . uk ) was used to detect evidence of correlated evolution between discrete traits . In order to account for phylogenetic uncertainty during the tree building process , 500 trees were randomly sampled from the post-burn-in posterior sample of MrBayes trees . The reversible jump MCMC analysis was run for 1x109 iterations . After a burn-in of 1x106 iterations , the chains were sampled every 10 , 000th iteration . The alpha and beta parameters of a Gamma distribution were seeded from uniform distributions on the interval 0 to 2 . To confirm the robustness of the results , the analysis was performed three independent times . Bayes factors were used to select between models [39] . Traditionally , a log Bayes Factor greater than 30 is considered as positive evidence for the model being tested .
In an attempt to understand differences in oncogenic potential , researchers have traditionally compared viral phenotypes between the prototypical HR and LR types ( HPV16/HPV18 vs . HPV6/HPV11 ) [13] . However , these viruses infect different anatomical niches on the human body and are separated by approximately 30 million years of evolutionary divergence , confounding simple comparison [10] . Phylogenetic analysis ( Fig 1 ) , clustered the members of the genus Alphapapillomvirus into two main clades [10] . Epidemiological evidence has shown that only a subset of the viruses in this HR clade ( highlighted in red ) is actually associated with cervical cancer [7 , 40 , 41] . The alignment of all known Alphapapillomavirus E6 proteins ( see panel to the right of Fig 1; 6 C-terminal residues are shown ) illustrates that the presence of a canonical type 1 PBM ( S/TXΦ where Φ indicates any hydrophobic residue [42] ) represents a synapomorphy of the high-risk papillomaviruses . In other words , all HR viruses , independent of associated oncogenic risk have a PDZ interacting domain . While most viruses in the LR clade do not contain comparable residues , HPV types belonging to the Alphapapillomavirus 8 species contain an E6 protein with a similar motif . The sequence analysis suggests that these alpha-8 E6 proteins may be able to target PDZ containing proteins for degradation . Thus , the E6 proteins from several non-oncogenic viruses contain putative PBMs , suggesting that the presence of a type 1 PBM does not allow for dichotomization between oncogenic and non-oncogenic HPV types . However , the presence of a sequence motif does not establish that these E6 proteins actually target PDZ proteins for degradation . Therefore , we examined the effects of E6 protein expression on steady state levels of a human PDZ-protein in vivo . In order to test whether different E6 proteins affect the steady state levels of a human PDZ protein , a novel isoform of MAGI was cloned from cervical cells ( S1 Methods ) . We tested the effects of co-expressing HPV E6 proteins with the PDZ containing human protein hMAGI1d ( Fig 2A ) . The figure shows a representative gel of the obtained results . These data demonstrate that all members of the HR-clade ( highlighted in red ) degrade hMAGI1d when compared to the empty vector control ( pQCXIN ) . In addition , HPV40 E6 , a representative of the alpha-8 species group , a viral type embedded in the LR-clade also degraded hMAGI1d in this system ( Fig 2 and S1 Fig ) . Thus , the alpha-8 PBM ( Fig 1 ) is associated with hMAGI1d degradation . It has been shown that the HPV E6 proteins target PDZ-proteins for proteasomal degradation [43 , 44] . We confirmed that the tested E6 proteins ( including HPV40 E6 ) degrade human MAGI1d in a proteasome dependent manner ( S1 Fig ) . HPV oncogenic potential and MAGI1d degradation is contrasted using mirrored trees in Fig 2B . The left tree is shaded according to epidemiological classification , while the tree on the right indicates the ability of HPV types to degrade MAGI1 as shown in Fig 2A . While this data indicates incoherence between both PDZ degradation and oncogenic activity , this hypothesis is formerly tested in the next paragraph . The main goal of this study was to test whether degradation of PDZ containing proteins was associated with oncogenic potential as determined by the clinical/epidemiological empirical cancer risk [7] . The biochemical degradation of PDZ-proteins and viral oncogenic potential were coded as discrete variables , and tested to determine whether these pairs of discrete binary traits evolved together . Briefly , the test of correlated evolution , as implemented within the BayesTraits package [39] , compares the fit of two models of evolution . In the independent model both traits are allowed to evolve independently on the tree . The dependent model requires that both traits evolve in a correlated fashion . In order to incorporate phylogenetic uncertainty we integrated the analysis over 500 trees randomly selected from the posterior distribution . We employed a reversible-jump ( RJ ) MCMC methodology [39] . The RJ-MCMC has the advantage that it travels through the posterior tree sample and tests all possible models of evolution in proportion to the likelihood of each individual model . This method provides a posterior sample ( n = 100 , 000 ) of models with their associated transition rates . We obtain support for each model through the calculation of Bayes Factors based on the frequency with which each model occurs in the posterior sample compared to the prior expectation of observing that model ( i . e . , the ratio of posterior to prior odds ) [45] . Bayes Factors are interpreted on a heuristic scale . Traditionally , Bayes Factors greater than 30 are considered very strong evidence in favor of one model . To ensure reproducibility of the RJ-MCMC analysis , the analysis was performed three independent times . Fig 3A shows the top 5 models selected by the RJ-MCMC approach . The selected model ( “0ZZ0ZZ0Z”; Fig 3A ) has a Bayes Factor of 130 . 41 ( +/- 1 . 99 ) demonstrating strong support for this model . We also reconstructed the ancestral combination of phenotypes at important nodes of the phylogenetic tree . This suggested that the most recent common ancestor ( MRCA ) of all high and low-risk Alphapapillomaviruses did not degrade PDZ proteins and was not oncogenic ( probability P ( 0 , 0 ) = 0 . 79; Fig 3B , left panel ) . Likewise , the MRCA of the LR clade did not degrade PDZ proteins . However , the ancestor of the extant HR viruses acquired the ability to degrade PDZ proteins . Importantly , the analysis suggests that this ancestral HR virus would not have been oncogenic ( P ( 0 , 1 ) = 0 . 74; Fig 3B , middle panel ) . Fig 3C shows the estimated instantaneous rates change between the four different phenotype combinations . The selected model suggests that the ancestor of the HR clade acquired the ability to degrade PDZ proteins , which is associated with a relatively low probability , likely reflecting the fact that a new interaction motif had to be created . This combination of phenotypes ( able to degrade PDZ proteins while not oncogenic ) appears to be highly unstable , and suggests that the ability to degrade certain PDZ proteins resulted in sub-optimal viral fitness . Under these assumptions , the model predicts two equally probable scenarios . In the first scenario , the progeny viruses would lose the capacity to degrade PDZ containing proteins . Since no extant members of the HR clade lost the ability to degrade PDZ proteins , we hypothesize that such viruses were unable to colonize the available niche and went extinct . The second scenario proposes that the ancestors of the extant HR viruses were able to colonize and thrive in a new ecological environment . It is likely that these viruses had to evolve ways to cope with the cellular milieu following the loss of PDZ containing proteins . Furthermore , the model predicts that it is highly unlikely to become oncogenic without degrading MAGI1 . Taken together the evolutionary data suggests that the ability to degrade PDZ proteins represents an enabling phenotype that had to be acquired prior to further adaptation . Some of these additionally adopted phenotypes inadvertently resulted in cellular deregulation , transformation and cancer .
In the present study we tested the ability of E6 proteins from different HPV types to degrade PDZ containing proteins . We used evolutionary hypothesis testing to reconstruct the plausible evolutionary events involved in E6-induced PDZ-protein degradation and HPV type cancer risk . These analyses indicated that the biochemical phenotype was not associated with the cancer risk , but rather with the evolution of extant viral types . Due to their small genomic size , the complete genomic sequence of over 200 papillomaviruses has been characterized [31] . Importantly , the empirically derived contribution of most HPVs to ( cervical ) cancer has been well documented . While ( cervical ) carcinogenicity of the HR-HPV types likely varies in strength along a continuum without clear demarcation , from extremely strong ( e . g . HPV16 ) to probably carcinogenic in rare situations ( e . g . HPV 68 ) , evidence suggests that only a small subset of viruses is responsible for virtually all cancers Of note , viruses closely related to these oncogenic HPV types are significantly less common in cancer , despite the fact that their overall prevalence is comparable . Suggesting that these related viruses are implicitly less carcinogenic . Taking into account some of the limitations associated with epidemiological classification , the present study only considered viruses as oncogenic if sufficient epidemiological evidence was available based on analysis by experts [41] . The analysis of biochemical assays in light of the available genomic information and epidemiological classification has allowed us to predict which viral phenotypes may be pathogenic and begin to explain why only a subset of these viruses is truly oncogenic . The use of evolutionary information has been valuable to interpret the relationship between HPV type specific phenotypes and HPV oncogenicity [8 , 9] . While developing the in vivo assay , we cloned and sequenced a novel isoform of the MAGI1 family , hMAGI1d . We provide evidence that this isoform is expressed in cervical cell lines and primary cervical tumors . We confirmed that , as was previously reported for other MAGI proteins , HPV16 E6 dramatically affects the steady state levels of exogenously expressed human MAGI1d [43] and that this reaction is dependent on the proteasome . Since hMAGI1d appears to be degraded by HPV E6 proteins in a manner similar to what has been reported for other PDZ containing proteins we used this human isoform as a representative member of the PDZ-protein family . Previous studies have shown that specific residues within the PBM and/or certain post-translational modifications may result in each E6 protein having a specific PDZ protein interactome [23 , 46 , 47 , 48] . However , the goal of this study was to relate the overall ability to degrade PDZ containing substrates to oncogenic potential . It was previously shown that most E6 proteins interact with MAGI1 with nano-molar affinity [49] . Therefore , by comparing the steady state levels of hMAGI1d , and not the kinetics of degradation , this approach allows for the representation of the degradation phenotype as a discrete variable . The analysis of 24 tested viral E6 proteins showed that all members of the HR-clade , independent of oncogenic risk-classification were capable of degrading hMAGI1d . In addition , we provide evidence that the E6 protein of HPV40 , a member of the alpha-8 species , contains a functional type 1 PBM . While HPV E6 proteins appear to prefer a Val or Leu residue , any hydrophobic residue can allow for interaction with PDZ containing proteins [42] . Indeed , the PBM found on the HBV core protein contains a terminal cysteine [50 , 51] . The observation that the members of the alpha-8 clade are the only low-risk viruses that encode E6 proteins with a functional PBM may be the result of convergent evolution . This is supported by the presence of a divergent cysteine ( as opposed to the canonical valine or leucine ) . Importantly , this implies that ( some ) members of the low-risk clade have the inherent ability to degrade cellular targets . Indeed , we have previously shown that the E6 protein derived from the low-risk virus HPV71 is able to degrade p53 [8] . Furthermore , when the HPV18 PBM was grafted onto HPV11 E6 , this chimeric protein was able to degrade cellular PDZ proteins [52] . This supports the notion that the ability to degrade cellular protein is inherently shared by all Alphapapillomaviruses [52] . We used evolutionary trait analysis to integrate biochemical , epidemiological and evolutionary data . This approach allowed us to model how two viral phenotypes ( the ability to degrade PDZ containing proteins and viral oncogenic potential ) evolved . Importantly , we were able to estimate the order of evolutionary events in the emergence of these traits in HPVs . Since all tested members of the High-risk clade are able to degrade hMAGI1d , the most parsimonious explanation suggests that the putative ancestor of this clade was likewise a degrader . Indeed , the trait analysis favors the hypothesis that the ancestor of the extant high-risk viruses gained the ability to degrade PDZ-containing proteins . Notably , this putative ancestral virus was likely not oncogenic . A key result indicates that the ability to degrade PDZ-proteins is a highly unstable phenotype . This suggests that even though the initial acquisition of a type I PBM lowered viral fitness , it may have allowed for the colonization of a new ecological niche . Furthermore , since the acquisition of PDZ degradation did not coincide with the ability of viruses to cause cancer , the oncogenic types must have acquired additional phenotype ( s ) that explain their association with human cancer . We have previously reported that the ability to increase cellular hTERT , the protein subunit of telomerase , shows strong association with epidemiological classification [9] . Importantly , since the oncogenic viruses do not form a monophyletic clade , oncogenicity may represent convergent evolution . Alternatively , the non-oncogenic types within the high-risk-clade may have reduced penetrance of the oncogenic phenotype ( s ) , thereby making them less oncogenic . The present study provides evidence that all members of the High-Risk clade are able to target PDZ containing proteins for degradation . Phosphorylation of the PBM by cellular kinases modulates the ability of viral E6 proteins to recognize PDZ containing substrates . The observation that evolutionarily related E6 proteins are substrates of divergent kinases [48] , suggests that the acquisition of a PBM predates the ( convergent ) evolution of regulation through post-translational modifications . By analyzing extant E6 proteins , a previous study did not find evidence for a strict correlation between in vitro PDZ protein degradation and oncogenicity [26] . However , in order to fully understand the importance of E6/PDZ-protein interaction and degradation in the malignant process , it is important to analyze this phenotype in relationship to the evolutionary history of these viruses . By taking evolutionary relationships into account we propose a model in which the ability to degrade PDZ proteins allowed for the colonization of a new cellular niche ( e . g the cervical transformation zone ) . A role for PDZ protein degradation during the viral lifecycle is supported by the observation that mutants of HPV31 unable to interact with PDZ proteins are less fit compared to wild type viruses . Specifically , cellular proliferation , viral copy number control and other early viral functions were affected [20] . A PBM is required for long-term replication of the viral genome . Interestingly , this requirement is alleviated when p53 is removed from the cell using shRNA [53] . While , all tested members of the HR-clade were capable of degrading both p53 and hMAGI1d [8] , low-risk viruses do not show clear correlation between both phenotypes . HPV71 , a virus previously shown to degrade p53 [8] , did not affect the steady state levels of MAGI1d . Inversely , HPV40 does not degrade p53 . Therefore , the association between the ability to degrade p53 and PDZ-proteins may be less clear than previously suggested [43] . In conclusion , the ability of E6 proteins to interact with PDZ proteins allowed papillomaviruses to colonize a new ecosystem in the host . However , in order to thrive within this new environment the virus evolved additional ways to usurp the host cells’ machinery . Through interfering with normal differentiation and cell cycle control pathways , long term persistent infection may prime the cell for malignant transformation . The acquisition of hTERT promoter activation ( and/or other interactions ) by the oncogenic viruses might begin to explain why these viruses are associated with significantly higher cancer rates compared to non-oncogenic types [9] . Finally , this study highlights the importance of combining epidemiological , biochemical and evolutionary data with phylogenetic analysis in attempting to understand the relative role of specific viral phenotypes with host pathogenesis .
|
It is thought that the ability to degrade PDZ domain containing proteins is a hallmark of oncogenic papillomaviruses . However , since papillomaviruses did not evolve to be oncogenic , this hypothesis does not address the evolutionary importance of this phenotype . The present manuscript attempts to address whether HPV induced degradation of PDZ containing proteins is associated with oncogenic potential as determined by the clinical/epidemiological empirical cancer risk . Using Bayesian approaches to model trait evolution we show that it is highly unlikely for a virus to become oncogenic without first acquiring the ability to degrade PDZ proteins . Furthermore , the ability to degrade PDZ proteins allowed ancestral viruses to colonize a new cellular niche . However , in order to thrive in this new environment , these ancestral viruses had to acquire additional functions . We hypothesize that some of these additional phenotypes lead to oncogenicity . Importantly , our study illustrates the power of combining epidemiological , biochemical and evolutionary data with phylogenetic analysis in attempting to understand the relative role of specific pathogen phenotypes with host pathogenesis .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Degradation of Human PDZ-Proteins by Human Alphapapillomaviruses Represents an Evolutionary Adaptation to a Novel Cellular Niche
|
The development of biomedical interventions to reduce acquisition of HIV-1 infection remains a global priority , however their potential effectiveness is challenged by very high HIV-1 envelope diversity . Two large prophylactic trials in high incidence , clade C epidemic regions in southern Africa are imminent; passive administration of the monoclonal antibody VRC01 , and active immunization with a clade C modified RV144-like vaccines . We have created a large representative panel of C clade viruses to enable assessment of antibody responses to vaccines and natural infection in Southern Africa , and we investigated the genotypic and neutralization properties of recently transmitted clade C viruses to determine how viral diversity impacted antibody recognition . We further explore the implications of these findings for the potential effectiveness of these trials . A panel of 200 HIV-1 Envelope pseudoviruses was constructed from clade C viruses collected within the first 100 days following infection . Viruses collected pre-seroconversion were significantly more resistant to serum neutralization compared to post-seroconversion viruses ( p = 0 . 001 ) . Over 13 years of the study as the epidemic matured , HIV-1 diversified ( p = 0 . 0009 ) and became more neutralization resistant to monoclonal antibodies VRC01 , PG9 and 4E10 . When tested at therapeutic levels ( 10ug/ml ) , VRC01 only neutralized 80% of viruses in the panel , although it did exhibit potent neutralization activity against sensitive viruses ( IC50 titres of 0 . 42 μg/ml ) . The Gp120 amino acid similarity between the clade C panel and candidate C-clade vaccine protein boosts ( Ce1086 and TV1 ) was 77% , which is 8% more distant than between CRF01_AE viruses and the RV144 CRF01_AE immunogen . Furthermore , two vaccine signature sites , K169 in V2 and I307 in V3 , associated with reduced infection risk in RV144 , occurred less frequently in clade C panel viruses than in CRF01_AE viruses from Thailand . Increased resistance of pre-seroconversion viruses and evidence of antigenic drift highlights the value of using panels of very recently transmitted viruses and suggests that interventions may need to be modified over time to track the changing epidemic . Furthermore , high divergence such as that observed in the older clade C epidemic in southern Africa may impact vaccine efficacy , although the correlates of infection risk are yet to be defined in the clade C setting . Findings from this study of acute/early clade C viruses will aid vaccine development , and enable identification of new broad and potent antibodies to combat the HIV-1 C-clade epidemic in southern Africa .
The development of effective biomedical intervention strategies to prevent HIV-1 infection remains a global priority . To support these efforts , two large immunization trials in high incidence , clade C epidemic regions in southern Africa are imminent . The first , a Phase 3 efficacy trial using a vaccine similar to the one used in the RV144 trial modified to include clade C antigens will be tested to determine if the protection observed in the RV144 vaccine trial in Thailand can be replicated in this high incidence setting ( http://vaccineenterprise . org/content/P5Partnership ) . The second is a Phase 2b trial to evaluate if passive administration of the VRC01 monoclonal antibody , that targets the viral CD4 binding site ( CD4bs ) , reduces HIV-1 acquisition [1] . Both interventions rely on the induction of HIV-specific antibodies against the HIV-1 envelope glycoprotein . HIV-1 is extraordinarily diverse , and evaluation of potential coverage by these intervention strategies would therefore need to take envelope diversity into account . As there is a severe HIV-1 transmission bottleneck that may affect viral phenotype [2–7] , studies that aim to elucidate the target for active and passive immunization should ideally be done on viruses that are collected soon after transmission . Although correlates of protection from HIV-1 infection are not fully understood , both neutralizing and non-neutralizing antibodies are known to play a crucial role . The importance of neutralizing antibodies is demonstrated in non-human primate models where passive administration of broadly neutralizing antibodies ( bnAbs ) conferred complete protection from simian-human immunodeficiency virus ( SHIV ) challenge [8 , 9] . While bnAbs with extraordinary coverage have been tested for safety in HIV-1 infected and uninfected humans [1] , the VRC01 trial will be the first to evaluate the effectiveness of bnAbs for prevention . While a vaccine should ideally elicit bnAb responses , no such vaccine has been developed to date . However results from the RV144 efficacy vaccine trial in Thailand found that non-neutralizing antibodies may offer an alternative route to protection [10] . The HIV-1 envelope is comprised of several variable domains , and antibody responses directed to V1/V2 variable loops [10 , 11] , as well as to V2 and V3 linear peptides [12] , were inversely correlated with infection risk in RV144 . Genetic signatures in breakthrough infections that were associated with reduced acquisition risk were also identified [11 , 13 , 14] . The HIV-1 global pandemic is highly diverse , comprising of many different clades ( also known as genetic subtypes ) and recombinant circulating forms ( chimeras between more than one clade ) . Although neutralizing antibody responses elicited by one clade are generally effective at neutralizing viruses across clades , there is evidence of enhanced potency when clades are matched [15–19] . Even within a clade , diversity has been shown to have an effect illustrated by the fact that the clade-matched neutralization advantage was more pronounced in regions with lower viral diversity such as Thailand , compared to epidemics in southern Africa with higher viral diversity [15] . The effects of viral diversification on neutralization targets have also been seen by tracking the epidemic over time . Two cohorts in Europe , both predominantly infected with clade B , showed that HIV-1 has evolved at a population level to become more resistant to serum and bnAb neutralization over a twenty-year period [20–22] . This high diversity and increasing neutralization resistant phenotype is a potential problem for both active and passive immunization . Immunization strategies need to block the virus ( es ) that establishes infection , and it is thus important to understand the properties of these transmitted founder viruses . There have been conflicting observations in cross-sectional studies as to whether viruses collected soon after transmission are more sensitive or more resistant to neutralization compared to viruses from chronic infection . In one study acute clade B viruses tended to be more sensitive to neutralization by VRC01 , as well as pooled IgG from HIV-1 infected subjects compared to chronic viruses [5] . A separate clade C study generally detected no significant difference in neutralization susceptibility to VRC01 or sera [6] . In contrast to the clade B study , a large multi-clade study showed that early viruses were more neutralization resistant than late viruses when tested against clade-mismatched plasma , with no difference observed for clade-matched plasma [15] . Small sample size in some studies together with differences in study design and reagents all potentially confounded these studies , motivating for a large standard panel of viruses collected soon after transmission to more accurately evaluate targets of passive immunization studies and vaccines . Here , using a large panel of 200 HIV-1 clade C Env-pseudotyped viruses generated from acute/early infection , we investigated the genotypic and neutralization properties of these viruses that may impact non-neutralizing and neutralizing antibody recognition . This collection is a good representation of viruses from southern Africa , the region of the world most affected by the HIV-1 epidemic and a major region for passive and active immunization efficacy trials . This panel , representing the largest collection of clade C Env-pseudotyped viruses from acute/early infection , is a valuable resource to the field , providing a reagent set that will enable establishing the cross-reactive potential of newly isolated monoclonal antibodies and the characterization of vaccine responses in the critically important HIV-1 clade C epidemic in southern Africa .
The CAVIMC-CAVD HIV-1 Clade C Virus Neutralization Phenotype Study was reviewed and approved by the research ethics committee of the Faculty of Health Sciences of the University of Cape Town ( 168/2007; 513/2012 ) . All participants provided written informed consent for study participation . Samples used to generate functional env clones originated from Botswana ( BW , n = 6 ) , Zambia ( ZM , n = 13 ) , Malawi ( MW , n = 23 ) , Tanzania ( TZ , n = 28 ) , and South Africa ( ZA , n = 130 ) . South African samples originated from eight provinces: Western Cape ( Cape Town , ZAwc n = 12 ) , Eastern Cape ( ZAec n = 2 ) , North-West ( ZAnw n = 7 ) , Kwazulu-Natal ( ZAkzn n = 68 ) , Mpumalanga ( ZAmp n = 6 ) , Northern Cape ( ZAnc n = 3 ) , Limpopo ( ZAlp n = 5 ) and Gauteng ( Soweto and Johannesburg , ZAgp n = 27 ) ( Table 1 and S1 Table ) . Of the 200 , 199 were assumed to be sexually transmitted , and one was transmitted by breast feeding . Samples were catalogued as originating from an individual soon after transmission if they had a documented HIV-1 negative test within the previous 100 days , or if they were HIV-1 PCR positive and antibody negative at the time the sample was collected . cDNA synthesis , followed by single genome amplification of env ( ~2 . 5kb ) , was performed according to a method described previously [4 , 23] . Both env strands were directly sequenced using an ABI PRISM 3100 Genetic Analyser using BigDye terminator reagents ( Applied Biosystems , Warrington , UK ) , and sequence reads were assembled , edited and consensus sequences generated using Sequencher version 5 . 2 . 3 ( Gene Codes Corporation , Ann Arbor , MI USA ) . Amplicons used for sequencing were generated using the limiting dilution approach , and for 90% ( n = 179 ) of samples , at least five SGA derived sequences were generated per sample allowing for accurate transmitted founder consensus inference [24–26] ( S1 Table ) . For pseudovirus production , full-length rev-env cassettes were cloned into one of two mammalian expression vectors , pcDNA 3 . 1 Directional/V5-His-TOPO ( Invitrogen , Carlsbad , CA ) or pTarget ( Promega , Madison , WI ) . The amplicon sequence selected for cloning was the one closest to a participant’s consensus that was generated from at least five sequences . The resulting clones were sequenced to ensure an exact match to the original amplicon sequence and where cloned inserts differed from the parental sequence , mutagenesis was performed to ensure a match with the parental sequence . Env-pseudotyped viruses were generated by co-transfecting envelope clones with a clade B backbone pSG3ΔEnv ( NIH AIDS Research and Reference Reagent Program ) in HEK293T cells as previously described [27 , 28] . Pseudovirus functionality was determined by measuring luciferase expression after infecting TZM-bl cells ( NIH AIDS Research and Reference Reagent Program , ARRRP ) . Relative luminescence units ( RLUs ) of ≥100 , 000 were considered ideal and 30 , 000 RLUs were accepted in cases where readings were 2 . 5 x the background; <2 . 5 times the background were considered negative . Co-receptor usage was inferred by Geno2Pheno [29 , 30] and webPSSM [31 , 32] using a false positive rate ( FPR ) cut-off of 5 for Geno2pheno [33] . Viruses predicted to use CXCR4 were tested for co-receptor usage in the Trofile assay [34] . Viruses used in this entry assay were produced in an identical manner as described above except the HEK293T cells were co-transfected with env clones and the backbone , pNL4-3 . lucRΔenv ( NIH AIDS Research and Reference Reagent Program ) . Positive tropism controls RP1 . 12 , an X4 virus [35] , and QH0515 . 1 , an R5 virus [36] were used . Viruses were normalized using the in-house p24 assay and standardized quantities of Env-pseudotyped viruses were tested for their ability to infect U87 cell lines expressing CD4 and either CCR5 ( U87_R5 ) or CXCR4 ( U87_X4 ) [37] . Sequences were aligned using GeneCutter ( http://hiv . lanl . gov/content/sequence/GENE_CUTTER/cutter . html ) [38] which enables the maintenance of alignments in codon space by employing HMMER [39] and is trained specifically for HIV-1 on a full-length genome alignment . Sequences were examined for evidence of inter-clade recombination using RIP [40] , Rega HIV subtyping tool ( 400bp sliding window with 20bp steps size ) [41] and with jpHMM [42] . Only sequences that were clade C throughout the entire env reading frame were included in the panel . For phylogenetic analysis and similarity comparisons , regions with more than 5% gaps were excluded . Candidate vaccine strains , Ce1086 ( FJ444395 from Malawi in 2004 ) ; TV1 ( AF391230 from South Africa in 1998 ) ; 96ZM651 ( AF286224 from Zambia in 1996 ) ; and other clade C and clade C-related sequences from India and China ( CRF07_BC , CFR08_BC and BC unique recombinants which were predominantly clade C across env , n = 59 ) were included as references . Phylogenies were computed using FastTree [43] on nucleotide sequences using the GTR substitution model . Gp120 amino acid distance calculations for comparing vaccine coverage and relatedness were determined using the HIVb substitution model in DIVEin [44] . The sequences from the placebo arm of the RV144 study ( n = 66 ) were the same as those analyzed previously [13] . For investigation of the outward evolution of protein sequences , additional un-rooted phylogenies were computed in PhyML , version 3 . 0 [45] with the HIVb substitution model [46] . Outward evolution or divergence measured in branch length was calculated from protein phylogenies inferred in PhyML with trees rooted using the minimum sum of variance method [47] . Trees were visualized in FigTree v1 . 4 . 2 ( http://tree . bio . ed . ac . uk/software/figtree/ ) and cluster support stated as >80% of a 100 resampled replicates . Env characteristics such as loop length , glycan density and net charge were determined for both conventional variable loops as well as for hypervariable regions within variable loops using a Los Alamos HIV-1 sequence database web tool which excises and characterizes either complete variable loops or hypervariable regions ( http://www . hiv . lanl . gov/content/sequence/VAR_REG_CHAR ) . Weblogos of signature sites and linear peptide portions V2 ( HXB2 161–179 ) and V3 ( HXB2 300–322 ) , were used to show amino acid frequency by position and were performed using a web tool at the Los Alamos HIV-1 database AnalyzeAlign ( http://www . hiv . lanl . gov/content/sequence/ANALYZEALIGN/analyze_align . html ) . Signature analysis to distinguish between viruses from pre-seroconversion , indeterminate and post-seroconversion was performed using a phylogenetically corrected signature analysis strategy [48] . Fifty-four serum samples were collected from antiretroviral ( ARV ) drug-naïve , chronically HIV-1-infected individuals originating from 3 SAAVI ( South African AIDS Vaccine Initiative ) clinical trial sites in South Africa: Durban , Kwazulu-Natal ( n = 16 , CAPRISA ) ; Cape Town , Western Cape ( n = 20; Desmond Tutu HIV Foundation , DTHF ) ; and Soweto , Gauteng ( n = 18 , Perinatal HIV Research Unit , PHRU ) ( S2 Table ) . Serum samples were pre-screened for potency and breadth against three clade C pseudoviruses ( CAP8 . 6F , CAP255 . 16 and Du156 . 12 ) , a clade C consensus ( ConC ) , a clade B pseudovirus ( 6535 ) , a clade B consensus ( ConB ) and a single clade A pseudovirus ( Q23 . 17 ) . From this , a panel of 30 sera was selected , 10 per site , representing differing breadth ( selecting some low , medium and high ) for assaying against the 200 Acute/Early clade C panel ( S3 Table ) . Serum samples were collected between March 2011 and October 2013 and all were heterologous , except for two samples from Durban . All analyses were adjusted to correct for the inclusion of these autologous measurements . Neutralization of Env-pseudotyped viruses was measured using the validated TZM-bl assay . Briefly , reductions in Tat-regulated firefly luciferase ( Luc ) reporter gene expression were measured in 96-well culture plates after a single round of virus infection in TZM-bl cells as described previously [27] . Assay stocks of Env-pseudotyped viruses were produced by transfection in HEK293T cells , as described above , and titrated in TZM-bl cells . Serum samples ( inactivated at 56°C , for 1 h ) and bnAbs were tested at 1:20 dilution and then at 3-fold dilutions up to seven times in duplicate wells for each dilution starting at either 10 μg/ml ( PG9 , PGT128 ) or 50 μg/ml ( 4E10 ) or 25 μg/ml ( CAP256-VRC26 . 25 ) . VRC01 was initially tested at10 ug/ml , and was later evaluated at 50 ug/ml . Neutralization was determined as the serum dilution or bnAb concentrations at which a 50% reduction in RLU was detected compared to virus control well RLUs , reported as the 50% inhibitory dilution ( ID50 ) for serum samples and the 50% inhibitory concentrations ( IC50 ) for bnAbs . In some cases we also report IC80 values for bnAbs ( 80% reduction in RLU ) . Background luminescence read from cells-only control wells was subtracted . Positive controls consisted of a HIV-1 IgG pool purified from five HIV-1 clade C-positive plasma samples ( HIVIG-C ) and a pool of purified IgGs from HIV-1 clade B positive plasma samples ( HIVIG-B , NIH AIDS Research and Reference Reagent Program ) . bnAbs PG9 and PGT128 were provided by D . Burton ( Scripps Research Institute ) . bnAbs VRC01 and CAP256-VRC26 . 25 were provided by J . Mascola ( NIH Vaccine Research Center ) . bnAb 4E10 was commercially obtained from PolymunScientific ( Klosterneuburg , Austria ) . All assays were conducted in laboratories adhering to Good Clinical Laboratory Practice ( GCLP ) . Statistical comparisons were performed in GraphPad Prism 5 . 0 ( GraphPad Prism version 5 . 00 for Windows , GraphPad Software , San Diego California USA , www . graphpad . com ) or in RStudio , version 0 . 98 . 501 ( R Core Team . 2013 . R: a language and environment for statistical computing , http://www . r-project . org ) . Statistical significance was considered where p-values were ≤0 . 05 . Differences in viral characteristics ( neutralization susceptibility measured as ID50 ) and divergence of vaccine strains from the acute/early clade C and RV144 breakthrough viruses ) were assessed using nonparametric Mann-Whitney tests for distributions between two un-paired groups . One-sided tests were performed where existing associations exist . Correlations of variable loop properties with IC50 values , phylogenetic branch length over time and ID50 values with branch length , were tested using Kendall’s rank correlation test as implemented in the R package Kendall v2 . 2 , statistics provided for one-sided tests unless stated otherwise . Differences in neutralization sensitivity between viruses over calendar time from three different periods ( 1998–2005 , 2006–2007 and 2008–2010 ) were evaluated using a non-parametric Jonckheere-Terpstra test for trend among ordered groups ( time interval groups in this case , with time periods pre-selected based on numbers ) . We tested an increase in neutralization resistance hypothesis . Following the strategy used in Seaman et al . , [16] tier categorization was performed by grouping k-means clustering whereby the 200 C-clade viruses were assigned to one of four subgroups ( tier 1A , & B , tier 2 and tier 3 ) ranging from highly sensitive to resistant phenotypes . Subsequently , rank ordering of viruses according to their average log10 ID50 titers to further resolve tier 1B and tier 2 was performed [16 , 49] . We also tested the association between the ID50 titers and the early stage of infection , as well as the presence or absence of the glycan at position 332 using a Gaussian fixed-effect generalized model ( GLM ) with the titers as dependent variable and stage , the presence of the 332 glycan and year as independent variables . Listed in S1 Table .
To generate a panel of 200 functional clade C envelope ( env gene ) clones from viruses collected soon after transmission , plasma samples were obtained from individuals estimated to be infected for less than a 100 days: 67 individuals were HIV-1 PCR positive but HIV-1 seronegative ( pre-seroconversion , or Ab- ) [50]; 29 were in early seroconversion with an indeterminate western blot ( indeterminate , or Ab+/- ) ; and the remaining 104 were HIV-1 seropositive with a negative diagnosis within the previous 100 days ( post-seroconversion , or Ab+ ) ( S1 Table ) . Samples originated from five southern African countries ( South Africa , Botswana , Zambia , Tanzania and Malawi ) ( Table 1 ) . Phylogenetic analysis showed that the viruses generally did not cluster according to country of origin , suggesting a generalized regional epidemic with intermixing of strains ( Fig 1 ) , although African and Asian C clade sequences form separate clades ( S1 Fig ) . One bootstrap-supported cluster of four sequences , and two clusters of two sequences each , were identified from Tanzania; and seven clusters of two sequences each were identified from South Africa . The majority of South African sequences ( 68/130 ) originated from one particular region ( KwaZulu Natal ) . We found no strong bootstrap-supported evidence of local founder effects within South Africa , although some clades consisting of sequences only of South African origin were evident , with one smaller clade of 10/68 sequences originated in Kwazulu-Natal observed ( highlighted in Fig 1 ) . S1 Fig provides a more detailed breakdown of all geographic origins within South Africa , and includes the vaccine strains that will be used in the upcoming trial , which are located in three phylogenetically distinct parts of the tree ( S1 Fig ) . Transmitted viruses almost exclusively utilize the co-receptor CCR5 [3 , 5 , 6] . In order to determine co-receptor phenotype in the clade C panel we inferred viral co-receptor usage based on the V3 loop amino acid sequence using two genotyping methods [29–33] . Of the 200 viruses , 192 were predicted to use CCR5 while eight were predicted as CXCR4-using by both methods ( S4 Table ) . Experimental analysis using U87 cells expressing either the CXCR4 or CCR5 co-receptor [34] , however found all eight to be exclusively CCR5-using with no evidence for dual tropism . Thus , all 200 clade C viruses were considered to be CCR5-tropic . HIV-1 tier classification is a useful way to define a virus’ neutralization susceptibility profile , with tier 1 viruses being highly sensitive and tier 3 highly resistant to neutralization [16 , 49] . Neutralization assays were performed using 30 clade C HIV-1+ serum samples from chronic infection , which were prescreened to reflect sera with a range of neutralization potency and breadth ( S2 and S3 Tables ) . Following a procedure published by Seaman et al ( 2010 ) [16] , Envs were grouped using k-means clustering , and 4 clusters each representing a group of viruses with similar patterns of sensitivity were identified ( Fig 2 ) . This was used for an initial tier classification , with tier 1A , 1B and tier 3 forming highly significant and robust clusters , and tier 2 capturing everything between . This classification was recapitulated when neutralization sensitivity of each of the 200 viruses was measured as the geometric mean ID50 titer ( GMT ) determined for all sera against which each virus was tested ( Fig 3A ) . The higher the GMT relative to other viruses in the panel , the more neutralization sensitive the virus . The rank of the titers closely reflected the tiers assigned by k-means clustering , although there was a small amount of intermixing between tier 2 and 3 . The way Envs are grouped in a k means is generally highly dependent on the sera used for the evaluation , as is evident by the large number of Envs that sometimes are grouped with a more sensitive , sometimes with the more resistant cluster , depending on resampling . Consistent with previous approaches [16] , the ambiguous calls that were not consistently highly sensitive in the bootstrap ( tier 1 ) , or consistently highly resistant ( tier 3 ) , were classified as tier 2 . Fig 3A illustrates the continuum of geometric means . The majority of viruses exhibited a moderately resistant tier 2 phenotype ( 75% , n = 150 ) , whereas 8 . 5% ( n = 17 ) were classified as possessing a more sensitive tier 1B phenotype , and 15 . 5% ( n = 31 ) classified as possessing a more resistant tier 3 phenotype ( Fig 3A ) . Phylogenetic analysis showed that the most sensitive or most resistant viruses generally did not cluster together ( Fig 3B ) . To further interrogate if there were signatures in the viral sequences associated with tier phenotype , we considered every amino acid in each position using either a simple Fisher’s exact test testing for the association between the presence and absence of the amino acid and the phenotype of interest , and a phylogenetically corrected Fisher’s test ( described fully in Gnanakaran et al , 2011 ) [51] . We identified no phylogenetic signatures associated with tier phenotype based on a false discovery rate ( FDR ) threshold of q < 0 . 2 , however , using a simple Fisher’s test , a single association with tier 3 was identified , K683R . Lys ( K ) is the most common amino acid in this position . Arg ( R ) was enriched among Tier 3 viruses , found 16/30 times ( 53% ) , while among Tier 1 or 2 viruses it was present only 28/170 times ( 16% ) , ( p-value of 0 . 00004 , with a q = 0 . 04 for the full set of signature comparisons ) . This amino acid is located in the membrane-proximal external region ( MPER ) embedded in the 4E10 bnAb epitope , and the enrichment of R in Tier 3 viruses suggests that this position and region may be implicated in neutralization resistance . The neutralization distribution of this clade C panel roughly approximated the tier distribution of a multi-clade panel of Env-pseudotyped viruses previously described ( 19% , 66% and 15% for tiers 1B , 2 and 3 respectively ) [16] . Unique to our study was the identification of 1% ( n = 2 ) of viruses with a highly sensitive tier 1A phenotype . This highly neutralization sensitive phenotype is not normally associated with circulating viruses . Both tier 1A envelope sequences matched a single genome derived amplicon , suggesting that these highly neutralization sensitive viruses do occur in vivo , although it remains possible that they were the result of an artificially introduced error during amplification , as both had a single mutation away from the derived transmitted founder sequence: the CH0505 . w4 . 3 clone has a W680G mutation in the MPER [52] and the SO032_A2 . 8–1 clone had an E406G mutation in V4 . We therefore do not know if these viruses were transmitted or if they evolved post-infection , or may have been the result of PCR error and we thus excluded them in all analysis pertaining to neutralization sensitivity . There are conflicting data on whether viruses sampled in early infection are more or less sensitive to serum neutralization compared to viruses sampled later in infection . We investigated whether , over the first 100 days of infection , we could detect any difference in neutralization sensitivity . To select viruses that best reflected the virus in vivo that had recently traversed the transmission bottleneck , we included only envelopes generated using the single genome amplification ( SGA ) approach [23]; and excluded viruses that were classified as multi-variant transmission or for which there was insufficient information to classify their multiplicity of infection ( S1 Table ) . This subset of 139 Env-pseudotyped viruses was classified into three groups according to time of infection where sequential gain of HIV-1 antibody responses was used as a marker of time from infection [50]: pre-seroconversion group ( n = 58 ) ( no detectable antibodies Ab-; estimated infection <15 days ) ; indeterminate group ( n = 26 ) ( evolving antibody responses Ab+/- , estimated infection <37 days ) ; and early post-seroconversion group ( n = 55 ) ( near/full seroconversion Ab+; estimated duration of infection < 100 days ) . We found that pre-seroconversion viruses were more resistant to neutralization by the 30 clade C sera compared to post-seroconversion viruses , with pre-seroconversion viruses having a significantly lower GMT ( p = 0 . 001 , one-sided Wilcoxon rank sum test ) ( Fig 4A ) . There was no significant difference in GMT between the indeterminate group and the post-seroconversion group . To confirm that the result was real , given the possibility of error in measurements , we built an error model based on the many duplicated neutralization data points available [53] . We randomly resampled from the Gaussian distribution based on the observed error to add noise to our individual points , and generated 10 , 000 replicate datasets using the uncertainty model . For both comparisons , acute ( Ab- ) versus early ( Ab+ ) or indeterminate ( Ab +/- ) , more than 97% of replicates show similar trends as Fig 4A , i . e . the acute group is less sensitive than both indeterminate and early groups . Furthermore , using 10 , 000 replicates of artificial ID50 data generated as above , we also found that the acute group ( Ab- ) had higher normalized sum of ranks from the Wilcoxon rank sum test statistic than the indeterminate group ( Ab+/- ) in 99 . 8% replicates , and than the early group ( Ab+ ) in 100% replicates ( p < 10−4 ) . This suggests that the results shown in Fig 4A are robust under realistic models of experimental uncertainty . We also found that the pre-seroconversion viruses were neutralized less frequently by 30 clade C sera compared to the post seroconversion viruses ( median breadth of 36% versus 50%; p = 0 . 0474; two sided Mann-Whitney test ) ( Fig 4B ) . This was largely driven by the resistant viruses as when only neutralization sensitive viruses were analyzed , there was no significant difference in sensitivity between pre-seroconversion , indeterminate and post-seroconversion viruses ( median IC50 85 . 90 μg/ml , 89 . 74 μg/ml and 95 . 06 μg/ml ) ( S8 Fig ) . This behavior was confirmed with purified IgGs from clade C infections ( HIVIG-C ) , however no difference was observed when assayed against pooled purified IgGs collected from clade B infections ( HIVIG-B ) ( S2 Fig ) , suggesting a possible clade-associated specificity may be driving this differential resistance pattern . Taken together , these data suggest that viruses present prior to seroconversion are less likely to be neutralized by clade-matched polyclonal HIV-1 sera than those viruses present post-seroconversion . Several studies have shown that increases in length and glycosylation density across the V1 , V2 and V4 regions of gp120 are associated with neutralization resistance [15 , 20 , 54–56] . We investigated whether these features could account for observed differences in neutralization sensitivity seen in the 200 clade C viruses . Similar to previous studies , when we analyzed all viruses together , we found that longer V1 , V2 and V4 length , and higher glycan density were strongly associated with neutralization resistance ( S3 Fig ) . However , we found no significant difference in V1 , V2 and V4 loop length or glycan density between the pre-seroconversion , indeterminate and post-seroconversion groups of viruses ( Fig 4C and 4D ) , indicating that observed differences in neutralization phenotype between these groups were not attributable to differences in variable loop length or glycan density . We then investigated whether differences in neutralization phenotype could be attributable to differences in particular antibody specificities . Viruses were assayed with bnAbs targeting: the V2-glycan site ( CAP256-VRC26 . 25 and PG9 ) , the gp41 MPER epitope ( 4E10 ) , the CD4 binding site ( VRC01 ) , and the V3/C3 glycan supersite ( PGT128 ) . Here IC50 was used , where low concentrations indicate increased viral sensitivity . For antibodies targeting the CD4bs , V2-glycan and MPER specificities we found no difference in neutralization susceptibility between pre-seroconversion and post-seroconversion viruses , suggesting that these epitope targets did not differ between the groups . However , pre-seroconversion viruses were significantly more resistant to neutralization by PGT128 compared to the post-seroconversion viruses ( median IC50 of 4 . 42 μg/ml compared to 0 . 06 μg/ml respectively , p = 0 . 0174 , Mann-Whitney two sided test ) ( Fig 5A ) , with 50% of pre-seroconversion and 27% of post-seroconversion viruses resistant to this bnAb . PGT128 specificity is largely dependent on the presence of a glycan at position 332 [58 , 59] , and similar to what we previously reported [59] , there was an underrepresentation of this glycan in pre-seroconversion clade C viruses ( Fig 5B ) . To determine the contribution of the 332 glycan in influencing neutralization resistance to polyclonal sera , we divided the 139 viruses into three groups: those with the glycan at position 332 ( N332+ ) , those with the glycan at position 334 ( N334+ ) ( the glycan at position 332 shifts to position 334 in some cases ) and those with no glycan at either position 332 or 334 ( N332-/N334- ) . Similar to previous studies , we found that viruses with the 332 glycan were indeed significantly more sensitive to serum neutralization compared to viruses lacking this glycan ( median ID50 of 48 . 98 μg/ml , 35 . 73 μg/ml and 41 . 30 μg/ml for N332+ , N334+ and N332-/N334- viruses respectively , p = 0 . 0085; Mann-Whitney two sided test ) ( Fig 5C ) . However , when we considered only viruses containing the 332 glycan , the pre-seroconversion viruses still had significantly lower sensitivity to serum neutralization ( median ID50 of 34 . 36 μg/ml , 62 . 52 μg/ml , and 56 . 36 μg/ml , for Ab- , for Ab+/- , Ab+ respectively , p = 0 . 0219 ) , indicating that the lack of the glycan at position 332 is not the sole determinant of increased resistance of pre-seroconversion viruses ( Fig 5D ) . We then fitted a Gaussian fixed-effect generalized model ( GLM ) to test for an association between GMT with infection stage and the 332 glycan . Partitioning viruses into pre-seroconversion ( Ab- ) and indeterminate with post-seroconversion ( Ab+/- and Ab+ ) , the model showed that both the presence/absence of the 332 glycan and the stage of infection reliably predicted GMT , without any interaction between infection stage and glycosylation state ( S4 Fig ) . The log10 GMT was on average 1 . 36 fold higher when the 332 glycan was present ( p = 0 . 0094 ) , and 0 . 72 lower when the sample was pre-seroconversion ( p = 0 . 0034 ) . PGT128-like antibodies have been shown to recognize other glycans centered around the 332 glycan , the so-called high-mannose patch , which include glycans , N136/7 , N156 , N295 , N301 and N334 [57] . N156 and N301 are conserved in clade C viruses and are therefore unlikely to contribute to differences in neutralization phenotype . N295 has been reported to substitute for N332 [60] , but was found in higher frequencies in pre-seroconversion viruses and thus unlikely to be playing a role here . This together with N136/7 , which occurs at similar frequencies in both pre-seroconversion and post-seroconversion groups , this could not account for increased neutralization resistance in the pre-seroconversion viruses ( Fig 5B ) . We sought to determine whether other genotypes were statistically associated with pre-seroconversion viruses . We identified a site in V5 , G464 , which was present in all post-seroconversion viruses in our panel , but in only 80% of pre-seroconversion/indeterminate viruses ( p = 0 . 00024 , q = 0 . 096; Fisher’s exact test ) . This site is located in the CD4 binding domain , however the relevance of this finding is not yet clear . As this large collection of clade C viruses from soon after transmission was collected over 13 years , we sought to determine whether the southern African epidemic has increased in resistance to neutralization over time . For this we tested serum collected from clade C infected individuals between 2011 and 2013 against the clade C panel . We first investigated whether viruses diverged over this period , by determining Gp160 phylogenetic tree branch length ( measured as distance from root ) from 1998 to 2010 . We found that early viruses ( warmer colors ) displayed shorter branches ( closer to the root ) compared to later viruses ( cooler colors ) ( Fig 6A ) . Even when measuring divergence conservatively by determining protein distances on alignments excluding all hypervariable regions , we found a significant positive correlation of branch length over time ( one sided Kendall’s τ = 0 . 157 , p = 0 . 0009 ) ( Fig 6B ) indicating that the clade C epidemic diversified appreciably over the 13 year time period . We also observed a significant negative correlation of branch length with serum neutralization sensitivity , as measured by geometric mean ID50 titer ( two sided Kendall’s τ = -0 . 141 , p = 0 . 0015 ) ( Fig 6C ) , but could find no direct correlation of polyclonal serum neutralization sensitivity over time ( two sided Kendall’s τ = -0 . 017 , p = 0 . 7338 ) ( S5A Fig ) . We also found no significant association between variable loop length and glycan density over time ( S5B and S5C Fig ) , suggesting that this trait has remained relatively constant . Taken together these data provides indirect evidence of diminishing sensitivity to within-clade sera over time . We then evaluated changes in virus susceptibility to known bnAbs . Similar to polyclonal serum , we identified a significant negative association of branch length with bnAb neutralization sensitivity ( measured in geometric mean IC50 titer for all bnAbs in aggregate ) ( Kendall’s τ = 0 . 1157 , p = 0 . 0075 ) ( S6A Fig ) . Viruses were grouped into three time periods according to date of collection to generate three groups of relevant sample size ( 1998–2005 , 2006–2007 and 2008–2010 ) . Although the first time period spanned 8 years , the majority of samples ( >73% ) were collected over a three-year period from 2003–2005 . When we evaluated each of the five bnAbs individually , we found increased resistance as measured over three time periods to PG9 , VRC01 and 4E10 ( p = 0 . 013 , p = 0 . 030 and p = 0 . 004 respectively , Jonckheere-Terpstra test ) , but not to PGT128 or CAP256-VRC26 . 25 ( Fig 6D ) . Thus , we demonstrated that viruses are becoming more neutralization resistant to certain bnAbs as the epidemic progresses . The role of VRC01 in preventing HIV acquisition is being evaluated in a Phase 2b study in Africa ( ClinicalTrials . gov Identifier: NCT02568215 ) , where trial participants will receive an intravenous infusion of VRC01 at a dose of 10 mg/kg or 30 mg/kg every 8 weeks . As VRC01 serum concentrations will decay over time between transfusions , we were interested in determining the effectiveness of this antibody at different concentrations including , 50 ug/ml , 10 ug/ml and 1 ug/ml . At these concentrations , 84% , 80% and 56% of the panel were neutralized respectively ( IC50 ) ( Fig 7 ) . Assayed with the same concentration range , but measuring the concentration that resulted in 80% inhibition ( IC80 ) , VRC01 neutralized 78% , 68% and 30% of viruses , respectively . The median IC50 titer against sensitive viruses at <10 ug/ml was 0 . 42 μg/ml . We have shown that viruses are becoming increasingly resistant to VRC01 as the epidemic matures , where overall potency at <10 ug/ml was shown to decrease from a median IC50 titer of 0 . 48 μg/ml to 0 . 69 μg/ml to 0 . 84 μg/ml over three time periods measured ( 1998–2005 , 2006–2007 and 2008–2010 ) ( Fig 6D ) . This was observed despite VRC01 breadth ( percentage viruses neutralized at IC50 < 10 μg/ml ) staying relatively constant over this time period ( 77% , 78% and 79% respectively ) . The RV144 trial in Thailand showed 31 . 2% vaccine efficacy with immune correlates of risk identified as antibodies predominantly against V1V2 region of gp120 [10] . Building on the success of RV144 , a Phase 3 trial is planned for South Africa using vaccines similar to those used in RV144 , however the canarypox vector has been modified to express the clade C ( 96ZM651 ) gp120 and the protein boost now comprises two clade C gp120 Env proteins ( TV1 and Ce1086 ) . To get insight into how such a vaccine will perform in the South African setting , we genotypically compared the clade C viruses to candidate C vaccines , and Thailand CRF01_AE breakthrough viruses to the AE gp120 immunogen used in the RV144 trial . Phylogenetic comparison of the three clade C candidate vaccine strains showed that they fall within the 200 clade C panel sequences from southern Africa and cluster separately from clade C viruses sampled in India and China and recombinant CRF07_BC and CRF08_BC viruses from China ( S1 Fig ) , suggesting they are broadly representative of the southern African epidemic . The mean gp120 amino acid sequence distance between the panel viruses and clade C candidate vaccine protein boost sequences ranged from 22 . 19% ( 95% CI of mean , 21 . 90%–22 . 48% ) for Ce1086 to 24 . 15% ( 95% CI of mean , 23 . 85%–24 . 44% ) for TV1 . This was significantly greater distances than that observed in RV144 , where the mean gp120 amino acid sequence distance between the CRF01_AE viruses from breakthrough infections in the placebo arm to the clade-matched vaccine immunogen , CM244 was 15 . 24% ( 95% CI 14 . 69%–15 . 79% ) ( Ce1086 and TV1 individually both to CM244 p<0 . 0001 ) ( Fig 8A ) . Taken together the mean gp120 amino acid sequence distance of Ce1086 and TV1 to panel viruses was 23 . 14% ( 95% CI of mean 22 . 91%–23 . 37% ) , which was 8% more distant when compared to distances between CRF01_AE viruses and the RV144 vaccine immunogens . Binding antibodies to 19-mer V2 ( HXB2 161–179 ) and 21-mer V3 ( HXB2 300–322 ) linear peptides ( so-called “hotspot” regions ) were correlated with reduced risk of infection in RV144 [12] . We were interested in estimating the potential coverage of vaccines by determining the protein distance across V2 and V3 linear peptides between the C-clade vaccine boost proteins ( TV1 and Ce1086 ) and the C-clade panel viruses . We found high mean V2 amino acid distances of 40 . 51% ( 95% CI 39 . 13%–41 . 89% ) , with lower mean V3 amino acid distance of 17 . 10% ( 95% CI 16 . 20%–18 . 00% ) ( Fig 8B ) . Considering amino acids that were conserved in at least 90% of viruses across these hotspot regions , only 53% ( 10/19 ) of the V2 residues were conserved compared to 62% ( 13/21 ) in the V3 region ( Fig 8C ) . We found that mean distances between the V2 hotspot region and viruses from the RV144 placebo group was significantly lower than in the clade C epidemic: 19 . 63% ( 95% CI 17 . 38%–21 . 88% ) compared to 40 . 51% ( 95% CI 39 . 13%– 41 . 89% ) respectively ( p < 0 . 0001 ) , but similar distance were observed in V3 , 19 . 22% ( 95% CI 16 . 78%– 21 . 66% ) compared to 17 . 10% ( 95% CI 16 . 20%–18 . 00% ) , Thailand and South Africa respectively ) ( Fig 8B ) . We cannot be sure however what the significance of this is , whether the absence of divergence in hotspots important in the RV144 trial will be critical for clade C . We have shown that the clade C epidemic is diverging over time and we were interested to see what effect this had on C-clade vaccines distances to clade C viruses across V2 and V3 hotspot regions . We found no significant difference between distances to 70 viruses collected after 2008 and the 130 remaining viruses collected during the preceding decade ( S7 Fig ) , indicating that V2 and V3 hotspot regions are not diverging significantly over the period measured . In the RV144 genetic sieve analysis , which compared sequences in the vaccine and placebo arm of breakthrough infections to the vaccine strains , a number of signature sites associated with reduced infection risk were identified [11 , 13 , 14] . Two of these genetic signatures ( K169; I307 ) present in the RV144 vaccine strain were associated with increased vaccine efficacy against breakthrough viruses with matching amino acid residues at these positions , and have supportive experimental data to show that mutations found in vaccine breakthrough infections were associated with decreased antibody binding [11 , 13] . The protective signature K169 was present in both TV1 and Ce1086 , whereas I307 was entirely absent . We determined the frequency of these protective signatures in our panel of clade C viruses , and compared them to the frequency in 66 viruses from the placebo arm of RV144 ( Fig 8D ) . Both genetic signatures were found at a lower frequency in clade C viruses compared to RV144: the K169 site was found in 69% ( 95% CI 62%-75% ) of clade C viruses compared to 86% ( 95% CI 78%–95% ) of CRF01_AE viruses , and the I307 was found in 60% ( 95% CI 53%-67% ) of clade C and 67% ( 95% CI 55%-78% ) of CRF01_AE viruses respectively ( Fig 8D ) . The variation in these sites in clade C is illustrated by a sequence logo ( Fig 8C ) . Two sites have been identified where a mismatch to the vaccine was associated with reduced risk of infection ( I181X; F317X ) . The protective signature I181X was present in Ce1086 but absent in TV1 , whereas neither Ce1086 nor TV1 contained the protective signature F317X . I181X was found at a similar frequency in clade C viruses compared to RV144 CRF01_AE viruses ( 33% , 95% CI 26%–40% compared to 27% , 95% CI 16%–38% respectively ) whereas F317X was found at a lower frequency ( 5% , 95% CI 2%–8% compared to 20% , 95% CI 10%–30% respectively ) .
The extraordinary diversity of HIV-1 is a barrier to achieving protection in active and passive immunization studies . Several large prophylactic trials in clade C epidemic regions of Africa that rely on antibody-mediated protection are imminent , including passive immunization with the broadly neutralizing antibody ( bnAb ) VRC01 , and a vaccine efficacy study of the RV144 regimen tailored for clade C . Using a large panel of 200 clade C pseudotyped viruses from acute/early infection , we observed that pre-seroconversion viruses were more resistant to antibody neutralization compared to post-seroconversion viruses . Additionally , we provide evidence of antigenic drift in certain bnAb targets including VRC01 , which we estimated will only block ~80% of clade C viruses at its most efficacious dose . Furthermore , the higher divergence of clade C viruses from candidate clade C vaccines , compared to CRF01_AE viruses to the clade matched vaccine used in the RV144 vaccine , may make protection harder to achieve in clade C epidemic regions . Our study therefore provides a comprehensive analysis of viral traits that affect antibody recognition and interrogates how planned clinical trials may perform in the clade C epidemic of southern Africa . Studies in non-human primate models have shown that vaccines have a limited window within which they can block infection , making the properties of pre-seroconversion viruses highly relevant [61] . The increased resistance of pre-seroconversion viruses seen in our study was only detected when viruses were assayed with clade C ( and not clade B ) polyclonal sera , suggesting clade-specific epitopes are either shielded or absent in these viruses . We propose that an under-representation of the N332 glycan ( critical for the PGT128-like bnAbs ) , shown here and previously [59] , is partially responsible for the more resistant phenotype . This is supported by the fact that antibody responses dependent on the N332 glycan are common in clade C sera [62] . The lack of the N332 glycan however was not the sole determinant of this resistant phenotype , as we still observed this effect even when only including viruses with the N332 glycan . Interestingly , an analysis using a large multi-clade panel of 219 viruses , tested against 170 polyclonal sera , also found viruses collected early in infection to be less sensitive to neutralization by polyclonal HIV-1 sera [15] , suggesting that this phenomenon is more generalizable . However , the signal was only detected for inter-clade comparisons in Hraber et al . , [15] , which may have been due in part to a smaller sample size for within clade comparisons , and/or differences in the classification of viruses according to infection stage . The reasons why pre-seroconversion viruses have a more resistant phenotype are unclear . It is possible that these traits provide the virus with a competitive advantage , either at the level of transmission , or in the early establishment of infection in the new host . Another possibility is that following transmission , and up to seroconversion replication in the absence of neutralizing antibodies , similar to cultivation in vitro [63] , lead to adaptations that result in reduced shielding and thus the evolution of a more sensitive neutralization phenotype . Other studies examining viral genotypic properties shortly after transmission have also observed rapid adaptation with initial shortening and subsequent elongation of envelope variable loops as HIV-1 antibody responses evolve [64] . If this viral property is advantageous to establishing clinical infection , it may enhance transmission risk in a scenario where transmission predominantly occurs from donors who are in the acute/early stage of infection , which is thought to occur in roughly 50% of cases [65 , 66] . It will nevertheless be important to further elucidate the factors responsible for the increased resistance of pre-seroconversion viruses , to ensure that vaccines are designed to elicit antibodies that target the most vulnerable bnAb sites . Persistent cycles of immune pressure have resulted in an accumulation of HIV-1 escape mutations at a population level . Recently , two studies , in predominantly clade B European cohorts , have shown that HIV-1 is also becoming increasingly neutralization resistant over time [20–22] . In our study we show for the first time that this is also occurring in the clade C heterosexual epidemic in southern Africa , where viruses are becoming increasingly resistant to some bnAbs , including those targeting the CD4bs ( VRC01 ) , V2-glycan ( PG9-like but not CAP256-VRC26-like epitopes ) and MPER ( 4E10 ) . While Bunnick et al . [20] found that the viruses were increasing in loop length and glycan density as the epidemic progresses , similar to Bouvin-Pley et al , [21] , we did not observe this genotypic change in our cohort . However , our study , as well as many others , have found these properties to be associated with neutralization resistance [2 , 15 , 55 , 67–69] . It is possible we did not detect significant changes in these properties due to our shorter sampling period of 13 years , as compared to a period of 20 years for other studies . These observations nevertheless , suggest that vaccine strains may need to be updated to be more representative of circulating diversity . It was concerning to note that approximately 20% of clade C viruses in this panel were resistant to VRC01 and that they have become substantially less sensitive to this bnAb as the epidemic has matured , with an observed nearly doubling of concentration needed to reach 50% inhibition over thirteen years ( Fig 6D ) . Thus , if antibodies are to be an effective prevention modality , they will need to have increased neutralization breadth , so as to reduce the impact of natural resistance . Furthermore , it may be necessary to use a mixture of antibodies that target multiple sites on the virus to both increase coverage and curb viral escape routes . This strategy is a focus of many research programs and there are several newer generation antibodies , and antibody combinations , currently in clinical development [9 , 70 , 71] . Utilizing this same virus panel with 15 bnAbs that target four regions ( CD4 binding site , V1V2-glycan region , V3-glycan region and MPER ) , a comprehensive evaluation was performed to identify the best-in-class single antibodies , together with optimal combination of antibodies for HIV prevention and treatment [72] . The clade C version of the RV144 vaccine regimen is currently being tested for safety and immunogenicity , with Phase 3 trials due to start in South Africa in 2016 . We analyzed envelope sequences at various sites or regions relevant to the correlates of risk in the RV144 vaccine trial , in order to consider how this vaccine may perform in the South African clade C epidemic relative to the vaccine used in the partially efficacious RV144 vaccine trial in Thailand . We found that the clade C gp120 protein boosts ( Ce1086 and TV1 ) were generally representative of currently circulating viruses , however protein pairwise distance scores between the viruses and candidate vaccine boosts were approximately 8% greater than between RV144 vaccine immunogens and breakthrough viruses . Together with the low frequency of the RV144 signature sites , this may make vaccine protection harder to achieve in southern Africa compared to Thailand , a suggestion previously proposed by Hraber et al . based on neutralization properties [15] . However , in RV144 the major immune correlate of risk was V2 binding antibody responses , and while these responses were clade sensitive , they were found to be cross-reactive with clade C V2 peptides , suggesting that high diversity in V2 may have a limited impact [11 , 12] . Together these findings have important implications for future vaccine development although a major caveat associated with this conclusion is that the proposed correlates of risk identified in RV144 may be different in a clade C setting . This clade C panel of acute/early Env pseudotyped viruses is the largest collection of functional clade C envelope clones available , and as such , provides a unique resource to the field . The samples are geographically representative of the southern African epidemic , and all samples were from acute/early infections . Together this makes them valuable reagents to support HIV-1 vaccine and passive immunization clinical trials , and important reagents for characterization of the breadth and potency of newly isolated bnAbs . However , one limitation of this work was that , due to the difficulty in obtaining very early samples , this study took many years to accumulate sufficient sample numbers , and only 63% of the panel was relatively current ( obtained between 2006–2010 ) . In conclusion , elucidation of viral traits associated with resistance to antibody responses , and how these change over time , will be important to inform vaccine design efforts . This panel will be used to select representative viruses for evaluation of nAb responses in vaccine trials conducted in southern Africa . Furthermore , it has been used extensively to inform passive immunization studies and to prioritize bNAbs for clinical testing in clade C populations . This study has highlighted aspects that are highly relevant to vaccine design and provides insight into how efficacy vaccine trials , currently underway or imminent , will fare in this region of the world .
|
Vaccine and passive immunization prophylactic trials that rely on antibody-mediated protection are planned for HIV-1 clade C epidemic regions of southern Africa , which have amongst the highest HIV-1 incidences globally . This includes a phase 2b trial of passively administered monoclonal antibody , VRC01; as well as a phase 3 trial using the clade C modified version of the partially efficacious RV144 vaccine . The extraordinary diversity of HIV-1 poses a major obstacle to these interventions , and our study aimed to determine the implications of viral diversity on antibody recognition . Investigations using our panel of very early viruses augment current knowledge of vulnerable targets on transmitted viruses for vaccine design and passive immunization studies . Evidence of antigenic drift with viruses becoming more resistant over time suggests that these prevention modalities will need to be updated over time and that combinations of antibodies will be necessary to achieve coverage in passive immunization studies . We further show that it may be more difficult to obtain protection in the genetically diverse clade C epidemic compared to RV144 where the epidemic is less diverse , although it should be noted that the correlates of infection risk are yet to be defined in the clade C setting .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"sequencing",
"techniques",
"viral",
"vaccines",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"immunology",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"vaccines",
"preventive",
"medicine",
"rna",
"viruses",
"phylogenetic",
"analysis",
"molecular",
"biology",
"techniques",
"antibodies",
"vaccination",
"and",
"immunization",
"antibody",
"response",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"immune",
"system",
"proteins",
"sequence",
"analysis",
"proteins",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"hiv-1",
"molecular",
"biology",
"immune",
"response",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"biochemistry",
"virology",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"lentivirus",
"organisms"
] |
2016
|
Features of Recently Transmitted HIV-1 Clade C Viruses that Impact Antibody Recognition: Implications for Active and Passive Immunization
|
How non-coding DNA gives rise to new protein-coding genes ( de novo genes ) is not well understood . Recent work has revealed the origins and functions of a few de novo genes , but common principles governing the evolution or biological roles of these genes are unknown . To better define these principles , we performed a parallel analysis of the evolution and function of six putatively protein-coding de novo genes described in Drosophila melanogaster . Reconstruction of the transcriptional history of de novo genes shows that two de novo genes emerged from novel long non-coding RNAs that arose at least 5 MY prior to evolution of an open reading frame . In contrast , four other de novo genes evolved a translated open reading frame and transcription within the same evolutionary interval suggesting that nascent open reading frames ( proto-ORFs ) , while not required , can contribute to the emergence of a new de novo gene . However , none of the genes arose from proto-ORFs that existed long before expression evolved . Sequence and structural evolution of de novo genes was rapid compared to nearby genes and the structural complexity of de novo genes steadily increases over evolutionary time . Despite the fact that these genes are transcribed at a higher level in males than females , and are most strongly expressed in testes , RNAi experiments show that most of these genes are essential in both sexes during metamorphosis . This lethality suggests that protein coding de novo genes in Drosophila quickly become functionally important .
Most new genes arise from the duplication or rearrangement - in whole or in part - of existing genes [1] , [2] . These new genes are typically structurally and functionally similar to their progenitors . In contrast , protein-coding genes may also evolve de novo from previously non-coding sequences , making them lineage-specific and unlike any existing protein . De novo genes were once thought to be vanishingly rare , or even impossible [3] . Subsequent work suggests instead that these brand-new genes may make up a significant proportion of novel genes and that some have important functions . The first experimental evidence of de novo genes in Drosophila came from studies identifying a handful of protein-coding genes apparently specific to the D . melanogaster [4] and D . yakuba [5] , [6] lineages respectively . Analysis of multiple genomes in Drosophila had previously indicated that intergenic DNA contained abundant protein-coding potential [7] , but many strongly predicted genes were not functional [8] . The early de novo gene papers identified proteins that were lineage-specific and were also were stably expressed in a specific tissue ( the testis ) . Because most functional genes were believed at that time to produce proteins , these early efforts focused on the de novo emergence of proteins from regions lacking that ORF ancestrally . Genes that had high similarity hits in close relatives were excluded , though conservation of synteny was required [4] . This prevented mischaracterizing novel genes that arose through some other mechanism - such as duplications of functional exons - as de novo evolved . A similar strategy was later used to identify de novo protein coding genes in yeast [9] and mammals [10] . In contrast to Drosophila , work focused on humans identified genes with high similarity matches in the comparison species coupled with a lineage-specific loss of a mutation disabling the open reading frame ( e . g . de novo proteins ) [11] , [12] . Regardless of the detection strategy used , the early work focused on the evolution of a novel protein from DNA sequence thought to be non-coding , and the evolution of lineage-specific transcription was largely ignored . As the increasing importance of non-coding RNA genes became broadly recognized , efforts to identify de novo evolution of non-coding RNA genes began . Heinen and colleagues [13] identified a case of novel transcription from a previously untranscribed region in mice . This novel transcript did contain an ORF , but the researchers argued that the short peptide encoded was unlikely to be functional . More recently , some human de novo proteins were found to have likely arisen from previously transcribed non-coding RNA sequences [12] , implying that the evolution of a de novo protein may occur either before or after transcription of a previously non-coding region begins . What is clear is that for a protein-coding gene to arise de novo it must evolve both transcriptional and protein-coding potential . In principle , these events could occur in either order ( Figure 1B ) . If a new open reading frame ( ORF ) evolves within a transcribed region ( such as a non-coding RNA ) , it is more likely to ultimately be translated than an ORF that evolves in a region of untranscribed DNA ( Figure 1B left ) . Alternatively , an ORF may exist in the ancestral state , but not be expressed until transcription is initiated through acquisition of regulatory machinery ( Figure 1B right ) . In either case , ORFs may subsequently expand through loss of stop codons and/or exon gain . These models are not mutually exclusive and intermediate models have been proposed – for example , occasional read-through transcription of genes [4] , translation of small ORFs from non-coding RNA , or other partial gene states are expected to occur commonly . Indeed , both Yeast [14] and Drosophila [15] contain hundreds of these “proto-genes” which may subsequently evolve into de novo protein coding genes . Despite the wide array of studies identifying de novo genes using multiple approaches in many taxa , the number of genes with functional characterization remains small . A recently identified yeast de novo gene , BSC4 , is important for DNA repair [9] , [16] . The Drosophila melanogaster de novo genes , CG31406 [17] and CG31909 [18] both showed pupal lethality in large RNAi screens and the mouse de novo gene Pldi affects male fertility [13] . The analysis of de novo gene function in humans has been restricted to analysis of previously existing gene expression and association with disease phenotypes in GWAS data , but are suggestive of function in the brain for one gene [19] . Here we combine an analysis of the evolutionary history – including analysis of sequence evolution and expression – with functional studies of six D . melanogaster de novo genes previously reported in the literature [4] , [20] . These six de novo genes represent a variety of “steps” in the evolution of de novo genes , consistent with previously described gradual models of de novo gene evolution [4] , [9] , [14] . Some de novo genes are specific only to D . melanogaster , D . simulans , and D . sechellia . Others have a deeper evolutionary history , with evidence of the evolution of transcription ( but not necessarily an ORF ) occurring in the common ancestor of D . melanogaster and D . yakuba/D . erecta or earlier . We find that two of the genes were clearly transcribed prior to the evolution of an open reading frame , supporting the concept that de novo proteins may evolve from non-coding RNA genes . In four other cases , an open reading frame and transcription appear to have co-occurred in the same evolutionary interval . Knockdown of de novo genes with RNAi showed that these de novo genes are important to organismal fitness . Finally , our data show that despite arising through different mechanisms , D . melanogaster de novo genes share evolutionary and functional similarities .
We investigated de novo genes previously described [4] , [20] as having arisen recently in the D . melanogaster lineage ( both D . melanogaster subgroup and D . melanogaster specific ) – along with other internal candidates ( Methods ) – and reassessed whether they qualify as de novo protein-coding genes using current genomic resources . For each gene , we determined whether proteins had arisen recently from apparently non-coding DNA by tBLASTn of the protein-coding regions to all 12 Drosophila genomes , as well as comparing to UCSC's BLASTZ alignments from D . yakuba , D . erecta , D . ananassae , D . simulans , and D . sechellia ) . This eliminated a number of candidates from consideration either because they were collinear to highly diverged putative protein-coding sequences in species previously analysed , or because one of the species in the 12 genomes that was not previously analyzed contained a potential ortholog ( see Table S1 for the full list of candidates ) . For the remaining six genes , we extracted the UCSC BLASTZ alignments for sections of each gene ( 5′UTR , all CDS exons , and 3′UTR ) , then used the pairwise sequence alignment program water to calculate the sequence identity and the proportion of the D . melganogaster sequence conserved between D . melanogaster and each of the other species in the alignment ( Figure 2 ) . We found that five of the six genes could be aligned to fragments of sequence from species as far diverged as D . yakuba or D . erecta , and in the case of CG34434 , CG31406 , and CG32235 , sequences that overlapped with the D . melanogaster open reading frame in these species were not interrupted by stop codons indicating that if transcribed and translated , a highly diverged protein or peptide may be produced in these closely related species . In addition , sequences collinear to portions of the CG34434 CDS and part of the CG32690 UTR could be found in D . ananassae ( Figure 2D and 2F ) . These sequences are highly diverged and major changes in size and structure were apparent in many cases . CG32582 and CG32690 can be distinguished from the other de novo genes because they appear to have an open reading frame that is unique to D . melanogaster alone . Collinear sequences in D . simulans and other species carry disabling mutations that greatly truncate any potential ORF ( Figure 2 , Supporting data ) . CG31909 is well-conserved in D . simulans and D . sechellia but no sequences similar to the CDS can be found in any other species . Interestingly , while the CG31909 CDS is novel , the 5′ UTR of CG31909 contains similarity to a short transposable element – perhaps sequence from elsewhere in the genome was inserted in the ancestor of D . simulans and D . melanogaster through movement of that transposable element . The lack of sequence similarity of the CDS for any sequence in any genome other than D . melanogaster and its two sister species makes it difficult to determine the origin of this sequence . CG31909 also has a near exact paralog ( 98% amino acid identity ) in D . melanogaster ( now annotated as CG43800 as of Flybase r5 . 45 ) that is specific to D . melanogaster . Interestingly , an RNAi screen of Notch signaling genes showed RNAi of CG31909 to be semi lethal [18] . The remaining genes ( CG31406 , CG33235 , and CG34434 ) have undergone structural changes after their origins resulting in increases over time in the size of the total gene ( CG31406 and CG33235 ) the size of the CDS ( all three ) , and the number of exons ( CG31406 ) ( Figure 2 ) . De novo protein-coding genes might evolve from previously non-coding but transcribed sequences ( “Transcription first” model , Figure 1 ) . Alternatively , a previously untranscribed ORF could arise through random mutation , and only later become transcribed ( “Proto-ORF” model , Figure 1 ) . Of course , these models are not mutually exclusive , and do not rule out other intermediate possibilities – such as transient transcription of an existing ORF later becoming stably transcribed ( see [6] , [9] , [14] ) . As described above , in all cases these sequences were highly diverged at both the sequence and structural level ( Figure 2 ) . We used qRT-PCR to measure transcription of these genes in species where collinear sequences could be found , regardless of protein-coding potential ( Figure 2 , with bolded text indicating species where transcription could be detected ) . With the exception of two genes , we were able to detect expression of transcripts in all species in which collinear sequence could be clearly identified ( CG31406 was expressed in D . yakuba but not D . erecta despite alignable sequence being present in both species; CG32582 was not expressed in either D . erecta or D . ananassae ) . These data suggest that the de novo evolution of expression can predate the evolution of the ORF and that existence of a proto-ORF was not a prerequisite for the evolution of transcription of the de novo gene . In the cases where an ORF was present ( CG31909 , CG34434 , CG31406 , and CG33235 ) , we surmise that the origin of the ORF and the evolution of stable transcription arose at around the same time . While these data are consistent with the hypothesis that transcription arose from nascent ORFs in the genome ( proto-ORF model ) , we cannot conclude that the proto-ORF existed first—transcription may have evolved first and then an ORF shortly thereafter . On the other hand , in cases where the sequence was clearly non-coding and stably transcribed prior to the evolution of an ORF ( CG32690 and CG32582 ) , we can conclude that the transcription-first model applies . We next mined the EBI PRIDE proteomic database for evidence that the extant ORFs were translated . Four of the six de novo genes – all but the newest ORFs , CG32582 and CG32690 –expressed peptides in early embryos ( [21]–[24] , Table S2 ) . It is unknown if the short proto-ORFs of these four genes are being translated in other species or if the other two genes are translated in other , less deeply surveyed tissues in D . melanogaster . All six genes have sequence features consistent with post-translational cellular localization - CG32690 , CG32582 , and CG34434 have secretory signals , whereas CG31909 has a nuclear signal and CG33235 is predicted to be localized to the mitochondria . In sum , we have evidence for translation of the ORF in all four of the “proto-ORFs” , but not for the two “transcription-first” genes . These data do not rule out the possibility that CG32690 or CG32582 are translated in D . melanogaster as only one tissue ( embryos ) was deeply surveyed , but these data are consistent with the interpretation that genes arising through a transcription-first mechanism are less likely to produce peptides and that their biological activity is tied to the evolution of a novel RNA , rather than a novel protein . Prior work shows that de novo genes in Drosophila tend to exhibit male-biased expression [4] , and are expressed at their highest levels in L3 larvae , pupae , adult males , and the adult reproductive system [25] . We compared expression in D . melanogaster in adult testes , male accessory glands , the remainder of the male tissues , and adult females . In addition , we sexed L3 larvae [26] and measured expression in male and female larvae . We found male-biased expression in all six genes . Expression of de novo genes was at its highest in the testes , and male larvae expressed at a higher level than female larvae ( Figure 3 ) . We also found that lack of a male germline ( Figure 3 sons-of-tudor , light green ) reduces but does not typically eliminate expression ( transcription of CG32690 was undetectable in the sons-of-tudor testes ) . This suggests that these de novo genes are contributing to the development and maturation of sperm , but likely perform other functions as well . Following on this result , we determined whether these genes were regulated downstream of a spermatogenesis specific gene by measuring expression in a tombola ( tomb ) mutant background . tombola is a transcription factor known to activate expression of a suite of genes important during male meiosis in Drosophila [27] . We found that expression of CG31406 was reduced in the tomb mutant background ( Figure 3 , red ) implying expression of this gene is partially dependent on an intact meiotic arrest pathway . The other genes , however , did not appear to be affected by tomb , suggesting that though they are expressed at a high level in the male germline they either operate up-stream of tomb or are regulated by a parallel pathway . Next , we compared expression levels of collinear expressed sequences in tissues ( testes , male carcass , and female ) from D . simulans , D . sechellia , D . yakuba and D . erecta ( Figure 4A–D ) . Despite radical structural and sequence changes , testes-biased expression of all de novo genes was conserved for species in which expression could be readily detected . It has been suggested that de novo genes might occasionally be transcribed spuriously ( possibly due to a permissive transcriptional environment [28] ) prior to recruitment of a more specific promoter upon evolution of a novel function . This idea predicts that expression levels should vary stochastically across species . Our results suggest instead that de novo genes have been expressed in a biased manner from the moment transcription originated . Additionally the sons-of-tudor and tomb data suggests that active regulation of these genes' expression evolved early . The consistency of testes-biased expression of the genes across species led us to hypothesize that these genes may function primarily as male fertility genes . Contrary to our expectation , we found that RNAi knockdown of the four de novo genes we were able to assay strongly affected viability . RNAi stocks from the VDRC's [29] phiC31 library ( also known as “KK stocks” ) crossed with a ubiquitous Actin5C-GAL4 driver ( y1 w*; P{Act5C-GAL4}25FO1/CyO , y+ ) , produced no RNAi offspring for the four genes assayed ( CG31406 , CG32582 , CG34434 , CG33235 ) , We further characterized the viability phenotype using a driver line that included a GFP marker ( y1 w*; P{Act5C-GAL4}25FO1 , UAS:CD8:GFP/CyO , y , donated by S . Chen ) and found that lethality occurred in all four cases at the late pharate adult stage , just prior to eclosion ( Figure 5 ) . Our observation of pharate-stage lethality is consistent with previous work showing RNAi of CG31406 leads to pharate-stage death [17] . This result suggests that these four de novo genes may be essential . To rule out spurious effects of RNAi , we crossed all RNAi lines to an additional ubiquitious Tubulin-GAL4 driver ( y1 w*; P{tubP-GAL4}LL7/TM3 , Sb1 , Bloomington #5138 ) as well as a driver that targeted testes and various essential larval tissues ( larval fat body , gut , leg discs , and salivary glands , w1118; P{GawB}c564 , Bloomington #6982 ) with the same result . We also drove RNAi expression of a negative control phiC31 RNAi stock ( Gr22c ) using the Actin5cGAL4 driver and saw no lethality , as expected . Finally , we measured the extent of RNAi knockdown for all lines and found that RNAi samples had weaker expression of the target gene than controls ( Figure S1B ) , whereas there was no significant knockdown of genes predicted to be potential off-targets by sequence similarity ( Figure S1C ) , which is consistent with other studies using these lines that show that off-target effects are rare [18] . We also obtained P-element RNAi lines from the VDRC ( also known as “GD stocks” ) for four of the six genes ( CG33235 , CG31406 , CG31909 , and CG34434 ) . Due to their random placement in the genome , the P-element library produce more variable knockdown than the “KK” stocks in which the construct is placed in the well characterized phiC31 site ( expression of the “GD” stock was weaker for two of the three genes for which we had both a “KK” and a “GD” stock , Figure S1 ) . Using the same design as above , all “GD” lines produced viable progeny of both sexes . We confirmed partial knockdown ( Figure S1A ) of the target genes in adults from three of the crosses ( P<0 . 05 ) , but CG31909 did not show knockdown ( P = 0 . 42 ) . This gene showed partial pupal lethality in an earlier study where its expression was driven by pannier promoter [18] , suggesting our ubiquitious driver did not express RNAi strongly enough to knock down expression . CG34434 GD-RNAi showed robust ( ∼40-fold ) knockdown and a semi-lethal phenotype in adults ( Table 1 ) , with males more affected than females , whereas CG31406 , CG31909 and CG33235 GD RNAi had no significant affect on overall viability . In addition , CG34434 GD-RNAi males had a dramatically reduced lifespan compared to control males ( Figure S2A ) . Although overall viability was not affected in the other three genes tested , female-biased skews in the sex-ratio of F1 adults were observed for three of the four genes tested ( compared to the expected 50∶50 sex-ratio and the observed sex-ratio of controls ) . As parents do not carry RNAi - only offspring - the skewed sex-ratios cannot be the result of sex-chromosome meiotic drive . Indeed , we saw no bias in sex ratio of F2 offspring in subsequent experiments ( described below ) . Instead , these findings could be the result of a male viability defect of the same type that caused complete lethality in the KK-RNAi lines , or in principle , increased viability among RNAi females . Using males from the three RNAi crosses that produced viable males , we proceeded to measure effects on male fertility and sperm production using two assays ( Figure 6 , Figure S2 ) . We mated single RNAi and control F1 males to w1118 females , and found that total fertility was reduced by RNAi of CG34434 ( Figure 6D , Student's t-test P<0 . 0001 ) but not CG31406 ( Figure 6A ) or CG33235 ( Figure 6C ) . We extended these findings using a sperm exhaustion assay [30] for two of the genes ( CG33235 and CG34434 ) . Sperm exhaustion measures the ability of a male to continue to produce viable progeny when challenged with multiple females over a five day period and can be more sensitive to subtle differences in fertility . CG34434 GD-RNAi males performed even more poorly during the later days of the assay than in the single-day mating experiments , but there was still no effect on the fecundity of CG33235 GD-RNAi males using this assay ( Figure S2B ) . Rather than having a direct effect on fertility , we suspect that CG34434 GD-RNAi males are weaker overall as indicated by their shortened lifespan ( Figure S2A ) and hence were less able to mate successfully . That said , RNAi of these genes using a more specific and powerful male germline driver might reveal specific defects in spermatogenesis or fertility that we were unable to detect in this preliminary analysis . Because we were unable to knock down expression of CG31909 using RNAi , we produced TILLing lines for CG31909 [31] obtaining an allele with a premature termination codon ( PTC , predicted to truncate 40% of the protein ) as well as a number of nonsynonymous mutations . We crossed the PTC line ( SH2_0024:R89* ) to a deficiency covering the CG31909 gene region ( w1118; Df ( 2L ) BSC291/CyO , Bloomington #23676 ) and the PTC allele did not alter expression ( data not shown ) , which was not unexpected as nonsense mediated decay in Drosophila does not typically affect expression if PTCs occur within ∼400 bp of the polyA signal [32] , [33] . None of the alleles appeared to affect viability . We used the same two fecundity assays described above to determine whether the PTC a protein-coding mutation ( D118>N ) reduced fertility and saw no effect of the flyTILL lines on performance compared to controls ( a D->N mutation at position 118 and w1118 crossed to the same deficiency , Figure 6B , Figure S2B ) . This could be for a number of reasons . First , CG31909 has a recently evolved D . melanogaster-specific near duplicate in that is also testes-expressed according to modENCODE and EST data ( BT023668 ) , and recently annotated as a protein-coding gene , CG43800 ( as of flybase release 5 . 45 ) . This duplicate's function may be redundant with CG31909 and sufficient to complement our TILLing mutant . Second , CG31909 may be expressed in the testes but not essential for male fertility . Third , given that knockdown of CG31909 by the Notch pathway promoter of pannier resulted in a lethal phenotype similar to other de novo genes , yet our nonsense and missense mutations had no effect on viability , CG31909 may function in viability as a long non-coding RNA gene , despite the fact that it produces a protein . The de novo genes in our analysis are identified in part as being lineage-specific by a lack of sequence similarity to protein-coding genes in other species . Thus , it is unsurprising that these genes are highly diverged at the sequence level when compared to those relatives harboring orthologous sequence ( Figure 2 , Dataset S1 ) . However , as we found many of these genes have become involved in essential functions , we expect that they have experienced strong selection as they acquire these functions . Where possible , we aligned the D . simulans and D . melanogaster extended gene region and compared with polymorphism data from D . melanogaster [34] ( lines collected from Raleigh , USA , “NA” and Malawi , Africa “AF” as part of the DPGP project ) using Variscan [35] . Divergence ( Figure 7 , κ , black bars ) was always highest over the part of the region including the gene , whereas polymorphism was usually lower or similar to background levels ( Figure 7 , π , dotted lines ) . Furthermore , regions overlapping the CDS of CG32582 and of CG32690 had elevated rates of divergence compared to the entire transcribed region . An increased rate of divergence without a similar increase in polymorphism is generally consistent with positive selection acting on a gene . However , polymorphism-based metrics ( Tajima's D and Fu and Li's D and F [36] , [37] ) failed to show significant deviation from neutrality for blocks containing the de novo genes ( Table S3 ) . Failure to reject the null could be due to low levels of polymorphism present within the open reading frames of the de novo genes and the small size of the genes combining to reduce the power of the test . We also tested whether protein-coding regions of four genes with D . simulans ORFs ( CG34434 , CG33235 , CG31909 and CG31406 ) show signs of recent positive selection . Each gene had high levels of both synonymous and nonsynonymous divergence when compared to D . simulans ( Table 2 ) , but dN/dS was below 1 in all cases , implying the genes are selectively constrained . None of the proteins tested show strong evidence that they have recently evolved under positive selection , though they are diverging rapidly at the sequence level . The DoS estimates and dN/dS indicate that CG31909 is the most likely of the four to be evolving under positive selection , though the McDonald-Kreitman test was not significant . On the other hand CG33235 and CG34434 show evidence of purifying selection ( DoS is negative and dN/dS are <1 ) , despite high levels of nonsynonymous divergence . This makes sense given the evidence that these genes are essential for viability in D . melanogaster . For our six candidate de novo genes , the DPGP data show no evidence that any variants that disrupt the open reading frame are segregating ( in the DPGP data set ∼3% of all genes harbor a segregating null [38] ) . In the case of CG31909 , the region overlapping the gene was not found in the DPGP dataset , but a broad ( 300 allele ) PCR-based survey of a natural population of D . melanogaster for deletions of CG31909 found that in all cases , the gene was intact . Combined with our RNAi data the absence of common null mutations reinforces our observation that de novo genes have become important to fitness .
Of the five D . melanogaster de novo genes we investigated in an RNAi screen , four RNAi lines resulted in lethality in our assay , three led to skewed sex-ratios in adults most likely due to sex-differential survival , and one showed altered male reproductive fitness ( though this case may be a side effect of the reduced male viability in the same cross , Figure S2A ) . In short , de novo genes are consistently evolutionarily and biologically essential . In contrast , the origins of these genes are divergent—some de novo genes clearly began as ( de novo ) long RNAs , whereas others may have emerged from a proto-ORF , although it is clear that a proto-ORF is not required for their evolution . After they arose , de novo genes' sequence and structure invariably evolved rapidly . However , we did not detect significant signatures of recent positive selection , but this may be due to problems with power in the data ( particularly the low levels of polymorphism ) . Earlier work suggested positive selection had acted on some of these genes [4] . RNAi knockdown caused lethality in four of five de novo genes tested , a surprising finding because these genes are very young—if these genes are essential , what function are they performing now that was apparently not needed by the ancestor ? The lethality consistently occurred during late pharate adult stages ( pre-eclosed adults ) , after full eye pigmentation and the appearance of bristles had begun ( Figure 5 ) . Expression of all the genes studied was high in both larvae and male adults , and this data suggests that the essential function of these genes begins prior to the adult stage . This implies that de novo genes are playing an important role in the development of the adult fly . Alternatively , during the sensitive pupation stage , the fly may not tolerate absence of a de novo gene even though this could be tolerated during larval development . RNAi can have off-target effects , but we did not find evidence of knockdown of any genes predicted to be off targets by sequence similarity or lethality in genetic controls ( Figure S1 ) . Other large RNAi screens using similarly generated lines suggest that such off target effects are rare [17] , [18] and that phenotypic effects produced by these lines are often confirmed with genetic mutants . It is impossible to completely rule out effects of RNAi on off-targets that have , for example , very weak sequence similarity to the double-stranded RNA , so extending this work using genetic mutants is a logical next step . These strong effects on viability may appear at first to be at odds with the finding that expression of these genes is often strongest in the testes ( Figures 3 and 4 ) . Contrary to our naïve expectation , only one of the RNAi lines produced a defect in fertility ( Figure 6 , Figure S2B ) and we interpret this effect to be a result of reduced robustness in RNAi males ( Figure S2A ) . This pattern may be explained by global gene expression patterns . While nearly 20% of Drosophila genes show male-biased expression – a huge excess compared to other tissues [25] , genes expressed in male germline stem cells prior to meiosis are typically also expressed in at least one other cell type [39] . Therefore , strong expression of a gene in the testes may not be a good indicator that a gene's function is testes or even male specific . For instance , we found that CG31406 was under the regulation of a meiotic arrest gene , tombola , which functions in sperm development ( Figure 3 ) . Yet this gene had a strong effect on viability . Examples like this suggest that genes may be expressed at a high level due to general transcriptional “permissiveness” in the testes [40] , [41] , but their expression may not be critical to male reproduction . Alternatively , the strong testes expression may reflect the evolutionary origins of these genes rather their current function in the fly – that is , expression patterns may be conserved through phylogenetic momentum . This would be consistent with the hypothesis that the testes act as an “evolutionary playground” for the emergence of new genes that are later adapted to other functions [28] . Researchers have speculated that de novo genes may function as non-coding RNAs [13] , [42] , as seminal peptides ( particularly in Drosophila , where they are often found to show expression in the male reproductive tract [4]–[6] ) , or may not be functional at all , but expressed as a side effect of nearby transcription or overly promiscuous transcription in particular tissues [28] . However , increasing evidence suggests that new genes of all forms , including de novo genes , are important to fitness . Our data suggest that in the time since these de novo genes arose they have integrated into some key developmental or physiological network and become critical to some basic function of the fly . These results parallel data from yeast [9] , [16] , which found that loss of a de novo gene in a synthetic lethal screen was lethal , and similar to work by Chen and colleagues showing that many types of young genes in Drosophila are essential [17] . Interestingly , although we tested only a handful of genes , this 80% “essentialness rate” is actually significantly ( P = 0 . 035 ) higher than the ∼30% lethality rate observed for all classes of young genes and the 35% observed for old genes by Chen and colleagues . Thus , when a de novo gene arises and persists it appears even more likely than most other young genes to be integrated into an essential aspect of fly biology . While our sample size is small and should be interpreted with caution , it is remarkable that so many of these genes appear to be essential . How can we explain this finding ? The appearance of a wholly new gene would seem more likely than other types of mutation to result in a large phenotypic change . Models of both phenotypic and genotypic evolution predict that larger than expected changes occur early during a bout of adaptive evolution [43] , [44] . While this may explain why the phenotypic effects of a new gene should be large it does not explain why these genes would become essential at a disproportionate rate . To become a gene that codes for a protein whose loss results in death , a de novo gene must become integrated into an essential physiological or developmental pathway . Unlike new duplicates - which often retain interacting partners with their parent genes - these genes are entirely novel and any interactions they have with other genes would be novel . Perhaps as the network adapts to the presence of a new member , the de novo gene becomes essential to network function and unlike new duplicates , if lost , interactions cannot be replaced by a parent copy . Interestingly , all of these proteins do have predicted interactions on the DroID database [45] , including a substantial number of interactions with small RNAs . CG31909 , for instance , is annotated as having interactions with six miR , including those important for development and ecdysone signaling ( miR-125 ) . Our data show that two de novo genes first arose as non-coding RNAs . Although their ORFs are disrupted in non-D . melanogaster species , CG32690 and CG32582 are transcribed with a similar expression pattern across species . This pattern is similar to that seen in the mouse de novo gene , Pldi . Heinen and colleagues [13] argued that it is unlikely that a protein arising from a novel RNA would be functional and annotated their newly evolved transcripts as non-coding RNAs despite the presence of short open reading frames in these genes . However , our data suggest that for the other four genes considered in this study , the open reading frame may have been present when transcription began . Proteomic data from the EBI PRIDE database [21]–[24] showed evidence these “proto-ORF” de novo genes we identified do produce peptides in D . melanogaster ( Table S2 ) . Thus it seems unlikely that de novo genes function solely as RNA genes/lncRNAs , although we cannot reject the hypothesis that these protein coding de novo genes began as functional lncRNAs that later evolved an ORF , or that they may produce non-functional peptides and function primarily as lncRNAs . Recent data suggests that a substantial fraction of non-coding DNA is experiencing natural selection [46] . Much of this selection is thought to be acting on regulatory sequences such as promoters and enhancers , and these types of changes are thought to be essential in adapting existing genes to perform new functions [47] . Our data suggests that selection is also shaping non-coding regions into functional protein coding genes are recruited into the basic and fundamental genetic pathways of the fly .
Using data from Levine et al [4] and Zhou et al [20] , we chose a number of published de novo genes to further characterize . In short , we combined the candidate genes from these two studies with an additional analysis comparing CDS of annotated D . melanogaster protein coding genes from FLYBASE ( v4 . 3 ) , which included a handful of partially annotated non-coding RNA genes , to the genomes of all other Drosophila species available at that time ( tBLAST ) . Proteins that failed to have similarity to the any genomes outside the melanogaster clade we considered candidates . These candidates were then filtered ( described below ) and candidates were ruled in or out as de novo genes using currently existing data ( Table S1 ) . For example , the CDS of the genes presented have no significant hits by translated BLAST ( e = 10∧−6 ) to genes outside of D . yakuba/D . erecta . We mined the NCBI trace archive to rule out the possibility that assembly error in species other than D . melanogaster had led to the misannotation of these genes as de novo and found no evidence these genes existed among the traces in species outside of what was previously reported . We searched UCSC's whole genome chained BLASTZ alignments , which are more sensitive to highly diverged hits than BLAST or BLAT [48] in order to find genomic regions collinear to the immediate gene regions in other species . We then used the UCSC [49] and Flybase [50] genome browsers to ask whether the D . annanassae , D . yakuba , D . erecta , D . simulans , and D . sechellia chained alignments covered annotated genes in whole or in part , despite not matching by BLAST/BLAT . Genes that were found to be collinear to annotated genes with similar structure in all five species were excluded as putative rapidly evolving loci ( Table S1 ) . In cases where gene structures were radically different , but there was overlap with an annotated gene , we used RT-PCR to verify ( or exclude ) the annotated gene models . In the case of CG34434 , we found that the annotation of the putative D . yakuba ortholog incorrectly connected the putative ortholog of CG34434 with a neighboring gene , and that the D . simulans gene had a second , unannotated exon similar to the second exon of the D . sechellia ortholog . These corrected gene structures were used in the presented analysis . Finally , the flybase annotation of the collinear D . sechellia CG34434 ortholog ( GM12640 ) had an incorrect splicing pattern leading to a frame-shifted second exon . Once corrected , GM12640 was similar in sequence and structure to CG34434 . We have contacted flybase and provided them with evidence for these updated annotations . We downloaded BLASTZ [48] alignments of the extended gene regions surrounding the six candidate de novo genes from the UCSC genome database . We used these alignments to determine which parts of the D . melanogaster putative lineage-specific genes and their flanking sequences were collinear to sequences in each of the other species . We extracted any portion of the alignment overlapping transcripts and realigned pairs of sequences ( D . melanogaster against each other species ) using the “water” pairwise alignment program , part of the EMBOSS suite [51] . We calculated the total sequence similarity and the proportion of alignable bases between sections of each gene ( e . g . CDS , UTRs , etc ) from these pairwise alignments . We also performed a global pairwise alignment of the D . melanogaster and D . simulans extended gene regions ( extracted from FlyBase genbank files ) using progressiveMAUVE [52] , [53] . We counted the number of fixed differences between D . melanogaster and D . simulans in 500 bp windows along the alignment , then aligned 39 D . melanogaster Raleigh genomes and 6–9 African genomes ( www . dpgp . org , [34] ) to these regions and calculated polymorphism ( π ) and divergence ( κ ) in each window . We looked for evidence of null alleles ( e . g . premature stop codons in the DPGP data ) and calculated Tajima's D [36] and Fu and Li's D and F [37] for 500 base pair windows across the region using Variscan [35] . For genes with intact proteins in D . simulans , we aligned the protein-coding regions using ClustalW and used these alignments to calculate the Neutrality Index ( NI ) and the Direction of Selection ( DoS , [54] ) , and to perform a Macdonald-Kreitman test [55] . SNAP [56] was used to calculate dN/dS relative to D . simulans , except in the case of CG33235 where the comparison was to D . sechellia as that species has a longer ortholog than D . simulans . Finally , in the case CG31909 , data from DPGP was not available for most of the gene's CDS . Instead , we screened 150 wild caught African flies for deletions of CG31909 , which would be expected to occur if the gene were non-essential . PCR was performed using primers ( CTTGGCCCTGCGAAGTGAACACC and CGCACTGGGCGCTGAAATCTGTG ) amplifying a ∼1 kb region surrounding CG31909 looking for a negative reaction or short product . Candidates were then sequenced to confirm or deny the null allele . Male reproductive tracts were dissected on ice from whole flies ( D . yakuba , D . simulans , and D . melanogaster ) in sterile PBS . Male reproductive tracts and carcasses were each pooled from at least 10 individuals and then flash frozen in liquid nitrogen . Whole females and males of each species were also collected , pooled and flash-frozen . D . melanogaster , D . simulans , and D . yakuba male reproductive tracts were further dissected into accessory glands and testes in PBS and flash frozen . D . melanogaster third instar larvae were sexed by identification of male and female genital discs following Drosophila protocols [26] , then flash-frozen . Testes were dissected from males carrying a null mutation at the gene tombola ( tombGS12862 , stock generously supplied by Dr . Helen White-Cooper ) , and sons of females mutant for the tudor gene ( Bloomington stock #1786 – sons of these flies lack a male germline ) . We extracted RNA from two or more biological replicates of each dissected tissue using TRIZOL reagent ( Invitrogen , Grand Island , NY #15596-026 ) , and synthesized cDNA using M-MLV reverse transcriptase ( Invitrogen , Grand Island , NY #28025013 ) . We performed relative qRT-PCR quantification using gene-specific primers and a single control primer that worked across all species ( Actin5c ) . All qRT-PCR Ct values were averaged across two technical replicates . In addition to our own data , we mined expression information from online databases - FlyAtlas [57] , modENCODE RNAseq data [25] , Baylor RNAseq data [58] , and FlyTED: Testes expression database [59] , and DroID [45] . Additionally , we mined Drosophila proteomic data from multiple sources [21]–[24] . These datasets are biased towards proteins expressed in early embryos as this constitutes ∼35% of available proteomic data and the handful of studies of testes and seminal fluid were of comparatively low depth [60]–[62] . Virgin Actin5C-GAL4 females ( y1 w*; P{Act5C-GAL4}25FO1/CyO , y+ , Bloomington 4414 ) were collected and crossed at 25C to lines carrying UAS-RNAi constructs for CG33235 , CG31909 , CG31406 , CG34434 , CG32582 and Gr22c - a control obtained from VDRC [29] ( stocks used: 19355 , 23550 , 39194 , 41772 , 102263 , 104704 , 105072 , and 110307 , 105051 ) . CyO ( control ) and straight winged ( RNAi ) progeny of both sexes were counted and collected . For RNAi knockdown in larvae , we crossed the same RNAi lines to a stock with Actin-GAL4 and CD8::UAS-GFP on the same chromosome ( y1 w*; P{Act5C-GAL4}25FO1 , UAS:CD8:GFP/CyO , y , donated by S . Chen , [17] ) . In these crosses , RNAi or control status can be ascertained at any stage ( RNAi larvae/pupae/adults will express GFP ) . We collected , sorted , and sexed larvae in the wandering stage and compared expression of the target gene using RT-PCR . To estimate effects on adult viability , we counted the number of control ( CyO ) and RNAi ( straight-winged ) progeny eclosing from the RNAi cross ( described above ) . To determine the stage at which lethality was occurring , we crossed the same RNAi lines to a GFP marked Actin-GAL4 line ( see above ) . We collected larvae from the cross during the late third instar wandering stage , and sorted by GFP expression and sex [26] ) . We then allowed RNAi ( GFP ) and control ( no GFP ) to continue development , and counted the number that survived or died prior to pupation or prior to eclosion .
|
De novo genes are protein-coding genes with no clear homology to previously existing protein-coding genes . Since their discovery in Drosophila and other species including humans , their existence has been controversial , with some doubt as to how they would arise , whether they produce proteins , and whether they could possibly perform any useful function . Here , we show that RNAi of several Drosophila de novo genes causes lethality – in fact , a higher proportion of de novo genes cause lethality than was found in a similar screen of other young and novel genes . Further , we find that de novo genes do produce proteins in the majority of cases and that in some cases , they were transcribed prior to the emergence of an open reading frame . Our data suggests that Drosophila de novo genes are an unexpected avenue for non-coding DNA sequences to contribute evolutionary and functional novelty .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2013
|
De Novo ORFs in Drosophila Are Important to Organismal Fitness and Evolved Rapidly from Previously Non-coding Sequences
|
This study was conducted to ( i ) determine the prevalence of African Animal Trypanosomosis ( AAT ) in tsetse challenged areas , ( ii ) compare conventional with qPCR detection systems and ( iii ) evaluate the host genetic background and biology as risk factors . AAT prevalence studies are often confronted with low levels of parasitaemia . Hence , we designed a novel qPCR assay using primers and species specific probes amplifying the Internal Transcribed Spacer 1 ( ITS1 ) gene . Thereby all three AAT species could be detected simultaneously . 368 individuals from three cattle types ( Baoulé , Zebu and hybrids ) originating from 72 farms in Burkina Faso were analysed . Farmers were interviewed and morphometric measurements of the cattle taken . A chi-squared test and a logistic regression model were calculated to detect associations with infection . In our study , the overall rate of prevalence detected with the novel qPCR assay was 11 . 14% . Compared to conventional PCR we identified a concordance of 91 . 30% . We tested 41 animals positive for trypanosome DNA , five animals showed multiple infections . Zebus were twice as often infected ( 21 . 74% ) compared to Baoulé ( 9 . 70% ) and hybrids ( 9 . 57% ) . Trypanosoma vivax is the dominant species ( 9 . 24% ) , as compared to T . congolense ( 2 . 44% ) and T . brucei ( 0 . 82% ) . The chi-squared tests linking the infection events to the breeds ( Zebu vs . Baoulé and Zebu vs . hybrids ) were on the border of significance . No significant association with other tested parameters could be detected . We introduce a novel qPCR technique for the fast , sensitive and simultaneous detection of the three AAT species . Our results suggest that associations with breed and infection exist since Zebu cattle are more likely to be infected compared to Baoulé and hybrids . Indigenous taurine cattle breeds , like the Baoulé , therefore provide a unique and valuable genetic resource .
Trypanosomiasis affects both humans ( sleeping sickness ) and animals ( nagana ) and occurs in 37 sub-Saharan countries . Approximately 60 million people and about 50 million cattle are currently living in risk of infection [1] . The International Livestock Research Institute ( ILRI ) has listed trypanosomosis among the top ten global cattle diseases impacting on the poor [2] . In tsetse challenged areas of Burkina Faso the African animal trypanosomosis ( AAT ) is ranked first among nine most important cattle diseases [3] . However , over thousands of years , and presumably under high tsetse challenge , some West African Bos taurus cattle breeds have developed a tolerance to trypanosomosis in the course of evolution [4] . One trypanotolerant breed , and thus represents a valuable genetic resource , is the Baoulé cattle . The trypanotolerance character enables them to control the development of parasites and to limit the associated pathological effects and level of parasitaemia [5] , [6] . In contrast , zebu ( Bos indicus ) cattle types are more susceptible to trypanosome infections and can only be maintained in tsetse challenged areas through the use of costly trypanocidal drugs . Three different parasite species , T . congolense ( subgenus Nannomonas ) , T . brucei ( subgenus Trypanozoon ) and T . vivax ( subgenus Duttonella ) are causative agents of AAT in cattle [7] . In most areas several trypanosome species can be found in sympatry resulting in single or multiple species infections . African trypanosomes are of great concern for public and animal health , particularly in regions where most of the pathogenic trypanosome species are present . Therefore , the discrimination of the trypanosome species , subspecies or strain can be necessary for medical , sanitary , taxonomic or epidemiological studies [8] . AAT prevalence has been previously assessed microscopically with blood smears [9] or by buffy coat examination [10] , [11] . Both methods lack sensitivity and are laborious . ELISA techniques can be applied on large sample sets with accurate precision results but it does not differentiate between present and past infections . Moreover , trypanosome species cannot be discriminated with these serological-based detection methods [12] . Following the increasing demand for accurate , fast , sensitive and efficient detection tools , molecular methods have been steadily improved in recent years . To identify the most appropriate detection system , prevalence studies comparing several techniques were performed . In Sideradougou , Burkina Faso a survey of the agro-pastoral zone showed that the parasitological prevalence in cattle detected with the buffy-coat method was 5 . 3% compared to 11 . 5% using PCR methods [13] . PCR is generally considered as an efficient tool to estimate the prevalence of AAT in affected areas [13] , [14] . Thereby infections can be detected with more sensitivity and valuable information for prevalence studies provided [14]–[17] . The internal transcribed spacer ( ITS ) region of ribosomal DNA ( rDNA ) is a preferred target for universal testing because of its size variability among trypanosome species and subspecies . To discriminate the different trypanosome species in one step , primers ( ITS1-CF and ITS1-BR ) were designed targeting the ITS1 region . These show high diagnostic sensitivities and are capable of detecting all pathogenic trypanosomes in a single PCR with amplicon lengths below 720 bp . This is particularly helpful since approaches to combine conventional PCR reactions into a multiplex PCR for species differentiation are disappointing due to a decrease in sensitivity and non-specific PCR products ( Njiru et al . , 2005 ) . As prevalence studies of AAT are confronted with low levels of parasitaemia in chronically infected or trypanotolerant cattle , sensitivity is a major objective in test design and evaluation . Low levels of parasitaemia can only be detected with high-performance molecular diagnostic tests . In trypanosomosis real-time PCR ( qPCR ) methods were described for T . evansi [18] , [19] , T . brucei [20] and T . cruzi [21] , [22] . However , no qPCR for the simultaneous detection of the three species of AAT has been developed so far . In this study a new qPCR assay was designed and its performance compared to the conventional ITS1-PCR . In addition , information on morphological traits , biology and genetic background of the hosts was collected to identify risk factors for trypanosome infection . The present study was conducted to ( i ) determine the prevalence and incidence of AAT in tsetse challenged areas of Burkina Faso , ( ii ) compare conventional and qPCR detection systems , ( iii ) determine levels of parasitaemia and ( iv ) evaluate the genetic background and host biology as risk factors .
All blood samples from cattle were taken during routine veterinary examination by veterinarians or trained personnel . The participating owners of cattle provided their consent and agreed to fill out the survey form . In this study 368 individuals from three cattle types , namely the indigenous Baoulé ( n = 134; Figure 1A ) , the Zebu ( n = 46; Figure 1B ) and their hybrids ( n = 188; Figure 1C ) were sampled covering the three provinces Cascades , Hauts-Bassins and Sud-Ouest inside the tsetse belt of Burkina Faso . 12 villages from the West and 12 villages from the South-West of Burkina Faso were selected and cattle from six farms per village adding up to 72 different farms sampled ( Figure 2 ) . Purebred Zebus are rarely held in tsetse challenged areas due to their greater risk of severe disease . Therefore the sample set was not equally distributed within the three populations but reflects the actual population structure in southern Burkina Faso . In addition to the 368 animals , 28 Zebu cattle from the North ( Seguenega n = 8 , Marmisga n = 8 , Yako n = 7 , Nommon n = 5 ) were used to screen for mechanical trypanosome transmission . Breed assignment was performed based on farmer's information , asking also for the breed status of the two parents and records confirmed by body measurements ( height at withers and chest circumference ) . In addition information on the genetic background was made available [3] . Whole blood was collected , 500 µL spotted onto each Whatman-FTA card sample area ( GE Healthcare , Wisconsin , USA ) air dried and stored at room temperature for subsequent DNA extraction . After transfer to the Vetmeduni Vienna , three discs of 3 mm each were taken from spotted blood on a total of 368×3 FTA cards and used for DNA extraction according to Whatman FTA Protocol BD08 ( www . whatman . com ) . For improved DNA yield a Whatman FTA Purification Reagent containing 60 µg/ml Proteinase K was added to the disks and incubated overnight . On the next day washing steps with Whatman FTA Purification Reagent without Proteinase K , 0 . 3 M sodium carbonite , 0 . 5% SDS and TE−1 buffer were performed . Ultimately , the disks were placed into a new tube , incubated at 95°C for 15 min in 120 µl 1% PCR elution buffer and the supernatant collected . In addition to blood samples , we furthermore collected data on several variables with potential importance for the acquisition of trypanosome infections . Positive controls for T . congolense Savannah , T . brucei brucei and T . vivax were generously provided by the International Atomic Energy Agency ( IAEA ) and the concentration was spectrophotometrically determined using Ultrospec 2000 ( GE Healthcare , Wisconsin , USA ) . The following primers were used to amplify the trypanosome ITS1 gene: ITS1 CF 5′-CCG GAA GTT CAC CGA TAT TG-3′ and ITS1 BR 5′-TTG CTG CGT TCT TCA ACG AA-3′ [16] resulting in species specific size products ( Figure 3 ) . PCR was performed in a 25 µL volume containing 5 µL genomic DNA , 600 nM of each primer , 5× PCR buffer ( including 3 mM MgCl2 ) , 0 . 8 mM dNTPs , and 1 U of Taq DNA polymerase ( Go Taq , Promega , Madison , USA ) . PCR cycle conditions consisted of an initial denaturation step at 94°C for 3 min , 30 cycles of 94°C for 30 s , 55°C for 30 s , and 72°C for 30 s , followed by a final extension step at 72°C for 10 min . The amplified products were detected by electrophoresis on a 2% agarose gel ( peqGOLD , PeqLab , Erlangen , Germany ) stained with ethidium bromide . Species specific PCR was applied for the sequencing of selected positive samples for verification purposes . TCF1 and TCF2 for T . congolense Riverine/Forest and TCS1 and TCS2 for T . congolense Savannah [17] were used . PCR products were purified using Illustra ExStar 1-Step ( GE Healthcare , Wisconsin , USA ) . Two samples with weak bands on the agarose gel were cloned into a TOPO vector ( LifeTech Austria , Vienna , Austria ) and sequenced . The resulting sequences were BLASTed using the basic BLAST search tool and aligned using CodonCode Aligner ( version 2 . 0 . 6 ) . In the qPCR assay , new primers and probes for the simultaneous detection of all three AAT species were designed ( Table 1 ) . Trypanosome ITS1 sequences were selected from NCBI database ( www . ncbi . nlm . nih . gov ) , including T . congolense ( GI:1040856 , GI:1040860 - Forest-type , GI:1040858 - Kilifi type ) , T . brucei ( GI:14276830–14276836 ) and T . vivax ( GI:1040857 ) . The sequences were aligned in CodonCode Aligner and primers and TaqMan probes designed with Primer Express software version 2 . 0 . The probes were placed spanning species specific sequences of the ITS1 region for subsequent species discrimination . The universal primers and species specific fluorescent labelled probes allowed the detection of parasite DNA upon binding on the ITS1 gene . The specificity of the probes and primers was evaluated using BLAST searches and Primer BLAST respectively . A simultaneous amplification of the cattle specific toll-like-receptor-8 ( TLR-8 ) gene was run as inhibition control to detected false negative results ( Table 1 ) . qPCR was performed in a 25-µL volume containing 5 µL genomic DNA , 200 nM of each primer , 160 nM of each trypanosome probe , 120 nM of the TLR-8 probe , 5× PCR buffer , 6 mM MgCl2 0 . 8 mM dNTPs , and 1 U of Taq DNA polymerase ( GoTaq Hot Start Polymerase , Promega , Madison , USA ) . All three trypanosome FAM-labelled probes were added in equal amounts . The qPCR cycle protocol started with the initial denaturation at 95°C for 10 min , 45 cycles of 95°C for 30 s , 61°C for 1 min , and ended with 72°C for 1 min . qPCR was performed in a Strategene Mx3000P ( Agilent , Santa Clara , USA ) thermal cycler . All amplifications were reproduced in duplicates whereas triplicates were used for the standard curve and limit of detection . Quantification cycle ( Cq ) values of ≤38 were regarded as potentially positive and those samples used for further analyses . For species differentiation all potentially positive samples were applied to single-plex qPCR assays containing only one species specific probe . Alongside , the level of parasitaemia was identified in the single-plex assays . Using serial dilutions of trypanosome DNA the sensitivity in detecting and quantifying trypanosome infections was analysed . For the limit of detection serial 10-fold dilutions ranging from 20 ng to 0 . 2 fg were tested for each trypanosome positive control . This covers 105 to 0 . 001 parasite equivalents when considering that one parasite cell consists of approximately 200 fg of DNA [21] . The assay performance was tested individually for each probe and efficiency , slope and RSq values checked over a standard calibration curve covering a fivefold 4× dilution series . The values were calculated using the MxPro - MX3000P v 4 . 01 software . The specificity of the assay was tested with animals positive for Babesia divergens ( n = 2 ) and Theileria sp . ( n = 2 ) . We identified variables potentially important for the acquisition of trypanosome infections and investigated associations of these with infection events . 72 farmers were interviewed with questionnaires and morphometric measurements of the cattle taken . A Chi squared test was performed in SAS version 9 . 2 . [23] to test for associations between cattle breed and AAT infection . In addition a logistic regression model was calculated using breed , coat colour , gender , age , trypanocidal treatment history and type of trypanosome prevention as fixed effects and village as random effect ( Table 2 ) .
The analysis of the 368 field samples ( 134 Baoulé , 46 Zebu , 188 Baoulé×Zebu hybrids ) with conventional PCR revealed an overall trypanosome prevalence of 10 . 87% ( 40/368 ) . 40 animals were positive for trypanosome DNA . Of these animals two showed multiple infections . Altogether 42 infections in 40 different animals could be detected . T . congolense ( including Savannah and Forest subspecies ) gave an approximate band size of 700 bp ( Figure 3 ) and were detected in nine cases . T . vivax ( 250 bp ) was detected in 33 cases . No infection with T . brucei ( 480 bp ) was discovered with conventional PCR . The highest infection incidences were identified in Zebu cattle with 10 of 46 Zebus ( 21 . 74% ) infected , one of which had multiple infections ( T . congolense/T . vivax ) . None of the 20 Baoulé from the West was infected , in contrast to 13 Baoulé ( 11 . 40% ) from the Southwest adding up to a total prevalence in Baoulé of 9 . 70% ( 13/134 ) . Of the 188 hybrids , 17 animals ( 9 . 04% ) were infected . As expected , no animal from the North regions outside the tsetse belt was positive . By analysing serial dilutions no signal could be obtained from samples containing ≤2 fg DNA . This detection limit corresponds to 0 . 01 parasite genomic equivalents . To increase the specificity of the T . brucei assay and avoid cross-reactions with the T . congolense probe , the primer annealing temperature was increased from 61°C to 65°C . The assay performance of the three probes was assessed in separate reactions . The T . congolense assay efficiency was 97 . 0% , slope Y = −3 . 395 and RSq: 0 . 995 , the T . brucei assay efficiency was 97 . 5% , slope Y = −3 . 384 and RSq: 0 . 988 and the T . vivax assay efficiency was 98 . 3% , slope Y = −3 . 364 and RSq: 0 . 986 . To test for specificity of the qPCR assay Babesia divergens and Theileria sp . positive samples were analysed but no cross-reactivity with related protozoan blood parasites could be detected . After establishing the three assays , the probes were multi-plexed and a total of 368 blood samples from three populations analysed in duplicates . With qPCR a total of 41 animals were tested positive for trypanosome DNA . Of these animals five showed multiple infections . Altogether 46 infections in 41 different animals could be detected ( Table 2 ) . Our results indicate that the qPCR results are comparable to the conventional PCR results with a concordance of 91 . 30% . All positive samples from conventional PCR were verified with qPCR . The overall rate of prevalence detected with qPCR increased to 11 . 14% ( 41/368 ) with 21 . 74% ( 10/46 ) of Zebus infected . Baoulé and hybrids show an infection rate of 9 . 70% ( 13/134 ) and 9 . 57% ( 18/188 ) , respectively ( Figure 4 ) . When analysing the parasitic load of the 46 detected infections , eleven ( two T . congolense , two T . brucei and seven T . vivax ) specimens were PCR-positive but outside the linear range of the standard curve , indicating a low level of parasitaemia ( Table 2 ) . A total of 39 sequences were obtained by sequencing the qPCR products directly or after applying species specific primers to verify the positive results . Of these 14 fulfilled the submission criteria ( ≥200 bp ) and were deposited in Genbank ( JX910370–JX910383 ) . In case of multiple bands ( T . congolense subspecies ) , only the species verified by sequencing was regarded as positive . In our study T . vivax is the dominant species across all analysed area with 34 animals infected ( 9 . 24% ) as compared to T . congolense ( nine animals; 2 . 44% ) and T . brucei ( three animals; 0 . 82%; Figure 2 and 4 ) . None of the animals from the North were positive . For the statistical analysis the variables village , breed , coat colour , gender , age , trypanocidal treatment history and type of trypanosome prevention were used ( Table 2 ) . The results of the questionnaire asking for the trypanocidal treatment history of the cattle revealed that isometamidium chlorides ( Trypamidium , Securidium ) and diminazene aceturates ( Berenil , Diminaveto , Veriben , Trypadim , Survidim ) were applied either by the farmers or by skilled personnel . The types of trypanosome prevention were fighting flies or traditional methods ( e . g . scarification , traditional medicines ) . A chi-squared test linking breeds with infection events ( rates of 21 . 74% for Zebu , 9 . 70% for Baoulé and 9 . 57% for hybrids , respectively ) gave a near-significant result ( p = 0 . 0507 ) . When comparing Zebu with Baoulè and Zebu with hybrids separately , the results were significant ( Zebu – Baoulé P = 0 . 0349 and Zebu – hybrids P = 0 . 0227 ) . However , after correction for multiple testing the results were not significant . In the logistic regression model no significant effects of the seven analysed variables potentially important for the acquisition of trypanosome infections could be detected at a significance level of P<0 . 05 .
AAT is a major constraint on livestock production in Africa [2] . To assess the actual impact of the disease , AAT prevalence was associated with potential risk factors of biology and husbandry of cattle [24] , [25] . Such studies rely on accurate detection methods for circulating parasites . qPCR strategies offer a high level of analytical sensitivity and can be multiplexed with several fluorescent labelled probes . Therefore qPCR is regarded as a key laboratory tool for monitoring parasitic infections [21] . This study introduces a novel qPCR assay for the simultaneous detection of the three AAT species T . congolense , T . brucei and T . vivax . The ITS1 gene , present in around 200 copies per genome , has already been used in conventional PCR studies to detect trypanosome DNA at a dilution equivalent to less than one parasite per ml blood [15] . By using this region we created a highly sensitive , reproducible and specific qPCR assay for the detection and quantification of AAT parasites in cattle blood . The overall prevalence of all 368 analysed animals from 72 farms in Burkina Faso , detected with the novel qPCR assay was 11 . 14% . Previously published studies have reported a similar prevalence of 7 . 54% [26] and 11 . 5% [13] in South-Western areas of Burkina Faso . As shown in our study , the results of the novel qPCR are comparable to the conventional PCR with a concordance of 91 . 30% . However , three additional co-infections and one additional infected animal were detected in the qPCR compared to conventional PCR results ( Table 3 ) . The discrepant results between PCR and qPCR , reflected by the slightly better assay performance of the qPCR , can be explained by low levels of parasitaemia , probably below the detection threshold of conventional PCR . We see that the samples with a positive qPCR result but negative for the conventional PCR present late Cq values and quantities outside the linear range . Positive qPCR results with low levels of parasitaemia have been previously observed where quantification of the parasitic load was unsuccessful due to values outside the linear range of the standard curve [18] . One limiting factor in parasite detection could be the FTA-card sample storage system . FTA-cards offer a convenient method for large-scale prevalence studies [15] . However the parasite DNA is not evenly spread across the filter-paper resulting in inaccurate results [27] . Thus , not only the diagnostic method , but also the procedure of sample preparation is a driving factor for improved sensitivity in epidemiological screening surveys . In our study we used three discs of three FTA card sample areas for DNA extraction . The median level of parasitic load was highest in Baoulé ( 0 . 135 parasite genomic equivalents/rxn ) , followed by Zebus ( 0 . 010 parasite genomic equivalents/rxn ) and hybrids ( 0 . 068 parasite genomic equivalents/rxn ) , calculated according to Duffy et al . , 2009 . The high parasitic load in Baoulé can be accounted to two highly infected animals BD22 and PT24 , probably due to acute infections . It is known that taurine breeds are more tolerant to trypanosome infections and can better cope with the anaemia following high parasitaemia levels [4] , [11] , [28] , [29] . The parasitic load was highest in Baoulé and in hybrid cattle with similar rates of infection ( 9 . 70 vs . 9 . 57% ) . When considering that all included animals were described as healthy by their owners , these results suggest that crossbreeding in the analysed tsetse challenged regions of Burkina Faso was successfully applied as a management tool of AAT . Zebus are more susceptible and twice as often infected ( 21 . 74% ) compared to Baoulé and hybrids . They show significantly higher infection rates compared to Baoulé ( P = 0 . 0349 ) . Trypanotolerance is a multilocus trait with a complex hereditary mode , particularly under field conditions and investigations using quantitative trail loci ( QTLs ) are still on-going [30]–[33] . In a complex statistical model linking breed , village , coat colour , gender , age , trypanocidal treatment history and type of trypanosome prevention , no risk factor could be associated with the infection status in our study . However , for reliable calculations of complex models a higher sample size and higher numbers of infected animals would be needed . The most prevalent AAT species in our sample set was T . vivax with 9 . 24% compared to T . congolense ( 2 . 44% ) and T . brucei ( 0 . 82%; Figure 2 and 4 ) . This is consistent with recent results from the Comoe district in Burkina Faso which show that trypanosome infections were predominantly due to T . vivax [26] , [34] . Interestingly , in Sideradougou T . congolense Savannah type was the predominant species [13] . When looking at Toussiana village located approximately 44 km ( 27 miles ) from Sideradougou in our dataset , we also detected T . congolense Savannah type ( animal HT312 ) without any other trypanosome species present . It is known that different Glossinidae have diverse vector competences for different trypanosome species [35] , [36] . G . morsitans submorsitans is regarded as a good vector for T . congolense Savannah [37] , [38] . An in-depth vector and host surveillance program in this area would be helpful to shed more light on tsetse type abundance and T . congolense Savannah infections . When looking further at the distribution of AAT incidences we noticed one interesting region in Western Burkina Faso . The village of Sindou was sampled with 14 cattle , all belonging to the hybrid group . This village has the highest infection rate in the study , with nine out of 14 cattle being infected . We thus assume that the infection pressure is exceptionally high in this area . In fact , Sindou located in the Cascades province , was affected by increased AAT incidences in the year 2006 resulting in losses of many heads of cattle ( Soudre pers . comm . ) . The present assay can be placed in line with established detection systems for various trypanosome species like T . brucei [20] , [39] , T . cruzi [40] , [41] and T . evansi [18] , [19] . Our novel qPCR assay can detect the three AAT species T . congolense , T . brucei and T . vivax simultaneously . The separate probes can be labelled with different dyes and species determination achieved in only one multiplex qPCR reaction . This method could even be applicable to distinguish trypanosome species not only in cattle hosts , but also in tsetse flies , where morphological traits of the parasite and tissue localisations in the vector are not accurate enough to provide reliable diagnosis [42] , [43] . The taxonomic groups T . congolense Tsavo and Kilifi type and T . simiae have never been identified in vectors and hosts of Burkina Faso [7] , [37] . These species show distinct sequence variations and are not amplified by our novel qPCR assay . In summary , the novel qPCR technique for the simultaneous detection of the three AAT species T . congolense , T . brucei and T . vivax offers a simple , fast and sensitive method compared to conventional PCR . Contaminations can be minimized since single-round PCR reactions are performed and false negative results detected with internal controls . Permanent carrier animals with low parasitaemia levels can be detected and parasitic loads quantified . Detecting the level of parasitaemia is relevant in epidemiological studies to investigate host-vector-pathogen interactions . Our results suggest that breed is a risk factor for AAT infection in contrast to village , coat colour , gender , age , trypanocidal treatment history and type of trypanosome prevention , where no correlations could be found . Zebu cattle are more likely to be infected compared to Baoulé and hybrids . Thus , indigenous taurine cattle breeds , like the Baoulé , provide a unique and valuable genetic resource which needs to be preserved .
|
African Animal Trypanosomosis ( AAT ) is a neglected tropical disease heavily impacting on the poor . Sensitive diagnostic tools are needed since actual parasitaemia levels can be very low , particularly in chronically infected or trypanotolerant animals . Hence , we present a novel real-time PCR ( qPCR ) assay for the simultaneous detection of the three AAT species ( T . congolense , T . brucei and T . vivax ) . Thereby infected animals can be accurately detected in one step . 368 individuals from three cattle types ( Baoulé , Zebu and hybrids ) originating from 72 farms in Burkina Faso were analysed . Farmers were interviewed and morphometric measurements of the cattle taken to detect potential risk factors of infection . In our study , the overall rate of prevalence detected with the novel qPCR assay was 11 . 14% ( 41/368 ) compared to 10 . 87% ( 40/368 ) with conventional PCR . Zebus are most often infected ( 21 . 74% ) compared to Baoulé ( 9 . 70% ) and hybrid ( 9 . 57% ) cattle . Except for breed , no significant correlation with other tested parameters could be detected . Baoulé show significantly less infections and therefore provide a unique and valuable genetic resource . In summary , with this novel qPCR technique the three AAT species can be simultaneously detected in a fast and sensitive manner .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"veterinary",
"diseases",
"veterinary",
"epidemiology",
"veterinary",
"parasitology",
"genetics",
"biology",
"evolutionary",
"biology",
"evolutionary",
"genetics",
"veterinary",
"science"
] |
2013
|
A Novel qPCR Assay for the Detection of African Animal Trypanosomosis in Trypanotolerant and Trypanosusceptible Cattle Breeds
|
Long noncoding RNAs ( lncRNAs ) have emerged as critical factors in many biological processes , but little is known about how their regulatory functions evolved . One of the best-studied lncRNAs is TER , the essential RNA template for telomerase reverse transcriptase . We previously showed that Arabidopsis thaliana harbors three TER isoforms: TER1 , TER2 and TER2S . TER1 serves as a canonical telomere template , while TER2 is a novel negative regulator of telomerase activity , induced in response to double-strand breaks ( DSBs ) . TER2 contains a 529 nt intervening sequence that is removed along with 36 nt at the RNA 3’ terminus to generate TER2S , an RNA of unknown function . Here we investigate how A . thaliana TER2 acquired its regulatory function . Using data from the 1 , 001 Arabidopsis genomes project , we report that the intervening sequence within TER2 is derived from a transposable element termed DSB responsive element ( DRE ) . DRE is found in the TER2 loci of most but not all A . thaliana accessions . By analyzing accessions with ( TER2 ) and without DRE ( TER2Δ ) we demonstrate that this element is responsible for many of the unique properties of TER2 , including its enhanced binding to TERT and telomerase inhibitory function . We show that DRE destabilizes TER2 , and further that TER2 induction by DNA damage reflects increased RNA stability and not increased transcription . DRE-mediated changes in TER2 stability thus provide a rapid and sensitive switch to fine-tune telomerase enzyme activity . Altogether , our data shows that invasion of the TER2 locus by a small transposon converted this lncRNA into a DNA damage sensor that modulates telomerase enzyme activity in response to genome assault .
The discovery of long noncoding RNA ( lncRNA ) has challenged the prevailing paradigm of protein-mediated regulation of gene expression and cell behavior . lncRNAs play essential roles in epigenetic regulation , stem cell biology and signal transduction and are emerging as key targets in human disease [1–3] . Unlike small regulatory RNAs ( e . g . miRNAs , siRNAs ) , lncRNAs are not subjected to purifying selection , and as a consequence they are very poorly conserved , tending to emerge quickly and evolve swiftly [4] . Although transcriptome analyses have uncovered a vast array of lncRNAs , just a tiny fraction of these have an assigned biological function , and fewer still an ascribed molecular mechanism . Little is known about the evolutionary pathways via which lncRNAs gain new functions . The telomerase RNA subunit TER is a lncRNA and an integral component of the telomerase enzyme . TER functions as template to direct the synthesis of telomeric DNA by the telomerase reverse transcriptase TERT . Telomerase continually synthesizes telomeric DNA in stem and germline cells to avert cellular senescence . Conversely , in cells with limited proliferation programs telomerase activity is repressed , an outcome in vertebrates that may have evolved to avert tumorigenesis [5 , 6] . Mechanisms of telomerase regulation are varied and complex , and include modulation of telomerase localization , recruitment to the telomere and enzymology at the chromosome terminus [7] . Within the telomerase ribonucleoprotein itself , the major target of enzyme regulation is TERT . However , TER is also implicated in telomerase control . In addition , different isoforms of core telomerase components influence telomerase behavior [8 , 9] . In conjunction with modulating telomerase action at natural chromosome ends , the enzyme must also be restrained from acting at sites of DNA double-strand breaks ( DSBs ) . Barbara McClintock coined the term “chromosome healing” to describe the acquisition of telomeres on broken chromosomes in maize [10] . Although de novo telomere formation ( DNTF ) protects the terminus from subsequent repair activities , it leads to loss of the centromere distal chromosome fragment . Thus , DSBs must be sheltered from telomerase action to prevent gross chromosomal rearrangements and loss of heterozygosity . Multiple pathways evolved to prevent the establishment of telomeres at DSBs in yeast [11] . For example , phosphorylation of the Cdc13 telomere binding protein decreases its affinity for DSBs [12] . In addition , the Pif1 helicase is activated by DSBs , resulting in removal of telomerase from DNA [13] . Less is known about how DNTF is repressed in multicellular eukaryotes . In mammals , DSBs trigger TERT phosphorylation leading to decreased telomerase activity [14] . In addition , ionizing radiation causes transient sequestration of TERT in the nucleolus [15] . In Arabidopsis thaliana , a non-canonical TER represses telomerase activity in response to DSBs [16] . TER ranges in size from 150 nt in Tetrahymena to >2 kb in certain fungi , and while the nucleotide sequence is highly variable across species , core secondary and tertiary structures are conserved and essential for TER interaction with TERT and for telomerase catalysis [17–21] . TER is transcriptionally regulated in mammals [22] , but the transcript is highly stable with a half-life of several days [23] . Recent data show that that 3’ terminus of Schizosaccharomyces pombe TER is generated by an additional RNA processing step termed slicing , which involves only the first step in mRNA splicing [24 , 25] . Conventional introns have not been associated with TER . Arabidopsis thaliana is unusual in that it harbors two TER genes , TER1 ( 784 nt ) and TER2 ( 748 nt ) [26] . Within TER1 and TER2 , there are two regions of high similarity spanning ~219 nt termed conserved region 1 ( CR1 ) and conserved region 2 ( CR2 ) . In TER2 , CR1 and CR2 are separated by a 529 nt intervening sequence . An additional unique 36 nts lie at the 3’ end of the TER2 CR2 termed 3’R . The intervening sequence and 3’R are removed in vivo to create a truncated isoform called TER2S [16] . Sequences flanking the intervening sequence do not adhere to consensus splice donor and acceptor sites , suggesting that removal of this element may not proceed via conventional mRNA splicing . Although the function of TER2S is unclear , TER1 and TER2 play opposing roles in the control of telomerase enzyme activity . TER1 serves as the canonical telomere repeat template necessary for telomere length maintenance in vivo [26] . Plants deficient in TER1 exhibit progressive telomere shortening , and mutations in the TER1 template alter the telomere repeat sequence in vivo . In contrast , TER2 does not direct telomere repeat incorporation in vivo . Instead , this RNA negatively regulates TER1-mediated enzyme activity . Telomerase activity is elevated in plants lacking TER2 , while in plants over-expressing TER2 , telomerase activity is decreased and telomeres shorten [16] . TER2 is regulated by DNA damage . Under standard growth conditions , the steady state levels of TER1 and TER2S are similar , and 10-20-fold higher than TER2 [16] . However , in response to DSBs , TER2 is rapidly induced and becomes the predominant TER isoform . The increase in TER2 is coincident with a reduction in telomerase activity . Indeed telomerase inhibition is dependent on TER2: ter2 mutants do not down-regulate telomerase in response to DNA damage [16] . Telomerase repression is not elicited by replication stress or telomere dysfunction , indicating that TER2-mediated telomerase regulation is specific for DSBs and thus may play a role in repressing DNTF . While the mechanism of TER2-mediated telomerase inhibition is not known , TERT has a higher affinity for TER2 than for TER1 or TER2S , and preferentially assembles into TER2 containing RNP complexes in vivo . Therefore , TER2 may serve as a molecular sponge to sequester TERT in a non-functional RNP in response to DSBs [16] . TER is evolving rapidly in Arabidopsis and its relatives . Analysis of sixteen closely related species within the Brassicaceae lineage revealed that these species contain a single locus that bears similarity to the 3’ end of TER1 and the 5’ end of TER2 from A . thaliana [27] . Remarkably , several of these TER-like loci lack a template domain altogether , indicating that a functional TER must be encoded elsewhere in the genome . The intervening sequence associated with A . thaliana TER2 is missing from the TER-like genes of other Brassicaceae . Thus , the appearance of TER2 and its intervening sequence represent recent events likely generated during a massive genome rearrangement that occurred on the branch leading to A . thaliana [28] . In this study we employ a comparative genomics approach to investigate the regulatory function of TER2 . Using data acquired from the 1 , 001 Arabidopsis genomes project , we show that the intervening sequence in TER2 has the characteristics of a solo long terminal repeat ( LTR ) from a Copia-like retrotransposon . The element is associated with most , but not all of the TER2 loci . We report that the unique regulatory functions of TER2 , including its responsiveness to DSBs , are derived from this transposable element . Consequently , invasion of the TER2 locus by a transposon transformed this lncRNA into a highly sensitive DNA damage sensor that modulates telomerase enzyme activity .
Since a clear TER2 ortholog could not be discerned in other members of the Brassicaceae , we analyzed genomic sequence data for different A . thaliana accessions , natural strains of A . thaliana collected from the wild . A . thaliana diverged from its closest relative 10 million years ago [29] . It is estimated that Col-0 and Ler-0 , the two best studied A . thaliana accessions , are approximately 200 , 000 years divergent from one another [30] . We retrieved TER1 and TER2 loci from 853 accessions compiled by the 1001 Arabidopsis genomes project ( http://signal . salk . edu/atg1001 ) and analyzed them for variation against Col-0 , the A . thaliana reference genome where a regulatory function for TER2 was first described [16] . The TER1 locus is highly conserved , including the 5’ and 3’ regions flanking CR1 and CR2 ( Fig 1A ) , which lie upstream of the RAD52 coding region or within a predicted intron [27 , 31] . TER1 exhibits 92% identity across the sequenced accessions , but a few polymorphisms are scattered across the RNA ( Fig 1A and 1B , S1A Fig ) . The most notable variations lie within the TER1 template domain ( S2A Fig ) . A transition of A to C occurred three times while a T-A transversion appeared in 44/853 accessions . In neither instance are the two variations found within the same TER1 gene . Because the A . thaliana TER template is 11 nt in length and encodes one and a half copies of the telomere repeat , these TER1 RNAs retain the potential to direct synthesis of TTTAGGG repeats . More intriguing is the C to T mutation in the middle of the template in Bela-1 ( S2A Fig ) . Whether this variation reflects a sequencing error or indicates that an alternative TER1 locus is present in this accession is unknown . Like TER1 much of TER2 is strongly conserved . CR1 retains high percent identity among the accessions ( 92% ) ( Fig 1C ) . CR2 and the 3’R are also very well conserved with complete conservation in >60% of the accessions analyzed ( S3 Fig ) . Conservation of 3’R was unanticipated since this segment of TER2 is eliminated in the production of TER2S ( Fig 1A ) . Nevertheless , the high degree of conservation in CR1 , CR2 and 3’R argues that these regions are important for TER2 function . Although the intervening sequence within TER2 is completely conserved in more than 60% of the accessions , striking sequence divergence was observed in many of the other accessions . Two islands of conservation with ≥ 50% identity were identified within the intervening sequence , one corresponding to 63 nt and a second of 123 nt ( S2B Fig ) . Hyper-variable sequences flank these regions within the 65 accessions bearing an incomplete intervening sequence . To verify the TER2 sequencing data , we performed PCR genotyping on a sampling of accessions predicted to harbor an intact intervening sequence ( Col-0 , Ws-2 ) , a partial intervening sequence ( Aa-0 , Ang-0 , Co-1 and Ei-2 ) or no intervening sequence ( Ler-0 ) . PCR primers were positioned within CR1 and 3’R ( S4A Fig ) . A 784 bp PCR product is expected for accessions bearing an intact intervening sequence , a 255 nt product for accessions completely lacking the intervening sequence , and an intermediate size product for accessions with a partial intervening sequence . Products of the expected sizes were obtained for loci predicted to contain an intact or no intervening sequence , but for all TER2 loci predicted to contain a partial intervening sequence , the genotyping results indicated that this element was completely absent ( S4B Fig ) . Genotyping repeated with siblings from accessions predicted to contain a partial intervening sequence gave the same result ( S4C Fig ) . Genotyping was performed on several additional accessions reported to contain a partial intervening sequence ( S1 Table ) . In all cases , the intervening sequence was absent . Finally , PCR products were sequenced from TER1 and TER2 reactions , with TER1 polymorphisms serving as a control to ensure that seed stocks were as expected ( S4B and S4D Fig ) . The sequencing results confirmed the PCR genotyping data . For all partial intervening sequence accessions tested , there was complete loss of this element . The sequencing data also revealed a substantial deletion ( ~20 bp ) within CR2 in two accessions ( S4D Fig ) . The simplest explanation for these genotyping results is that the TER2 locus was mis-annotated in some of the A . thaliana accessions . However , we cannot exclude the possibility that the intervening sequence within TER2 is extremely labile and between the time the genome sequencing was performed and our acquisition of seeds , partially deleted elements were completely eliminated . For reasons discussed below , we named the intervening sequence within TER2 DSB responsive element ( DRE ) . BLAST analyses against the A . thaliana genome using DRE as a query returned two hits , one on the left arm of chromosome 3 ( adjacent to At3G30120 ) bearing 94 . 6% identity to DRETER2 termed DRE3L , and another on the right arm of chromosome 3 ( adjacent to At3G50120 ) showing 63 . 4% identity called DRE3R ( Fig 2A ) . Both DRE3L and DRE3R are found within intergenic regions and display a number of single-nucleotide polymorphisms among A . thaliana accessions ( S5A Fig ) . BLAST was performed to determine if the DRE is present in other species within the Brassicaceae family . Arabidopsis lyrata , A . thaliana’s closest relative , contains 32 copies of DRE dispersed throughout the genome ( Fig 2B ) . A significant fraction of these elements exhibit a high degree of similarity within the 5’ 200nt of DRETER2 , and are associated with open reading frames encoding typical retrotransposon proteins ( S6 Fig ) . Three DREs were also detected in Capsella rubella , four in Brassica rapa , and ten in Eutrema salsugineum ( Fig 2B ) . The presence of multiple copies of DRE in A . thaliana and its relatives suggests that it is a transposable element ( TE ) . Consistent with this conclusion , sequences at the 5’ and 3’ borders of DRETER2 contain a 5 nt tandem inverted repeat of TGTTG/ACAAC ( Fig 2C , brown bar ) . The tandem inverted repeat at the 5’ and 3’ boundaries of DRETER2 and DRE3L are highly conserved across the A . thaliana accessions and are present at the boundaries of DREs detected in other species ( S6 Fig ) . In addition , a target site duplication of TCGTC is present at the 3’ end of CR1 and the 5’ end of CR2 of TER2 ( Fig 2C , green bar ) . Tandem site duplications flank all three DREs in A . thaliana , ranging in length from 5 nt for DRETER2 and DRE3L to 18nt for DRE3R ( Fig 2C , green bar ) . The tandem site duplication sequence varies , consistent with the hypothesis that these insertions represent unique TE insertion events rather than gene duplications . The small size of DRE and its association with tandem inverted repeats and target site duplications suggest that DRE is derived from a solo LTR of the abundant Copia family . Based on synteny mapping with Arabidopsis lyrata we confirmed that all three Copia-like solo LTRs in A . thaliana ( TER2DRE , DRE3L , and DRE3R ) are unique insertion events and are of approximately the same age ( S7 and S8 Figs ) . Since the large majority of A . thaliana accessions apparently harbor an intact DRE within the TER2 locus , it is likely that the element was inserted soon after the TER duplication and was subsequently lost in a small subset of accessions . The presence of two distinct TER2 alleles in A . thaliana provided us with an opportunity to study the functional impact of DRE . We previously showed that two RNA transcripts are derived from the Col-0 TER2 locus: the primary TER2 transcript and a processed isoform , TER2S , in which DRETER2 is removed along with 3’R [16 , 26] . In the Ler-0 accession , the TER2 locus lacks DRE , and thus the primary transcript is predicted to be TER2Δ . To assay for TER2Δ , RT-PCR was performed on RNA from Ler-0 seedlings using primers directed at CR1 and 3’R , which is unique to TER2 ( Fig 3A and 3B ) . A product of the expected size was generated , indicating that a Ler-0 transcript containing CR1 , CR2 and 3’R is present . Sequence analysis confirmed this conclusion . Notably , the CR1/CR2 junction in Ler-0 TER2Δ is distinct from Col-0 TER2S [26] as it contains only a single 5’ TCGTC 3’ motif instead of the two found in Col-0 ( Fig 3B bottom , underlined sequence ) . Although a faint signal for TER2 was observed in Col-0 using our PCR conditions , TER2Δ was not ( Fig 3A ) , suggesting that TER2Δ is either a transient processing intermediate , or is not generated during the conversion of TER2 to TER2S . Col-0 TER2 is a poorly expressed transcript ( Fig 3A ) and is substantially less abundant than TER1 or TER2S [16] . To assess the relative abundance of Ler-0 TER2Δ , we performed qPCR ( Fig 3C ) . The steady state level of TER1 was similar in Ler-0 and Col-0 . However , Ler-0 TER2Δ was approximately 6–8 fold more abundant than Ler-0 TER1 . By comparison , Col-0 TER2 was 15–20 fold less abundant than Col-0 TER1 ( Fig 3C ) . Thus , Col-0 TER2 and Ler-0 TER2Δ are differentially regulated in vivo . In Col-0 , TER2 but not TER1 or TER2S is rapidly induced by DSBs [16] . Therefore , we asked if regulation is confined to TER2 by examining TER2 and TER2Δ in other A . thaliana accessions ( Fig 4A ) . Seven day-old Ler-0 and Col-0 seedlings were treated with 20μM zeocin and qPCR was performed . In control reactions , BRCA1 mRNA was induced in both accessions after 2 hours and peaked at 4 hours , confirming that a DNA damage response was elicited ( Fig 4B ) . As expected , the level of TER1 was unchanged in Ler-0 and Col-0 following zeocin treatment ( S9 Fig ) . In addition , Col-0 TER2 increased 2 . 5 fold after 2 hours in zeocin relative to untreated seedlings ( Fig 4C ) . In marked contrast , there was no significant change in TER2Δ over the 4 hour zeocin treatment ( Fig 4C ) . To test if DSB-mediated regulation of TER2 is a peculiarity of the Col-0 accession , we examined TER2/TER2Δ transcripts in two additional accessions: Ws-2 , which contains DRETER2 and Co-1 , which lacks it ( Fig 4D ) . Consistent with the findings in Ler-0 and Col-0 , there was no change in Co-1 TER2Δ , while Ws-2 TER2 was induced ( Fig 4D ) . We conclude that the effect of DSBs on TER2 varies across A . thaliana accessions , and correlates with the presence of DRETER2 . We next asked if transcripts were derived from the other two DRE-like sequences in Col-0 , and if so whether they responded to DSBs . Semi-quantitative RT-PCR was performed with primers specific for DRE3L and DRE3R on seedlings in the presence or absence of zeocin ( S5B Fig ) . DRE3L transcripts could not be detected under either condition . However , transcripts from DRE3R were observed in the presence of zeocin ( S5B Fig ) , indicating that a DNA damage-sensing element resides within DRETER2 as well as DRE3R We previously showed that Col-0 TERT displays a hierarchy of binding favoring TER2 > TER1 >> TER2S both in vitro and in vivo [16] . The molecular basis for the enhanced affinity of TERT for TER2 is known . Since DRETER2 and the 3’R are unique to TER2 , it seems likely that one of these elements influences TERT binding . To investigate this possibility , we examined the relative affinity of TERT for TER2Δ . Col-0 and Ler-0 seedlings were subjected to immunoprecipitation with TERT antibody followed by qPCR ( Fig 4E ) . We set the ratio of TER2 to TER1 in the Col-0 TERT IP to 1 , and then assessed the change in TERT-bound TER2 following zeocin treatment . The relative abundance of TER2 containing TERT complexes increased ~ 7-fold in response to DSBs ( Fig 4E ) . Since the input level of TER2 increased by only 2 . 5-fold under these conditions ( Fig 4C ) , the data raise the interesting possibility that other DNA damage-induced factors promote TER2 assembly with TERT . In marked contrast to TER2 , we found that TER2Δ is not a preferred binding partner for TERT in vivo , and further zeocin treatment did not change the relative abundance of TER2Δ containing TERT complexes ( Fig 4E ) . These results argue that the increased affinity of TERT for TER2 in Col-0 reflects the presence of DRETER2 and not 3’R . Since Col-0 plants lacking TER2 do not down-regulate telomerase activity in response to DSBs [16] , we asked if DSB-induced telomerase regulation is dependent on DRETER2 by comparing the level of telomerase activity in Ler-0 and Col-0 in the presence of zeocin . As expected application of quantitative telomere repeat amplification protocol ( qTRAP ) to Col-0 seedlings treated with zeocin for 2 or 3 hours showed reduced telomerase activity ( 70% decrease ) compared to untreated seedlings ( Fig 4F and S10 Fig ) . Although there was an alleviation of the inhibitory effect after 3–4 hours of treatment , enzyme activity was maintained at 50% of untreated level . In contrast , under the same treatment regime , telomerase activity was unaltered in Ler-0 ( Fig 4F ) . Similar results were obtained with Ws-2 ( plus DRETER2 ) and Co-1 ( minus DRETER2 ) accessions , respectively ( Fig 4G ) . These findings imply that DRE is necessary for DSB-induced telomerase repression . To further assess the role of DRE in telomerase regulation , we generated two transgenic A . thaliana lines . First we asked if the presence of TER2 was sufficient to alter the level of telomerase activity in Ler-0 by expressing TER2 from its native promoter in this accession . In one of the transformants , the steady state level of transgenic TER2 was higher ( 2 . 5 fold ) than the basal level of endogenous TER2 in wild type Col-0 ( Fig 5A ) . qTRAP revealed a small , but statistically significant decrease in telomerase activity in the transformant ( Fig 5B ) , indicating that Ler-0 telomerase can be down-regulated by Col-0 TER2 . Next we asked if over-expression of TER2Δ altered telomerase activity in Col-0 . TER2Δ expression was driven by the powerful CaMV promoter in wild type Col-0 . As expected , there were no change in TER1 or TER2 , but the steady state level of transgenic TER2Δ was ~8-fold higher than endogenous TER2Δ in wild type Ler-0 . However , qTRAP showed no change in telomerase activity relative to untransformed Col-0 controls ( Fig 5A and 5B ) . We conclude that the regulation of telomerase by TER2 is dependent on DRETER2 . The rapid induction of Col-0 TER2 in response to DSBs could occur through increased TER2 transcription or increased RNA stability . Because the sequences upstream of all TER2 genes are highly conserved , we considered the former possibility less likely . Indeed , when TER2 transcription was monitored in seedlings expressing a fused GUS reporter to a TER2 or TER2Δ promoter in Col-0 and Ler-0 , respectively , approximately the same level of GUS staining was observed in the presence or absence of zeocin ( S11 Fig ) . Hence , TER2 induction in response to DNA damage is not caused by increased transcription . We assessed TER2 stability using six day-old seedlings treated with the transcription elongation inhibitor cordycepin . TER1 and TER2 RNA levels assessed by qPCR showed that Col-0 and Ler-0 TER1 have similar half-lives , t1/2 = 75 and 84 min , respectively ( Fig 6A ) . The stability of TER2Δ was even greater with t1/2 = 244 min ( Fig 6B ) . TER2 , on the other hand , had a much shorter half-life than either TER2Δ or TER1: TER2 t1/2 = 13 min ( Fig 6B ) . Thus , TER2 is an intrinsically unstable transcript . To test if DSBs reduce TER2 turnover , Col-0 seedlings were treated with cordycepin to pause transcription and then zeocin was added after 90 min to produce DSBs . Although there was a slight change in the abundance of TER1 and BRCA1 mRNA in the presence of zeocin , this change was not statistically significant ( Fig 6C and 6D ) . In contrast , TER2 abundance declined sharply over the 3 . 5 hour time course , but immediately after the introduction of zeocin , TER2 was stabilized ( Fig 6E ) . These data implicate DRETER2 as the causal factor in destabilizing TER2 and in turn negatively regulating telomerase activity during bouts of DNA damage .
When the insertion of a TE within or adjacent to a gene leads to a change in gene function the process is termed “exaptation” [32] . Exaptation can alter gene regulation through myriad different mechanisms . A prominent example in plants is the insertion of multiple TEs adjacent to teosinte branched1 ( tb1 ) , which gave rise to domesticated maize [33] . One of the TEs disrupts a regulatory region of tb1 , leading to increased expression and enhanced apical dominance . In vertebrates , exaptation of TEs is more prevalent at lncRNA loci than in protein-coding genes [34] . Approximately 41% of vertebrate lncRNA sequence is derived from TEs [35 , 36] , leading Johnson and Guigo to propose that TEs can behave as pre-formed functional RNA domains , and further that exaptation of TEs is a major driving force in lncRNA evolution [36] . A recent systematic survey in vertebrates catalogued multiple instances of TEs altering lncRNA promoters , splice sites , and polyadenylation sites [37] . LncRNAs can also acquire novel interaction partners as a direct result of TE exaptation [32] . For instance , TEs within XIST facilitate interaction with a host of protein complexes including PRC2 and splicing factor ASF2 [38] . Here we show that invasion of a small TE ( DRE ) into the A . thaliana TER2 locus profoundly altered the function of this lncRNA ( Fig 7 ) . This exaptation event is not fixed , as the TER2 genes in 9% of the 853 accessions examined lack DRE . Insertion and subsequent loss of TEs is not uncommon in Arabidopsis . Some 80% of the annotated TEs in A . thaliana were lost in one or more accessions [39] . In the 200 , 000 years since Col-0 and Ler-0 diverged , at least 200 TEs have been active , and the unique insertions/deletions between the two accessions have biological implications [30] . One illustrative example of TE exaptation occurred at the Flowering Locus C ( FLC ) in Ler-0 . Insertion of a Mutator-like transposon in this accession decreased FLC transcription , causing early flowering [40] . In this study we exploited the natural genetic heterogeneity within the TER2 locus , and discovered that many of the unique functions ascribed to this lncRNA derive from DRE . First , DRE destabilizes TER2 . A survey of ~800 lncRNAs in mouse revealed that only a small fraction are unstable , defined as RNAs with a half-life of less than 60 minutes [41] . By this criterion , TER2 is a highly unstable transcript with a half-life of only 13 minutes ( Fig 7 ) . TER1 ( t1/2 = 80 min ) and TER2Δ ( t1/2 = 240 min ) , on the other hand , are categorized as stable RNAs . Unstable lncRNAs , like their unstable mRNA counterparts , are typically associated with regulatory functions , while stable RNAs are thought to serve housekeeping roles [42] . With Col-0 A . thaliana TER1 and TER2 , this paradigm also holds . A second key observation is that the instability of TER2 arising from DRE is reversed in response to DNA damage ( Fig 7 ) . The abundance of TER2 , but not TER1 or TER2Δ is elevated in response to DSBs , and this change is largely , if not entirely , dependent on RNA stabilization rather than new transcription . Exaptation of a TE is known to endow host genes with the capacity to respond to environmental cues . For example , a cold-sensitive TE was inserted into the promoter of Ruby , a transcription factor that regulates flesh color in Citrus sinensis ( blood orange ) . Cold activates the transposon , which in turn activates Ruby and downstream anthocyanin production [43] . In the case of TER2 , DRE imparts DNA damage sensitivity , which increases TER2 abundance . How TER2 is regulated in response to DSBs is unknown . One possibility is that DRE carries binding sites for one or more interaction partners responsive to DNA damage , which then stabilize TER2 . RNA binding proteins can play a significant role in the DNA damage response by regulating specific target genes post-transcriptionally [44] . TER2 turnover might be controlled through the small RNA regulatory pathway . A 24 nt RNA is associated with DRETER2 [45] . This finding is particularly intriguing given the recent discovery that small RNAs modulate the response to DSBs in both vertebrates and Arabidopsis [46] . Finally , it is possible that DNA damage blocks the RNA processing steps ( e . g . splicing ) that lead to production of TER2S ( Fig 7 ) . Splicing machinery has emerged as a target of the DDR [47] . The third key observation from this work is that DRE increases the affinity of TER2 for TERT ( Fig 7 ) , and correlates with the down-regulation of telomerase activity . DRE could modify TER2 structure in a manner that enhances its inherit affinity for TERT . Alternatively , DRE may make independent contacts with TERT to increase TER2 affinity . Intriguingly , zeocin treatment causes an even greater enrichment of TER2 containing TERT complexes than expected based on the fold induction of TER2 , suggesting that a TER2 associated factor that is also responsive to DNA damage might drive the assembly of TER2-TERT RNPs . Altogether , our data are consistent with a model in which exaptation of a TE into the A . thaliana TER2 locus gave rise to a new mode of telomerase regulation . Specifically , we propose that the DRE converted TER2 into a DNA damage sensor that controls telomerase enzyme activity through sequestration of TERT . Furthermore , because this regulatory pathway is regulated by changes in RNA stability , it is both rapidly responsive and reversible , allowing the A . thaliana accessions that carry DRE to fine-tune telomerase activity when the plant is under genome assault . These discoveries provide a fresh perspective on the role of TE exaptation in shaping lncRNA function and evolution .
For experiments with seedlings , seeds from different accessions ( Col-0 , Ler-0 , Ws-2 , etc ) were sterilized in 50% bleach with 0 . 1% Triton X-100 and then stored in 4°C for 2–4 days . Liquid Murashige and Skoog ( MS ) medium were used for germination and growing [16] . After transferring cold-treated seeds to MS , plants were grown at 22°C under long day light condition for ~7 days . The Col-0 TER2 gene including 3kb upstream sequence and 300bp downstream sequence was cloned in the pMDC99 vector for transformation in the Ler-0 background . Hygromycin MS plates were used for selection . For Col-0 transformation , TER2Δ together with 300 bp downstream flanking region was cloned into the pBA002 vector with 35S promoter . BASTA MS plates were used for the selection . Sequences corresponding to TER2 ( Genbank accession number: HQ401285 . 1 ) were obtained using the genome browser at http://signal . salk . edu/atg1001 . The search query AT5G24660 was used to pinpoint the region of interest , and all available tracks ( accessions ) were selected . Two sequences were removed from our analysis . Hov 3–2 was removed because it was the only accession with two deletions in the 5’ end , corresponding to 20 nt from the 5’ start of TER2 , and a 100 nt deletion starting at nucleotide #101 . The template region was not disturbed in this accession , possibly indicating a functional TER2 is generated . The Tottarp-2 accession was removed because the sequence corresponding to our search region did not contain sequences corresponding to TER2 , most importantly , a template region . Sequences were trimmed in MEGA5 , and then analyzed using Geneious v6 . 0 ( Biomatters ) . Sequence conservation and alignments were performed using Geneious . DRE-like sequences were obtained by BLAST searches of the A . thaliana ( www . arabidopsis . org ) , A . lyrata , Capsella rubella , Brassica rapa , and Thellungiella halophila genomes accessed via www . phytozome . net v9 . 1 [48 , 49] . A . thaliana seedlings ( 5–7 day old ) were transferred to fresh MS liquid medium with 20 μM zeocin ( Invitrogen ) as described [16] . Seedlings were kept in the dark with gentle agitation for 1 , 2 or 4 h . Multiple seedlings were combined and flash frozen in liquid nitrogen for RNA extraction or protein extraction for TRAP . The combined sample was treated as a single biological replicate . DNA samples were prepared from the leaves of different accessions . Both TER1 and TER2 loci were used for genotyping . PCR samples were resolved in 1% agrose and gel purified and sequenced . RNA was extracted from seedlings using the Direct-zol RNA MiniPrep kit ( Zymo Research , Epigenetics ) according to the manufacturer’s instructions . 1 μg total RNA was used for preparing cDNA . For RT-PCR , cDNA was synthesized by SuperscriptIII Reverse Transcriptase ( Invitrogen ) using random primers . For qRT-PCR , reverse transcription was performed using the Superscript cDNA master mix ( Quanta ) , according to the manufacturer’s instructions . 1:5 diluted cDNA was used for qPCR . qPCR was performed on a Bio-Rad CFX-1000 using the following primers: qTER2Δ F: 5’-AGAACGTTGACGGCTAAAGG-3’; qTER2Δ R: 5’- TGTGGCATAAGGCAAACTGA-3’; TER2 , BRCA1 , TER1 and GAPDH primers are used as described before [16] . Data were analyzed using Bio-Rad’s CFX manager software . ΔΔCT values were obtained by comparing against GAPDH levels . qTRAP assays were performed as described [50] . Data were normalized against untreated Col-0 . For immunoprecipitation , TERT antibody [50] was conjugated with Dynabeads Protein A ( Invitrogen ) then incubated with protein extracts in 4°C . RNA was recovered from the IP sample using phenol/chloroform followed by ethanol precipitation [16] . qPCR was performed on TER1 and TER2/TER2Δ . The ΔCT value was used to determine the relative level of TER2 or TER2Δ against TER1 . 5–7 day old seedlings were treated with cordycepin ( 100 ng/μl as a working concentration ) for 2 h before RNA extraction . RNA was analyzed by qPCR normalized to eIF-4a [51] . RNA abundance was converted to the decreased level relative to untreated . RNA half-life was determined by the absolute value of inverse of the slope of the equation plotted by untreated and treated data . For the combined cordycepin/zeocin experiment , seedlings were pre-incubated with cordycepin for 1 . 5 h followed by zeocin and the incubation was continued for 2 h . RNA extraction and qPCR were used to determine RNA abundance . RNA half-life was determined by plotting RNA abundance versus time as described in [51] . 3 kb of sequence upstream of the TER2 5’ terminus was cloned in a GUS reporter vector pMDC163 . The construct was transformed into A . thaliana Col-0 and Ler-0 as described [52] . After selection in hygromycin , transformants seedlings were treated with zeocin for 2 h and then subjected to GUS histochemical staining as described [53] .
|
Telomerase is a highly regulated enzyme whose activity is essential for long-term cellular proliferation . In the presence of DNA double-strand breaks ( DSBs ) , telomerase activity must be curtailed to promote faithful DNA repair . We previously showed that the flowering plant Arabidopsis thaliana rapidly down-regulates telomerase in response to DSBs , and further that this mode of regulation is dependent on TER2 , a non-canonical telomerase RNA subunit . Here we demonstrate that the unique regulatory properties of TER2 are conveyed by a transposable element ( TE ) embedded in the TER2 gene . A comparison of A . thaliana accessions with and without the TE revealed that the element increases the binding affinity of TER2 for the telomerase catalytic subunit TERT relative to the canonical telomerase RNA subunit . The TE also increases TER2 turnover . In response to DSBs , TER2 is induced and accumulates in TERT containing complexes in vivo . Thus , invasion of a TE endows TER2 with a DNA damage sensor to rapidly and reversibly modulate enzyme activity in response to genotoxic stress . These findings provide an example of how exaptation of a TE altered the function of a long noncoding RNA . In this case , a duplicated gene ( TER2 ) was used as the platform , and the TE as the tool to engineer a novel mode of telomerase regulation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
A Transposable Element within the Non-canonical Telomerase RNA of Arabidopsis thaliana Modulates Telomerase in Response to DNA Damage
|
Biological protein interactions networks such as signal transduction or gene transcription networks are often treated as modular , allowing motifs to be analyzed in isolation from the rest of the network . Modularity is also a key assumption in synthetic biology , where it is similarly expected that when network motifs are combined together , they do not lose their essential characteristics . However , the interactions that a network module has with downstream elements change the dynamical equations describing the upstream module and thus may change the dynamic and static properties of the upstream circuit even without explicit feedback . In this work we analyze the behavior of a ubiquitous motif in gene transcription and signal transduction circuits: the switch . We show that adding an additional downstream component to the simple genetic toggle switch changes its dynamical properties by changing the underlying potential energy landscape , and skewing it in favor of the unloaded side , and in some situations adding loads to the genetic switch can also abrogate bistable behavior . We find that an additional positive feedback motif found in naturally occurring toggle switches could tune the potential energy landscape in a desirable manner . We also analyze autocatalytic signal transduction switches and show that a ubiquitous positive feedback switch can lose its switch-like properties when connected to a downstream load . Our analysis underscores the necessity of incorporating the effects of downstream components when understanding the physics of biochemical network motifs , and raises the question as to how these effects are managed in real biological systems . This analysis is particularly important when scaling synthetic networks to more complex organisms .
A longstanding question about signal transduction and gene transcription networks is how modular are they . Here modularity means relative insulation of small subgraphs or motifs of the main network from each other [1] . This question is especially relevant for synthetic biology that aims to build artificial circuits from the bottom up [2] . It is also relevant for molecular biologists that aim to arrive at a quantitative understanding of a cellular decision , by , for example , isolating a crucial network module [3] . For synthetic biologists the challenge is now to move from simple network motifs such as pulse generators [4] , genetic switches [5]–[8] , logic gates [9] , [10] , and oscillators [11]–[13] to more complicated networks combining multiple motifs and networks in more complex organisms . Novel applications currently being explored include plant biosensors [14] , hazardous waste remediation [15] , clean fuel technology [16] , and numerous medical applications [17]–[20] . Synthetic biologists hope to utilize biological modules in a manner similar to electrical circuit board components – plugging them together to attain a specific , and novel , function [21] . At the core of the concept of either breaking down complex biological systems into small modules , or even building complex systems from modules , is the belief that these modules will behave predictably in isolation and in connection . Recent theoretical and experimental work however [22]–[25] suggests that the functioning of modules may not be independent of the downstream components that they are connected to . Adding an additional binding reaction to the output of a gene regulatory network ( or loading the network ) may decrease system bandwidth [24] and substrate sequestration in covalent modification cycles may result in signaling delay [26] . In vitro studies find that there is significant load-induced modulation of the upstream module in an enzymatic signal transduction cascades [24] . Theoretical analysis has also shown that a load can change the fundamental properties of an oscillating circuit [27] . Thus understanding the effects of adding a load to the output of these technologically important network modules is required for a thorough understanding of the challenges of scaling up synthetic networks to higher levels of complexity . Loads could also have noteworthy unrecognized effects in natural systems . In fact all natural systems have loads in some ways or the other . Motifs in signal transduction networks are connected directly to a transcriptional response , or to downstream proteins that may function as transcription factors or go on to activate transcription factors . Motifs in gene transcription networks have transcriptional outputs with protein domains that bind nonspecifically and specifically to binding sites on the DNA , apart from interacting with other transcription factors . Circuits that function as switches play an important role in all biological signaling and gene transcription networks because they encode decisions . This change of state can be brought about by an external signal , or an internal accumulation of a protein , which can drive the system to a different steady state . Examples are the regulatory circuits for the cell cycle in yeast [28] , mitogen-activated protein kinase cascades in animal cells [29]–[31] , and the lysis-lysogeny switch in the λ phage [32] . Since many small circuits can show this kind of behavior , switches are among the earliest and most well studied of protein interaction circuits [33] . The genetic toggle switch , which was one of the first two synthetic circuits constructed , is a well-known synthetic example [5] . Given the ubiquity and importance of switch-like motifs , it is important to understand how their function could be affected by binding downstream partners . These reasons prompted our theoretical study of the behavior of a simple genetic toggle switch [5] , a toggle switch with positive feedback as well as a common positive-feedback based switch involving Ras activation in lymphocytes [29] , [30] under a load on either one or both of its outputs . These circuits are shown in Fig . 1 and described below . The simple toggle switch is a widely studied and emulated synthetic network motif based on the mutual repression of two repressor proteins . However , naturally occurring toggle switches are often found connected to an additional positive autoregulatory component . For example in the competence system in B . subtilis , ComK represses the production of Rok and Rok represses the production of ComK; however ComK also has a strong positive feedback upon its own production [34] . Another example is found in the apoptosis network of many multicellular organisms , including mammals . Within the pathway controlling intrinsic apoptosis is a set of genes with double-negative repression , Casp3 and XIAP , again accompanied by positive autoregulation of Casp3 [35] . The Ras protein is a G-protein found on mammalian cellular membranes that is important in many cellular processes and is an upstream activator of the MAPK pathway . Ras goes from a GDP-bound inactive form to a GTP-bound active form , often in a digital manner [30] , and previous studies in lymphocytes have shown that RasGDP is activated to RasGTP via a bistable switch that arises from a positive feedback loop on its own activation via SOS ( Son of Sevenless ) [30] . However the Ras switch very naturally has an associated load , since to transduce the cellular signals down along the MAPK/ERK pathway , RasGTP naturally binds to Raf kinase . Thus the Ras switch system contains all the elements we need to study the effects of adding loads to a bistable switch which is based on a positive feedback loop .
The basic genetic toggle switch consists of two mutually repressing genes as shown in Fig . 1 along with an additional system to toggle the states . As shown in previous studies , with the right combination of parameters , the toggle switch will stay in one of two stable states , each characterized by a high concentration of one of the repressor proteins , and strong repression of the other . The toggle switch can now be induced to switch states using two possible strategies for inducing a transition: decrease the level of highly expressed protein [5] , [36] or increase the expression of one of the repressed proteins ( Fig . 1 ) using an additional inducible system [36] . For a model which utilizes the latter protocol we obtain a system of four differential equations [36] after including a load . The load may be a protein , a small molecule or a binding site on DNA such that the bound complex prevents the repressor from binding to and repressing its conjugate promoter . In order to make the simplest and the most general model , we have assumed here that the repressors reversibly bind the load only in one copy . We assume that the total load L1T is a constant , L1 is the free load and conservation gives us the bound load as L1T−L1 . ( 1 ) ( 2 ) ( 3 ) ( 4 ) These four equations are presented in de-dimensionalized form , with representing the dimensionless concentrations of Repressor 1 , Repressor 2 , Load1 and Load2 respectively and τ the de-dimensionalized time . The basal parameter values that we use are as follows: α1 = α2 = 0 . 2; β1′ = β2′ = 4; n = 3; kon1′ = kon2′ = 0 . 5; koff1′ = koff2′ = 0 . 5; k1 = k2 = 1; [L1T] and [L2T] are variable . Note that Equations ( 1 ) and ( 2 ) without the last two terms incorporating the load are the standard equations for analyzing the toggle switch that have been widely used in both empirical and theoretical work [5] , [36] . These equations are discussed in more detail in Supplementary Text S1 Section 1 . 1 . The derivation of this model follows that of Kobayashi et al [33] . All parameters excluding load binding rates were sourced from Kobayashi et al [36]; extensive parameter sensitivity of the load binding rates was performed and are discussed in the Supplementary Text S1 section 1 . 4 and Figs . S1 , S2 , Table S1 and Figs . S15 and S16 . The effect of a load arises from the binding competition between the promoter where the repressor binds and the load . This competition is not directly incorporated into the Hill function , since the binding step with the promoter is not explicitly modeled and is treated in an effective way . In reality however the concentration of the promoter is so small compared to that of the load , that the use of Hill functions is justifiable [37] . There are possibly exceptional cases such as a high copy number of plasmids compared to load concentrations where this assumption does not apply . Note that the Hill function is an effective phenomenological equation describing gene transcription and protein production , and standard Law of Mass Action ( LMA ) methods to derive the Hill functional form may not apply for many transcription factors that nevertheless show Hill kinetics [38] . Thus it is preferable to use Hill function forms for this analysis . To calculate transition times , we first start the system in one state , say high Repressor 1 . After the system has reached steady-state , we add a constant concentration of the inducer and measure the time taken for Repressor 2 to go from 10% of its maximum value to 90% of its maximum value . This is the “rise time” . Similarly the “decay time” is the time taken for Repressor 1 to go from 90% of its maximum value to 10% of its maximum value . The level of the inducer remains fixed . In practice the inducer may decay and the transition would depend upon there being inducer present for a sufficiently long time to induce transition . In such cases the amount of inducer required may be of interest . When the inducer is applied as a bolus with a first order decay rate , it appears as an exponentially decaying pulse . We thus included a fifth differential equation governing the amount of Inducer . ( 5 ) Here is the ratio of the inducer degradation constant to the repressor degradation constant . We used Eq . 5 only when estimating the amount of inducer required to switch states for different loads and different decay rates of the load ( Supplementary Text S1 section 1 . 4 and Supplementary Tables S3 , S4 ) . A genetic toggle switch can be induced to change states by the alternative method of repressing the highly expressed repressor , and in fact the original toggle switch used this form of induction [5] , [33] . We repeated our calculations for the basic model for the case of alternative induction , but found no qualitative differences . The alternative induction model along with the equations is detailed in the Supplementary Text S1 section 1 . 4 . Equations 1–4 assume that the load itself stays in steady state during the switching of the toggle between one state and another . However in reality if the load is another protein , it is also synthetized and degraded by the cell , and therefore its level could be dynamic . We also simulated this situation by incorporating a synthesis and a degradation rate for each load . This resulted in Equations 3 and 4 being replaced by: ( 6 ) ( 7 ) Here is the load-repressor complex and and are the synthesis and degradation rates respectively for Repressor 1 , and correspondingly for Repressor 2 . The parameters are defined in the Supplementary Text S1 , section 1 . 5 . Since the total load is no longer conserved , we need to include additional equations for the load repressor complex . ( 8 ) ( 9 ) Our model assumes that when the repressor protein is bound to the load , it is protected from degradation . However it is also possible that even when the protein is bound to the load , it can still degrade . To check the impact of removing the protection assumption , we also consider an additional model where the repressor can still degrade with the same rate constant when bound to the load . The equations for that model are slightly modified versions of the equation above , and are presented in detail in the Supplementary Text S1 , section 3 . 2 . We conducted parameter sensitivity analysis on models utilizing both forms of induction; these did not show any qualitative change on wide variation of key parameters ( Tables S1 , S2 , S3 , S4 and Supplementary Text S1 ) . A positive feedback was added to the R1 side of the toggle switch as an inducible promoter with a Hill coefficient of 1 . We assumed that the positive feedback acted on the same promoter as the repression , resulting in a composite term for production of R1 from promoter 1 where ρ is the strength of positive feedback . ( 10 ) The derivation of this equation can be found in the Supplementary Text S1 , section 1 . 6 . 1 . As before , α1 is the leaky production of R1 while α1+β1 represents the activity of the promoter in the absence of repression or positive feedback . We chose k2 and k5 = 1 , d1 = 0 . 2 , and for the figures in the main paper we chose ρ = 3 . 5 . We address other values of the positive feedback in Fig . S6 and the Supplementary Text S1 , section 1 . 6 . 2 . We perform stochastic simulations and histogram the concentrations of the repressor proteins to construct their probability distribution . The quasi-potential of the toggle is given by the negative logarithm of this probability distribution [39] . In order to construct the probability distribution we make use of the phenomenon of noise-induced switching . Recent theoretical work has shown that multiplicative noises due to stochastic fluctuations can induce switching [40]–[42] . Experimental results demonstrate bimodal populations that correspond with theoretic predictions arising from noise-induced switching [41] . Stochastic simulations were carried out using a modified Gillespie algorithm using the standard rate expressions for every reaction ( Table S5 ) . We chose a reaction volume that would correspond to a small number of molecules in the system . Stochastic fluctuations then drive the system to transition between states rapidly , allowing us to collect sufficient data points . In order to make sure that the system was not being biased by the small volume , we also repeated the calculations for a five times larger volume ( and hence molecule number ) and found qualitatively similar results ( Fig . S4 ) . For the positive feedback toggle switch the same equations were used except for the repressible production of Repressor 1 , where we used instead the rate expression given by the right hand side of Eq . 10 in the Monte Carlo simulations . For our study we adapted the minimal model of the Ras switch proposed by Das et . al . [30] with the addition of a reversibly binding load in the form of the Raf protein ( Fig . 1C ) . The model contains three proteins , Ras , which exists as RasGDP or RasGTP , SOS , the guanine exchange factor ( GEF ) that catalyzes the transformation from RasGDP to RasGTP and a GTPase , RasGAP . SOS on its own has very low GEF activities . However , the activity of the GEF pocket is strongly influenced by the binding state of an allosteric pocket in Cdc25 domain [29] , [30] . When the allosteric pocket is bound by RasGDP , the GEF activity is increased by 5 times . If the allosteric pocket is bound by RasGTP , its GEF activity is increased by 75 times . In this way , RasGTP can upregulate its own production rate by binding to SOS , thus constituting a positive feedback loop . RasGTP is deactivated by GTPase's such as RasGAPs that are constitutively present . After Raf binds RasGTP , the complex catalyzes the phosphorylation of Raf leading to a phosphorylation cascade . For this study we ignore Raf activation and only consider the effects of Raf as a binding partner for RasGTP . The Das paper [30] also models the systems using Michaelis-Menten ( MM ) forms for the actions of the enzymes which is quite standard for modeling systems of enzymatic reactions . However since in this model the load competes not with a promoter , as in the toggle switch , but with another protein , it is possible that the quasi-steady state assumption of the MM form could be introducing some inaccuracies in the results . To account for this possibility we wrote the entire model using the Law of Mass Action . We separately simulated the model using the MM functional forms ( Supplementary Text S1 section 2 and Figs . S7 and S9 ) . The equations for the MM forms are listed and discussed in detail in the Supplementary Text S1 . The reactions and rate constants for this model are listed in Table S6 and Table S7 . We use the following notations for the species involved in the system: ( 11 ) ( 12 ) ( 13 ) ( 14 ) ( 15 ) ( 16 ) ( 17 ) ( 18 ) ( 19 ) ( 20 ) ( 21 ) Moreover , four of the basic protein species along with the complexes they participate in have associated conservation laws . These are as follows: ( 22 ) ( 23 ) ( 24 ) ( 25 ) In the Ras model too we implicitly assume that when RasGTP is bound to Raf , it is protected from de-activation by a RasGAP . We also study the effects of relaxing this assumption on both the LMA and the PSSA models . The modifications to the original model are detailed in the Supplementary Text S1 section 3 . 3 . 1 . We used XPPaut to perform a bifurcation analysis of the Ras switch with changing levels of SOS , with and without a load . The quasi-potential landscape does not provide useful insights into load induced modulation of the Ras switch and hence the probability distributions are not reported .
The presence of a binding partner for either Repressor 1 or Repressor 2 ( which we refer to thereafter as the load ) introduces new terms in the differential equations describing the toggle switch , i . e . the last two terms in Eq . 1 and in Eq . 2 , as well as two new equations , Eq . 3 and 4 , in the dynamical system . However it can be easily seen that in steady state Eq . 3 and 4 are also independently set to zero , and therefore do not affect the bifurcation properties of the switch . Even in the case of a dynamic load , since Eq . S13 and S14 are set to zero to ensure the load-repressor complex is in steady state , the additional terms in Eq . S9 and S10 are also zero . Thus the load makes no difference to either the bistability of the switch or to the parameter values where the bistability is seen . The exception is when the repressor molecule can degrade even when bound to the load , which may be relevant in some experimental situations . As Fig . 2A shows , when a load is added symmetrically to both sides of the toggle switch , the two stable states approach each other and eventually annihilate , leaving a monostable system . Fig . 2B shows that when a load is added only to one side , the system again goes from bistable to monostable at some critical value of the load . In effect , the upper stable point vanishes and is no longer accessible due to leakage of the repressor affected by the load . The reason for the change in steady state behavior is made clear on examining the equations of the system . Here we need to incorporate additional reactions that represent the decay of the repressor-load complex into the load alone . This leads to an additional term in the equation for the load and the repressor-load complex ( Eq . S44 and S45 ) . However this term does not appear in the equation for the repressors , which continue to be governed by Eq . 1 and Eq . 2 . As a consequence in the steady state , the additional terms in Eq . 1 and 2 no longer equal zero and the steady state properties of the switch are influenced by the presence of the load . As can be seen from an examination of the chemical reaction system , this mechanism of abrogation of bistability arises whenever the load-repressor complex participates in a non-reversible ( from the repressor's point of view ) chemical process that leads to an unbalanced leakage of the repressor from its function as a repressor by the presence of the load . A more interesting example of such a process could be provided by a chemical reaction system where the load is an enzyme for one of the repressor molecules , which is transformed by the enzymatic action into a protein no longer capable of repression . The mathematical analysis of this case is exactly the same as the model we are currently discussing hence we do not consider it separately here . However a load can significantly change the dynamic response of the basic genetic toggle switch as we shall see below . We examined two different measures of dynamic response , response time for state switching and the amount of inducer required for state switching . We measured two response times , the rise time which quantifies the time taken for the concentration of Repressor 2 to increase from its low or zero level in state 1 to its high level in state 2 , and the decay time which measures the time taken for Repressor 1 to decay from its high level in state 1 to its low level in state 2 , in both cases in response to a constant inducer . Specifically the rise time measured the time to go from 10% to 90% of the steady state maximum , while the decay time measured the time to go from 90% to 10% of the steady state maximum . These measurements were made using the deterministic model in the cases when the load was applied only to one side and to both sides of the switch . We found that both the rise time and the decay time increase with increasing load concentration . Interestingly , this relationship was approximately linear in all cases ( Fig . 3A & B ) . The slope of the linear relationship represents the increase in response time due to unit increase in load . We found that the slope of the line was larger when the load was applied to the opposite side of the system before the switching rather than the same side ( Fig . 3A ) , indicating that it is harder to switch out of a state without a load to a state with a load than the reverse . However when a load was applied to both sides , the slope of the linear fit was higher than when the load was only on the opposite side , suggesting that both the “opposite side” and the “same side” delays are operating . While we also found an approximately linear relationship between the decay time and the concentration of the load , there was little difference between the decay times for the state with the load ( “same side load” ) and the state without a load ( “opposite side load” ) at our base parameter values . Thus the load affects rise time and decay time differently . When a load was applied to both sides of the switch , the slope of the decay time linear fit was larger , again indicating the operation of both delays . We tested these results by changing parameter values for the binding of the load ( Table S1 ) and found that in all cases we obtain a good linear fit for the response time . For the rise time , the slope was uniformly larger when the load was applied to the opposite side as compared to the same side , and it was the largest when loads were applied on both sides . For the decay time , the slope could be larger or smaller when the load was applied to the opposite side of the decaying state compared with the same side , but it was always larger than both when a load was applied on both sides . The slope depended non-monotonically upon the dissociation constant ( Kd ) of the binding between the repressor protein and the load , with both low Kd and high Kd having a smaller effect that those in between ( Fig . S1 ) . This was because when the Kd was low , i . e . strong binding , the concentration of the load-repressor complex was unaffected by the state of the switch . However when the Kd was high , the maximum concentration of the load-repressor complex was smaller , thereby having a lesser effect on the system ( Fig . S2 ) . Thus response times are maximized when the load acts as a dynamic sink , i . e . it takes up newly synthesized repressor when the state changes from the unloaded to the loaded side , and releases the bound repressor when switching from the loaded side to the unloaded side . Previous studies of response times of biochemical networks with and without a load have also seen monotonic increases in the response time of simple transcriptional circuits [37] . However the extremely consistent approximately linear response we see under wide variation in parameter values is extremely intriguing . An increase in response time should also imply that the concentration of inducer required to shift states should also be affected , especially when it can decay . In accordance with this expectation we also found that the concentration of inducer required to switch states increased exponentially with increasing load , as seen in Table S2 . The parameter of the exponential fit was dependent on the inducer decay rate , indicating that the amount of time the inducer remains above a threshold is the key factor governing the switching . We find that this response to a load is unaffected by the mode of switching the toggle , and induction by repression of the current state yields the same qualitative results ( Table S2 & S3 ) . In our analysis so far we have assumed that the total concentration of the load is fixed . We now analyze the case when the load is generated by a constitutively active promoter and can decay at a first order rate . We find that in this case too the qualitative features of the transition time remain the same as the toggle switch with a fixed load , i . e . it was approximately linear in all parameter regimes tested ( Supplementary Text S1 section 1 . 5 , Fig . S3 and Table S4 ) . The reason why we do not see a difference from the basic toggle switch is that the transition times ultimately measures time between steady states , and we wait for the system to come quite close to the steady state value ( 90% ) . Thus the concentration of the load has also reached a steady state value and the system behaves as it would with a fixed load . We also tested the response times when the repressor can leak away from the systems after binding with the load . Here we find that ( Fig . 4 ) when a load is applied to the same side , the rise time continues to increase monotonically linearly with the load but the decay times decreases monotonically with the load . However when a load is applied to both sides , we find a negative linear relation between the transition times for both rise and decay and the load . The reasons for the change in behavior is because as we saw previously , when the repressor can leak away from the repressor-load complex , a load has a dramatic effect on the bistability properties of the switch , abrogating bistability very quickly ( Fig . 1 ) . When only one repressor has a load , the high state of that repressor approaches the unstable state , indicating a decrease in the domain of attraction . Shifting out of that state thus becomes easier with increasing load . When both sides have loads , both stable states approach the unstable state , therefore shifting out of either state becomes easier , and both transition times decrease . The modulation in the dynamic properties of the basic genetic toggle switch discussed above suggests that the load has altered the potential energy landscape of the toggle switch , making it harder to switch . For two-dimensional and higher systems , such as the toggle switch , analytical methods to construct the potential landscape are not available , but a quasi-potential can be constructed from the probability distribution function of the concentrations of the repressor molecules , where the quasi-potential is given by the negative of the natural logarithm of the probability distribution [43] , [44] . To calculate this we performed Monte Carlo simulations of the toggle switch using a Gillespie type algorithm elaborated in the Methods section . When the toggle switch is symmetrically balanced , both the probability distribution function and the potential energy landscape are completely symmetric . If the system is started in State 1 , random fluctuations can drive it into State 2 and vice versa . The probability distribution can then be constructed by counting the frequencies of these random fluctuations . However since the genetic toggle switch can be very stable , a numerical computation of the potential energy landscape requires impractically long simulation times ( as we show below ) . While computational methods to sample rare trajectories in such cases exist , they are very sensitive to choices of parameters [42] , [45] . We developed a computational protocol in order to numerically obtain the probability distribution function of both protein concentrations and the transition times . We chose an appropriate volume for the genetic toggle switch such that exactly the same parameters as in the deterministic simulations led to the operation of the toggle switch with only a small number of proteins . The toggle remains bistable in this regime but the small protein numbers vastly increases spontaneous stochastic fluctuations arising out of multiplicative noise in the system and allows the simulation to explore parameter space and collect enough data . Our simulations showed that the switch switched states a large number of times . In order to account for differences in the time step in different states , the probability density function of the concentrations was constructed using a time trace collected after approximately 1 second intervals . As Fig . 5 shows , for a symmetric switch we obtain a symmetric bi-modal probability distribution that corresponds to a double-well potential . When we add a load to the system asymmetrically , in the form of a binding partner for the Repressor 1 , we find that the probability distribution becomes extremely skewed , and the total weight of the probability distribution corresponding to the other side , i . e . Repressor 2 , dramatically increases ( Fig . 6A ) . This indicates that the underlying double well potential has become skewed and the state 2 , corresponding to high Repressor 2 , has increased its stability at the cost of State 1 ( Fig . 6C ) . When a load is applied to both sides symmetrically , the concentration probability distribution reverts to a symmetric bimodal distribution corresponding to a symmetric double-well potential ( Fig . 6B & D ) . In order to test this directly we calculated the distribution of lifetimes in state 1 and the lifetimes in state 2 . As shown in Fig . 7 , when the switch is symmetric with no load , the lifetime distribution is exponential , as should be expected for a simple two-state system . However when the load is applied to Repressor 1 , the probability distribution of the lifetime in state 2 increases dramatically . The average lifetime of state 1 also increases but only by a very small amount . The time spent in state 2 does not appear to saturate , and continues to increase with increasing load . When loads are applied symmetrically to both sides , the lifetime histogram in Fig . 7 indicates that both sides have been stabilized since the system spends significantly longer time in each state . Note that in an equilibrium system this would have been indicated by the deepening of the potential well . However in non-equilibrium systems the potential well picture does not completely capture the dynamics and there is an additional contribution from a “curl flux” [43] , [46] that needs to be taken into account . For our purposes calculating both the distribution of concentrations and the distributions of lifetimes captures the dynamics of the toggle switch . To test whether our results change for higher protein concentrations , we increased protein concentrations about fivefold and recalculated the probability distribution function . We find that our qualitative results remain robust despite the increase in protein concentrations ( Supplementary Text S1 section 1 . 3 and Fig . S4 ) . Switching between states is rare at these protein numbers , with a mean residence time in state R1 for the unloaded switch being approximately min against about 700 min for the basal case considered , a difference of almost three orders of magnitude . However as for the basal case , the quasi-potential landscape skews significantly with the addition of a load on the switch . These results allow us to interpret the dynamic results that we obtained earlier . If the system is in state 2 and there is a load on state 1 , a transition requires an increase in Repressor 1 concentration in order to suppress the production of Repressor 2 . A load on Repressor 1 however competes with the promoter of Repressor 2 for binding with Repressor 1 , and thereby reduces the effective concentration of Repressor 1 . This effectively stabilizes state 2 . The dynamic analysis shows that state 1 not only remains an attractor state but in fact it takes a longer time , and more inducer , to shift out of state 1 as compared with the no-load situation . This is because the load also acts as a reservoir for Repressor 1 , and in fact increases its total concentration . This slows down the transition to state 2 . Interestingly this “same side effect” is generally weaker than the “opposite side effect” above . In agreement with this picture , the stochastic simulations show that the distributions of lifetimes in state 1 broaden slightly on addition of a load . If the load is present symmetrically on both sides , the concentration histograms in Fig . 6 and the time histograms in Fig . 7 indicate that both states have been stabilized , due to a combination of the ‘same side’ and the ‘opposite side’ effect now acting together to stabilize each state of the switch . In the dynamical simulations this is seen by the increased slope of the response time line for the case of a load on both sides . Results for additional parameter values are shown in Fig S15 and Fig S16 . When a positive feedback moiety is introduced in the toggle switch , we again see a linear relationship between the rise time and the decay time of the two states of the switch and the load ( Fig . S5 ) . Therefore here too the load appears to be skewing the underlying potential landscape of the switch . Using stochastic simulations we constructed the probability distribution function of this toggle switch as described above . We found that even in the absence of a load , when a positive feedback moiety is introduced on one side of a toggle switch , the probability distribution for the toggle switch , and hence the quasi-potential landscape , becomes extremely skewed in favor of the state with positive feedback as shown in Fig . 8A . Even with no load on the system , the switch is biased to State 1 and the lifetime spent in State 1 is much longer than in State 2 . If a load is added to R2 , the opposite side effect additionally favors State 1 . If a load is added to R1 however , the opposite side effect favors State 2 ( Fig . 8B ) . It is possible to balance these effects resulting in a more even distribution by adjusting the load on R1 and the strength of positive feedback . As the load on R1 is increased beyond this balance point , the opposite side effect dominates and the probability distribution becomes skewed toward State 2 ( Fig . 8C ) . As the opposite side effect increases with increasing load , the lifetime in State 2 also increases in agreement with the findings for the regular toggle switch ( Fig . 8D ) . The lifetime in State 1 also increases by a smaller amount , as for the regular toggle switch ( Fig . 8E ) . For the toggle switch with the positive feedback moiety , we can also check the consequences of allowing repressor leakage through the repressor-load complex . As shown in Fig . S13 , this addition to the system affects the steady state properties of the switch and bistability is abrogated after the load increases beyond a critical value , when load is present for both sides or only one side . The RasGTP system shows a bistable transition from a low RasGTP state to a high RasGTP state as the activating signal , in our case the number of SOS molecules , are varied . As Fig . 9 shows , a system with no Raf shows a classic Z-shaped bifurcation diagram with two bifurcations as SOS is varied . The first bifurcation marks the transition from a monostable low-RasGTP state to a bistable system with a “high” RasGTP state ( and an unstable intermediate state ) . The second bifurcation marks the transition from the bistable state to another monostable state with a high concentration of RasGTP . When Raf is added to the system , the bifurcation diagram changes and the two bifurcations start approaching each other . This is because the effect of adding Raf is equivalent to sequestering away some of the activated RasGTP in an “inactive” complex . When Raf concentration crosses a threshold , the bifurcations annihilate each other and disappear . This system is now characterized by a single stable point for all concentrations of SOS , and the disappearance of the threshold for Ras activation . While there appears very little free Ras , in reality , even for low SOS concentrations there is a large concentration of the activated RasGTP-Raf complex ( since RasGTP in these complexes is also protected from the action of the Ras GTPases ) . This can be seen in another way in Fig . S8 where the stable state of RasGTP is plotted against the level of total Raf in the system , keeping the level of SOS constant . Again we see that a bistable system is transformed into a monostable system when Raf increases beyond a threshold . These results are exactly the same for the model which assumes Michaelis-Menten kinetics , except for small changes in molecule numbers , as can be seen in Fig . S7 and S9 . Results do not change on changing load-binding parameters ( Fig . S10 , S11 ) Thus the addition of the Raf scaffold , which is an integral part of the MAPK cascade , fundamentally changes the qualitative behavior of the positive feedback switch . The main reason why the steady state bifurcation properties are affected here in contrast to the basic genetic toggle switch is that for this signaling circuit , as seen in Eq . 22–25 , total Raf and Ras are conserved , as is typical for a short timescale signal transduction system . These conservation laws couple Raf concentration to RasGTP concentration even at steady state . Therefore adding Raf to the system effectively reduces total Ras concentration since Raf sequesters away Ras from the switch . To see this more generally , consider for example a chemical reaction system comprising of n-species . Let us assume without loss of generality that the species is coupled to a downstream circuit through a binding reaction with a load , . The ( n+2 ) differential equations describing this system are: ( 26 ) ( 27 ) ( 28 ) ( 29 ) Note that for simplicity of notation we have not indicated the dependence of the dynamical system on its own parameter values . Now in the steady state , if the set of equations is complete , the left side uniformly goes to zero and we recover the result that the steady state remains exactly the same with or without a load , as for the genetic toggle switch . However let us now assume that we have an additional conservation law , say , ( 30 ) This conservation law implies that one equation in our dynamical system is redundant , and we need to drop one equation to make the system linearly independent . We can decide to drop Eq . 19 , and substitute in Eq . 20 and Eq . 21 and solve the resulting ( n+1 ) equations for the ( n+1 ) unknowns , , obtaining as a residual from Eq . 22 . Thus the steady state solutions of the now involve the amount of the load . Clearly , the existence of the conservation law has led to a change in the steady state properties of the dynamical system . Note that itself would usually enter ( by itself or in the form of other complexes , which then would also need to be accounted for in the conservation law Eq . 22 ) into one or more of the equations for the remaining species , . This would result in the equations for those other species explicitly involving , and thus depending upon the level of the load . For the Ras system above , Eq . 16 couples the load , Raf , to the concentration of Ras . However Ras concentration and SOS concentration are also coupled . Thus the load explicitly affects the steady state values of all species concentrations in this system . This leads to a fundamental qualitative change in the bifurcation properties of the system .
It has been pointed out previously that significant sequestration effects can abrogate zero order ultrasensitivity [26] , [47] , [48] , can change the dynamics of simple phosphorylation circuits [23] , [24] and change oscillatory behavior in some circuits [27] . We add to this body of work by demonstrating that the addition of a simple binding partner to the output protein of a genetic or signaling switch can have dramatic effects on its properties , and can fundamentally change the operation of the switch . For a genetic toggle switch with two mutually repressing proteins such as the classic switch built by Gardner et al . [5] we showed that even though the presence of the binding partner does not alter steady state properties of the switch , it can drastically change the dynamic properties . Using a novel potential landscape analysis , we showed that this is because the addition of the binding partner skews the underlying quasi-potential , making one state significantly more stable than the other . In practice therefore , a genetic toggle switch that is significantly skewed towards one side may never properly function as a switch . Thus the downstream consequences of such loads need to be taken into account when designing larger synthetic circuits with the toggle switch as one of the elements . On the other hand this phenomenon actually provides a way of making artificial switches tunable . It is possible to engineer a biased switch merely by adding a load on the opposite side of the toggle , which is a useful device when engineering a switch that is designed to be switched on only in special circumstances . A load on both repressor proteins similarly stabilizes both sides of the toggle switch . This could be useful when working with synthetic components with low concentrations in cells , especially those that display stochastic switching . A load on both repressor proteins can significantly increase the stability of such a toggle . In natural systems , mutually repressing toggle switches are often found with other complexities , such as a positive feedback motif on one side . The positive feedback motif by itself biases the toggle switch by stabilizing the side it is on at the expense of the other side . A load on the same side then stabilizes the opposite side , and can re-establish balance between the two quasi-potential wells . For engineering circuits in multi-cellular organisms , it is worth noting that that feedback between the load on a toggle switch and the strength of the positive feedback may ensure that the switch operates efficiently even in the presence of cell to cell variability in the load . How loads vary between cells and in multi-cellular organisms is an interesting question to explore in future work . The presence of the positive feedback provides a potential target for evolutionary fine-tuning of the switch . In the above analyses we use novel potential landscape methods that have proved useful and insightful in fields such as protein folding to discuss the fundamental properties of a dynamical system that shows not apparent changes in its stability properties . We demonstrate that these methods , though still relatively underdeveloped for use with non-equilibrium chemical reaction systems , hold promise for understanding the dynamics of such systems beyond what linear stability analysis can provide . However there are certain conditions when addition of a load changes the stability properties of the genetic toggle switch . One class of such effects happen when the repressor can leak away from the repressor-load complex , as can happen either when the repressor can decay or degrade when bound to the load , or when the load can modify the repressor and make it unable to repress . We show , employing standard bifurcation analysis , that additional loads in this system can abrogate the switch-like properties of the toggle switch entirely . In switches based on autocatalysis or positive feedback with an enzymatic deactivation , such as is often found in signaling systems , the effects of a load are equally dramatic . We show that in a simple model of Ras activation , adding a small concentration of Raf molecules changes the bifurcation diagram of the signaling circuit and can completely abrogate the bistability in the system . While we have chosen a specific example of Ras activation , our simplified model , with an autocatalytic forward reaction and an enzymatic backward reaction is a minimal model for a many positive feedback switches . The change in the bifurcation diagram arises from the conservation laws that couple the concentration of the load with the concentrations of the proteins in the upstream module . Given the sensitivity of non-linear dynamical systems to initial conditions , it should probably be expected that many , if not all , positive feedback based switches that operate at the short timescales of signal transduction , and therefore must possess these conservation laws , should exhibit this sensitivity to the effect of a load . Our results throw up an interesting puzzle for quantitative biologists . In many natural signal transduction systems such as the MAPK cascade , the concentration of the output of a bistable switch is quite comparable to the concentration of the load , thus significant changes in load concentrations could have dramatic effects on the behavior of the switch . However it has also been shown that there is a significant cell to cell variability in protein concentrations [49] . How do cells ensure that positive feedback based switches such as the Ras switch continue to operate robustly in the bistable regime ? Additional regulatory mechanisms involving feedback between the load and its partner protein may exist that confer robustness to the qualitative behavior of the biochemical switch . Arguably some of the bells and whistles of natural protein networks that are often disregarded when analyzing the network may in fact be performing this role . In other words , self-assembled switches have to be complex ! In this context it is worth mentioning that it has been persuasively argued [50] , [51] that some biological circuits maintain robustness of “fold-change’ behavior rather than absolute levels of protein concentration . It is possible that additional protein-protein interactions that couple concentrations of loads with output proteins may end up in performing this function . Another significant factor that needs consideration is the role of spatial segregation in producing feedback from the downstream module to the upstream one . In fact it has been shown experimentally that MAPK substrates sequester activated MAPK in the nucleus , and thus protect it from cytoplasmic phosphatases . Changing the concentration of one substrate therefore affects the concentration of activated MAPK [52] . Previous discussions of the effect of loads on the operation of circuits have suggested the use of insulators , that is circuit elements that insulate the upstream module from the downstream module [22] . The initial suggestions for building insulators in Ref . [22] involved incorporating signal amplification along with negative feedback in the upstream circuit . Another way of insulating the circuit is to ensure that the demand of the load for its cognate repressor is never significant compared to the total amount of repressor . For a genetic switch therefore , a possible insulating mechanism is if the link to the downstream circuit is through a promoter . For example , consider making an AND gate from an output of the toggle switch . This can be done by inserting a constitutively produced protein Y that binds to R1 such that the complex is a transcription factor for another protein , say Z . Thus there is an AND relationship between the two inputs , Y and R1 and the output Z . To offset the effect of load induced modulation of the dynamics of R1 , an additional step can be inserted such that R1 first binds to the promoter region of another gene that codes for protein X and activates its transcription , and it is the protein X , rather than R1 , that can bind to Y and activate production of Z . The advantage of adding this extra step is that the concentration of the promoter for X is very small compared to the concentration of R1 , and therefore load induced modulation of the upstream toggle can be kept at a minimum . Note however that this cannot be done without the additional cost of the time delay required for the transcription and translation of X . As can be seen , any additional step or series of steps that can amplify a weak signal can act as an insulator . Another standard example of an amplifying circuit is a phosphorylation cascade which is especially relevant when considering Ras activation since it directly leads to the MAPK phosphorylation cascade . Phosphorylation cascades are also very fast , and therefore do not face the additional time delays of an additional transcriptional step . From the point of view of synthetic circuit design , the insulating mechanism here could be constructed by designing a weak binding affinity of Ras ( or the synthetic protein that plays that role ) for Raf ( or the equivalent protein ) . The bound complex then catalyzes a phosphorylation cascade that ends by connecting to the downstream circuit . Note that this method of insulation does not have the same time delay costs as the additional transcription steps . However it does come with the metabolic costs of having to produce large amounts of proteins that are essentially serving no useful physiological purpose for the cell . This cost could be relevant in some synthetic biology applications , and certainly needs to be evaluated during circuit design . It has been shown in the context of phosphorylation cycles that insulation always carries a metabolic cost , and in general better insulation carries a greater metabolic cost [53] . The existence of the MAPK phosphorylation cascade however begs the question whether it serves the purpose of insulation of the upstream Ras circuit from the downstream circuit . While it is not possible to answer this intriguing question without further experiments , it does appear that the Ras-Raf complex is present is quite large numbers on activated cells . This would suggest that insulation is not the function for which the cascade may have evolved . Our own analysis of the genetic toggle switch with the positive feedback motif suggests that Nature may prefer more complicated forms of regulation that balance the different components of the circuit . However there is no reason why both methods cannot be utilized . To our mind this is a very exciting question that requires more attention from experimentalists and theorists alike . It should also be noted that due to non-specific binding of transcription factors with DNA as well as between proteins , every circuit in the cell , real or synthetic , operates in the presence of a load . Variability in the functioning of circuits that are seen when transferring synthetic circuits between species , or even in different cells , may be a result of not only differences in basic protein concentrations , but also of this undervalued but nevertheless tangible load . Based on this reasoning we predict that some of the host-dependent effects that complicate synthetic biology , i . e . a synthetic circuit that works in one organism not performing well in another , are in fact due to changes in the intrinsic load due to non-specific binding when changing hosts . Our analysis underscores the importance of incorporating loads when simulating models of switches in natural and synthetic systems . Mathematical analysis of switch-like motifs therefore would do well to at least include a load on their output proteins , in order to incorporate the possible effects of load induced modulation on the circuit .
|
Cells rely on complex networks of protein-protein interactions in order to carry out life functions . Scientists believe that these networks are organized in a modular fashion; that is they are made up of functionally distinct parts like an electronic circuit . Modularity implies that just as we put together electronic parts to make an amplifier that we can use in many different circuits , we can put together biochemical reactions to make an amplifier , or a switch or an oscillator , which perform the same function in different organisms . This assumption is important in synthetic biology , where we engineer and assemble synthetic genetic circuits in living organisms in a modular fashion . We show that for important modules like genetic and signaling switches , the assumption of modularity has a crucial limitation . We show that if one simply connects a biological switch to another downstream circuit , the presence of the connection changes the operation of the switch , which in some cases may stop behaving like a switch . Our work underscores the importance of taking into account these downstream connections and suggests that natural systems may be balancing some of these components in order to ensure that despite diversity , modules continue to work in different systems with fidelity .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"systems",
"biology",
"biochemistry",
"signal",
"transduction",
"biochemical",
"simulations",
"computer",
"and",
"information",
"sciences",
"cell",
"biology",
"theoretical",
"biology",
"network",
"analysis",
"synthetic",
"biology",
"biology",
"and",
"life",
"sciences",
"regulatory",
"networks",
"signaling",
"networks",
"computational",
"biology",
"molecular",
"cell",
"biology",
"cell",
"signaling"
] |
2014
|
Loads Bias Genetic and Signaling Switches in Synthetic and Natural Systems
|
During the entry process , the human immunodeficiency virus type 1 ( HIV-1 ) envelope glycoprotein ( Env ) trimer undergoes a sequence of conformational changes triggered by both CD4 and coreceptor engagement . Resolving the conformation of these transient entry intermediates has proven challenging . Here , we fine-mapped the antigenicity of entry intermediates induced by increasing CD4 engagement of cell surface–expressed Env . Escalating CD4 triggering led to the sequential adoption of different pre-fusion conformational states of the Env trimer , up to the pre-hairpin conformation , that we assessed for antibody epitope presentation . Maximal accessibility of the coreceptor binding site was detected below Env saturation by CD4 . Exposure of the fusion peptide and heptad repeat 1 ( HR1 ) required higher CD4 occupancy . Analyzing the diverse antigenic states of the Env trimer , we obtained key insights into the transitions in epitope accessibility of broadly neutralizing antibodies ( bnAbs ) . Several bnAbs preferentially bound CD4-triggered Env , indicating a potential capacity to neutralize both pre- and post-CD4 engagement , which needs to be explored . Assessing binding and neutralization activity of bnAbs , we confirm antibody dissociation rates as a driver of incomplete neutralization . Collectively , our findings highlight a need to resolve Env conformations that are neutralization-relevant to provide guidance for immunogen development .
Human immunodeficiency virus type 1 ( HIV-1 ) envelope glycoprotein ( Env ) trimers , composed of three gp41-gp120 heterodimers , initiate the virus entry process by binding to the primary receptor CD4 via a high-affinity binding site ( referred to as CD4bs ) on gp120 . CD4 engagement triggers conformational changes that facilitate the binding of a coreceptor [1] . Subsequently , a second wave of conformational changes occurs within the gp41 subunit that leads to the release of the hydrophobic fusion peptide ( FP ) and finally the fusion of the viral and target cell membranes [2 , 3] . Important discoveries in recent years have highlighted that native Env is conformationally flexible even in the absence of receptor triggers , shifting dynamically between a closed ground state and activated open states resembling those triggered by CD4 engagement [4 , 5] . CD4 triggering exposes neutralization-vulnerable epitopes shielded on the native trimer , including the third hypervariable ( V3 ) loop crown and the CD4-induced site ( CD4i ) [6–8] . Paradoxically , V3 crown and CD4i antibodies are abundant in HIV-1 infection but have only a weak or no detectable neutralization activity at all because of their limited capacity to access their epitopes on the closed Env trimer [9–12] . In contrast , the rare broadly neutralizing antibodies ( bnAbs ) that are elicited [13] overcome the shielding restriction and neutralize a wide spectrum of global HIV-1 strains [14] . Although the capacity to bind to the closed ( pre-CD4-bound ) Env is generally thought to be critical for HIV-1 neutralizing antibody ( nAb ) activity [4 , 15–17] , the precise modes of action differ between individual nAbs and include direct interference with CD4 or coreceptor engagement [18 , 19] , arrest of Env in the ground state [4 , 5 , 20] , or acceleration of trimer decay by capturing Env in an activated state [4 , 21–23] . Despite detailed structural information on multiple Env conformations [2 , 24–26] , certain aspects of the entry process have not yet been fully unraveled . Soluble Env trimers , the basis of much of our current knowledge of Env structure , do not fully represent the wild-type membrane-embedded Env trimer [27–32] . Structural information on the native CD4-bound Env is comparatively limited and stems from a small number of lower-resolution cryo–electron tomography ( cryo-ET ) reconstructions [33 , 34] . Theoretically , each of the three gp120 protomers of the Env trimer has the capacity to bind CD4 and the coreceptor . However , how many CD4 and coreceptor molecules need to interact with an Env trimer in order for gp41 to undergo the necessary conformational rearrangements remains unclear [35] . A requirement of fewer than three CD4 proteins per Env has been proposed based on functional assays with mixed Env trimers that contained one or two CD4 binding–deficient gp120 subunits [35–41] . Yet , only one structural study so far has provided information on an Env conformation with partial CD4 occupancy using highly stabilized soluble Env trimers [42] . Here , we developed a strategy to investigate the antigenicity of native , membrane-embedded HIV-1 Env upon triggering with soluble CD4 ( sCD4 ) . The “on-cell sCD4-triggering assay” we introduce allows fine antigenic mapping of the conformational dynamics of the full-length Env trimer and resolution of dynamic changes in the antigenic landscape upon receptor triggering . Assessing the capacity of bnAbs in recognizing differentially CD4-triggered Env intermediates , we obtained novel insights into the interdependencies of bnAb binding and neutralization .
Recent discoveries have highlighted the intrinsic flexibility of the unliganded HIV-1 Env trimer complex [4 , 5] . Here , we sought to derive detailed information on the shifts in epitope exposure in CD4-unbound and CD4-bound forms of native , membrane-embedded HIV-1 Env by employing an experimental procedure we refer to as on-cell sCD4 triggering assay . The assay assesses the capacity of Env-directed monoclonal antibodies ( mAbs ) to bind to Env transiently expressed on cells in the presence of increasing concentrations of sCD4 . Binding of mAbs ( S1 Table ) to the differentially CD4-triggered Env was assessed by flow cytometry . We used mAbs as indicators of specific conformational states and therefore sought to avoid conditions in which the mAbs themselves would influence Env conformation . To define the appropriate conditions , we titrated each individual mAb over a wide concentration range in the absence and presence of varying concentrations of sCD4 using cells expressing the Env of the Tier 1B subtype B strain BaL . 01 ( Fig 1 , S1 , S3 , S4 and S5 Figs ) . As exemplified by the V3 crown-specific mAb 1-79 ( Fig 1 and S1 Fig ) , the sCD4-enhanced exposure of gp120 V3 crown on BaL . 01 Env is readily detectable after 10 minutes and increases up to a mAb dose of approximately 10 μg/ml for all probed sCD4 concentrations ( Fig 1A ) . Whereas 1-79 concentrations higher than about 10 μg/ml resulted in increased staining of CD4-unbound and low-level sCD4-triggered Env , the signal in presence of medium and high sCD4 concentrations started to decline ( Fig 1A ) , likely reflecting the onset of gp120 shedding , the release of gp120 from Env trimers [21] . We next normalized all staining curves of 1-79 to their individual maxima ( 1 . 0 = highest value measured at any particular sCD4 concentration ) ( Fig 1B ) . At lower 1-79 mAb concentrations ( 0 . 006–0 . 152 μg/ml ) , the normalized curves converge and tightly overlap ( Fig 1B ) . We refer to this as the basal epitope exposure curve of a mAb . The 1-79 basal epitope exposure curve rises along with increasing sCD4 concentration , peaks at approximately 5 μM sCD4 , and decreases thereafter ( Fig 1B ) . The native Env is known to continuously sample different conformations [5] . Antibodies specific for a conformation can bind and stabilize it , which is referred to as “conformation capture . ” As the conformational states are often short-lived [5] , the antibody concentration is expected to be a limiting factor in recognition . Employing the basal epitope exposure curves rather than staining curves obtained at higher antibody concentrations is thus crucial if native and not antibody-enforced conformational states are to be investigated . Extending the triggering/staining step to 60 minutes ( Fig 1C and 1D ) revealed an increase in gp120 shedding , with maximum 1–79 staining of sCD4-triggered Envs decreasing by up to 50% ( Fig 1C ) . The increased staining of CD4-unbound and low-level sCD4-triggered Envs highlighted that elongated reaction times provide more opportunities for conformation capture ( Fig 1C ) . These time- and antibody concentration–dependent effects eventually amount to a flattening of the basal epitope exposure curve , obscuring its otherwise distinct peak ( Fig 1D ) . To gain a better insight into the time development of the basal epitope exposure curve , we performed the on-cell sCD4 triggering assay with 1–79 mAb and second hypervariable ( V2 ) loop apex mAb PGT145 at a low concentration ( 0 . 1 μg/ml ) and stopped the reaction at frequent intervals ( Fig 2 and S2 Fig ) . Although the staining intensity by both mAbs increased during the 20-minute observation time , they were influenced differently by the sCD4 concentration ( Fig 2A and 2C ) . The PGT145 staining curves converged to a singular shape after approximately 10 minutes ( Fig 2D ) . In contrast , the 1–79 staining curves quickly reached maximum staining around the 5 μM sCD4 peak while continuing to increase at low or no sCD4 throughout the experiment ( Fig 2B ) . Therefore , although one can limit the influence of conformation capture by minimizing the detection mAb concentration , the on-cell triggering assay is inherently dynamic , and the conformation capture will inevitably manifest itself in the varying mAb binding on-rates depending on the sCD4 concentration . Nevertheless , after a minimum incubation time of approximately 5 minutes , which is necessary to establish a clear basal epitope exposure curve maximum , the sCD4 concentration corresponding to the curve maximum remains stable even as the signal continues to increase ( Fig 2B and 2D ) . It is this property of the basal epitope exposure curve development that allows for the retrieval of information on the conformational preference of antibodies by comparing the positions of their respective basal epitope exposure curve maxima . Based on these analyses , we chose an optimal triggering/staining step of 20 minutes because we found no evidence for increased signal loss , and thus gp120 shedding , versus shorter incubation times . In addition , the higher signal provided by the longer , 20-minute incubation step allowed us to derive basal epitope exposure curves even for weakly staining mAbs and inhibitors ( S3 and S4 Figs ) . The basal epitope exposure curves retrieved in independent experiments showed an identical pattern ( S5 Fig ) , allowing a robust readout also from single experiments . We acquired basal epitope exposure curves of BaL . 01 Env for a panel of anti-Env mAbs directed against antigenic sites on Env that are exposed upon CD4 engagement ( Fig 3A , S3 and S4 Figs ) . This included mAbs directed to the V3 crown , the CD4i domain on gp120 , and the immunodominant cluster I of gp41 [7 , 8] . To probe the accessibility of the heptad repeat 1 ( HR1 ) binding groove , which depends on CD4 engagement and is linked with a fusion-competent activated Env state that allows HR1–heptad repeat 2 ( HR2 ) interaction [43 , 44] , we used the HR2 peptide Fc-fusion protein C34-IgG1 [45] . To monitor the saturation of the cell surface–expressed Env with sCD4 , we utilized the tetrameric CD4-IgG2 molecule [46] as a detector ( Fig 3A ) . As sCD4 and CD4-IgG2 compete for the same binding site , a decrease in CD4-IgG2 staining intensity should reflect the decrease in free CD4 binding sites . Comparison of the basal epitope exposure curves of the CD4-dependent antibodies revealed a close similarity for the CD4i mAb 17b and V3 crown mAb 1-79 . Epitope exposure curves of both antibodies peak at approximately 5 μM sCD4 , at which binding of CD4-IgG2 is already greatly diminished , indicating a high CD4bs occupancy by sCD4 ( Fig 3A ) . Considering that the coreceptor binding site has substantial overlap with CD4i epitopes and includes the base of the V3 loop [47–49] , the peak of the basal epitope exposure curves of 1-79 and 17b indicates the predominance of an Env conformation that is optimally primed for coreceptor interaction , to which we refer hereafter as the optimal CD4i-triggered state . Binding of cluster I mAb 4B3 and of C34-IgG1 is not at its maximum at the optimal CD4i-triggered state and increases further upon elevation of sCD4 concentrations ( Fig 3A ) . We next investigated the epitope exposure for a selection of bnAbs during sCD4 triggering ( Fig 3B , S3 and S4 Figs ) . In line with its known preference for intact closed trimer , the V2 apex bnAb PGT145 showed high binding of CD4-unbound BaL . 01 Env but no reactivity with the optimal CD4i-triggered Env . The gp120-gp41 subunit interface bnAb PGT151 and the V3 high-mannose patch bnAb PGT121 also recognized CD4-unbound Env preferentially but maintained substantial binding activity toward the optimal CD4i-triggered state . Whereas the basal epitope exposure curve of bnAb 2G12 ( known to interact exclusively with V3 glycans ) remained largely unaltered across the probed sCD4 triggering range , the two V3 high-mannose patch bnAbs—PGT128 and PGT135—showed a preference for the optimal CD4i-triggered state . Binding of the two gp41-directed bnAbs , 4E10 ( directed against the membrane-proximal external region [MPER] ) and VRC34 . 01 ( directed against the FP ) , was gradually enhanced by increased sCD4 triggering . Thus , similar to MPER mAbs [50 , 51] , VRC34 . 01 appears to preferentially bind the CD4-bound Env . However , the basal epitope exposure curves of the two bnAbs differ: 4E10 reached its epitope exposure plateau in an sCD4 range close to the optimal CD4i-triggered state , whereas VRC34 . 01 epitope exposure did not reach a plateau within the tested sCD4 doses . To assess the exposure of gp41 epitopes following CD4 engagement in detail , we probed the binding patterns of a broader panel of gp41-reactive mAbs comprising cluster I ( immunogenic loop ) , cluster II ( HR2 ) , and MPER and FP-proximal region ( FPPR ) mAbs in the on-cell sCD4 triggering assay ( S6 Fig ) . With the exception of MPER mAbs and the reported cluster II mAb 98–6 that showed overlapping patterns , all other gp41 mAbs tested displayed a reactivity pattern identical to the cluster I mAb 4B3 ( S6B and S6C Fig , Fig 3A ) . Though none of the tested gp41 mAbs preferred the CD4-unbound Env conformation , the MPER-targeting bnAbs bound it most efficiently ( S6C Fig ) . Next , we probed the effects of sCD4 triggering on virus strains with different levels of general neutralization sensitivity ( tiers 1 and 2/3 ) . We included the highly neutralization-sensitive Env MN . 3 ( subtype B , tier 1A ) [52] , the relatively neutralization-resistant tier 2 Env JR-FL ( subtype B ) [53] , and the highly neutralization-resistant Env clone BG505 . W6M . ENV . C2_T332N ( BG505_T332N ) ( subtype A , tier 2/3 ) [27 , 54] . Exposure of epitopes upon sCD4 triggering was assessed at a single antibody concentration ( 10 μg/ml ) that provided a satisfactory signal intensity across all Env/antibody combinations tested . We probed epitope accessibility during gradual sCD4 triggering for three groups of antibodies: ( 1 ) weakly neutralizing antibodies ( wnAbs ) and nonneutralizing antibodies ( nnAbs ) , ( 2 ) CD4bs-targeting mAbs , and ( 3 ) bnAbs ( Fig 4 , S7 and S8 Figs ) . wnAb/nnAb epitope accessibility in the CD4-unbound stage and the sensitivity to sCD4-triggered conformational changes reflect the known neutralization sensitivity/resistance , open/closed conformation of the four virus strains ( Fig 4 and S7A Fig ) . In particular , binding of wnAbs/nnAbs to MN . 3 is already substantial in the CD4-unbound state and further boosted at low sCD4 concentrations . Compared to MN . 3 , the basal exposure of wnAb/nnAb epitopes on BaL . 01 is lower and can be enhanced at relatively modest sCD4 concentrations . In contrast , the accessibility of wnAb/nnAb epitopes on JR-FL is low , and their full exposure requires high concentrations of sCD4 . Binding of all wnAbs/nnAbs to BG505_T332N is negligible in the native state , and the sCD4 concentrations necessary to induce their binding are so extreme that we were unable to reach saturation in our assay . Thus , sCD4 doses that suffice to unshield tier 1 and tier 2 Envs failed to fully unmask neutralization-sensitive epitopes on the more neutralization-resistant BG505 trimer . bnAbs showed similar reactivity patterns in response to sCD4 triggering for all four Envs ( Fig 4 and S7B Fig ) , demonstrating that the conformational preference of bnAbs is well conserved irrespective of the Env genotype and its general neutralization sensitivity/resistance . The sCD4 triggering assay revealed that the open conformation of MN . 3 extends even to gp41 epitopes , as both cluster I and MPER epitopes are already accessible on CD4-unbound Env , and their exposure does not increase upon sCD4 triggering ( Fig 4 and S7 Fig ) . Assessment of epitope exposure by a larger gp41 mAb panel and C34-IgG1 ( S9 Fig ) revealed that the MN . 3 trimer assumes a unique , minimally shielded conformation with fully accessible FPPR , cluster I , cluster II , and MPER but effectively concealed FP and HR1 ( S9A–S9C Fig ) . FP and HR1 exposure upon sCD4 triggering closely paralleled the induction of V3 crown and CD4i epitopes on MN . 3 ( S9D Fig ) . We next probed the relevance of binding to closed ( CD4-unbound ) Env conformation for antibody neutralization . A correlation between binding to native BaL . 01 Env ( mean of fluorescence intensity [MFI] with staining antibody at 10 μg/ml ) and neutralization activity ( area under the inhibition curve [AUC] ) against BaL . 01 in the TZM-bl pseudovirus assay signified a wide spectrum of nAbs ( Fig 5A ) . V2 apex bnAbs , 2G12 , and to a lesser extent PGT135 differed in this respect , showing comparatively weak neutralization activity despite high binding efficacy . These antibodies exhibited incomplete neutralization of the BaL . 01 virus , i . e . , an inhibition curve with a top plateau below 100% ( Fig 5B and 5C ) . We observed a similar interrelationship of V2 apex , 2G12 , and PGT135 antibody binding and neutralization for MN . 3 but not for the more neutralization-resistant viruses JR-FL and BG505_T332N ( S10 Fig ) . The overall positive correlation between binding ( MFI ) and neutralization activity ( AUC ) was evident for all four viruses probed irrespective of the open/closed conformation of their native Envs ( S10A Fig ) . In line with the low-level impact of sCD4 triggering on BG505_T332N Env , inhibitor binding in the presence of increasing doses of sCD4 still showed a high correlation with neutralizing activity against the respective virus ( Table 1 ) . Notably , this association was rapidly lost for JR-FL and BaL . 01 , which are more sensitive to sCD4-induced conformational changes ( Table 1 ) . The correlation for MN . 3 Env was comparable across the entire sCD4 range , highlighting the lack of effective conformational shielding ( Table 1 ) . A plausible explanation for the disparity in binding and neutralization observed for 2G12 and V2 apex bnAbs would be high dissociation rates of these antibodies [22 , 55 , 56] . Alternatively , differences in Env binding and neutralization capacity could also result from the respective assay conditions that may impact antibodies differentially . The on-cell sCD4 triggering assay records binding at room temperature after a relatively short incubation , which may favor the measurement of antibody on-rates . Measurement of neutralization activity involves prolonged incubation , including a preincubation of antibody and virus at 37 °C , creating conditions that can promote antibody dissociation . Prolonged incubation may also irreversibly inactivate trimers ( e . g . , by arresting in nonfavorable conformations ) or cause trimer dissociation accompanied by gp120 shedding [4 , 21–23] . If present , these effects can impact the readout in both binding and neutralization assays in a virus strain– , antibody- , time- , and temperature-dependent manner . We thus sought to create an assay that enables measurement of antibody dissociation in absence of gp120 shedding . We generated a mutant Env , BaL . 01 SOS , which contains a disulfide bond linking gp120-gp41 , thereby preventing gp120 dissociation [57] . BaL . 01 SOS can be triggered by sCD4 to the same extent as BaL . 01 ( S11 and S3 Figs ) . A time-dependent loss in gp120 antibody signal was observed for BaL . 01 wild type , which is in line with the ability of gp120 to dissociate from BaL . 01 but , as expected , not for BaL . 01 SOS ( S11 Fig and Fig 1 ) . As BaL . 01 SOS expression was higher , BaL . 01 SOS/BaL . 01 binding ratios were normalized for expression based on antibody 2G12 and the reactivity of both Envs with a large panel of antibodies compared ( S12 Fig ) . The cell surface–expressed wild-type BaL . 01 and BaL . 01 SOS proved antigenically very similar . We only observed localized differences in antibody binding to BaL . 01 SOS compared to BaL . 01 wild type that centered on CD4i and gp41 epitopes . CD4bs , V3 high-mannose patch , and V2 apex and subunit interface reactivity was comparable ( S12C Fig ) . Probing the dissociation of a panel of mAbs , CD4-IgG2 , and sCD4 after binding to BaL . 01 SOS Env–expressing cells at 37 °C ( Fig 6 and S13 Fig ) , we observed the highest dissociation rates for V2 apex mAbs , PGT135 , and 2G12 ( Fig 6A–6C ) . In contrast , the signal for MPER bnAbs decreased very slowly or even increased slightly over time ( Fig 6A and 6B ) . Antibody dissociation curves for wild-type BaL . 01 closely resembled the patterns observed for the SOS mutant with the expected exception of a rapid decline following sCD4 or CD4-IgG2 treatment known to induce gp120 shedding ( S14 Fig ) . The fact that we observed equivalent dissociation patterns of antibodies on both BaL . 01 wild type and BaL . 01 SOS confirms that the decrease in binding we observe for certain antibodies is not due to shedding but indeed reflects high dissociation rates . Neutralization of HIV-1 before CD4 attachment is considered critical , as access of antibodies to the CD4-engaged Env on cells may be more constrained , and rapid progression toward fusion likely limits possibilities of antibody interference [58] . However , antibodies that retain post-CD4-attachment activity may increase their window of opportunity and also potentially neutralize the virus better in the setting of cell–cell transmission [18 , 59] . It is thus of high interest to determine if and which bnAbs have a post-CD4-attachment activity . CD4 triggering has been shown to have a negative effect on the binding of V2 apex bnAbs PG9 and PG16 [60] , but neutralization capacity of these and other bnAbs in presence of CD4 has thus far not been systematically investigated . Probing neutralization of BaL . 01 by antibodies and other Env-directed entry inhibitors in the presence of increasing doses of sCD4 ( Fig 7A and S15 Fig ) , we observed diverse patterns of both positive and negative neutralization cooperativity ( Fig 7B and 7C , S16 Fig ) . To assess if bnAbs retain neutralization activity post CD4 triggering , we calculated the antibody-contributed inhibition in the coinhibition with sCD4 ( Fig 7B and S16 Fig ) . When an antibody exhibited no additional inhibitory effect at the highest antibody and sCD4 dose tested , we recorded this as loss of antibody-mediated neutralization ( Fig 7C ) . Antibodies with positive CD4 cooperativity gained in neutralizing activity . These included antibodies directed to the V3 crown , the CD4i and HR2 peptides , in line with previous findings [61–67] . CD4bs bnAbs showed no cooperativity with CD4 , consistent with their shared binding site with sCD4 . Cooperativity of sCD4 with MPER bnAbs was positive but small ( 4E10 ) or negligible ( 10E8 ) . The effect of CD4 triggering differed for V3 high-mannose patch bnAbs . PGT135 and 2G12 showed a positive cooperativity with a slightly improved neutralization capacity in presence of CD4 . In contrast , PGT121 and PGT128 displayed an intermediate negative cooperativity with CD4 . Both showed reduced capacity to neutralize post-CD4 engagement but retained considerable neutralization activity . bnAbs with strong negative CD4 cooperativity completely lost neutralization activity . The FP-directed bnAb VRC34 . 01 and all probed V2 apex bNAbs fell into this category . Collectively , this suggests that the high off-rate of V2 bnAbs in binding to the native Env and their incapacity to interact with CD4-triggered Env are reflected in the incomplete neutralization commonly observed for these bnAbs .
Definition of the various conformations the HIV-1 Env trimer can adopt is one of the main gaps in the knowledge of the HIV-1 entry process and its neutralization [68 , 69] . Most approaches toward this are technically complex and not easily scalable [31 , 33 , 34 , 41 , 42 , 70 , 71] . Here , we provide information on the antigenic landscape of multiple distinct receptor-triggered Env forms with varying CD4bs occupancy that is based on a comparatively simple binding assay setup . Our study builds on a tremendous body of work dedicated to unraveling the interactions of the Env trimer with CD4 and the conformational stages the trimer adopts upon CD4 engagement [43–45 , 72–74] . To be able to explore Env in a close to native setting , we assessed cell surface–expressed Env trimers by flow cytometry , a widely used setup [6–8 , 51 , 60 , 75–77] . A fine-tuned composition of the assay—which controls for a range of factors including temporal kinetics of CD4 triggering , allosteric effects of antibodies , and Env inactivation through gp120 shedding—allowed us to trace and systematically investigate the varying conformations of native and CD4-bound Env . Increasing the average occupancy of CD4bs on the trimers manipulates the equilibrium of Env conformational states . The successive CD4-triggered conformations that are generated at different sCD4 concentrations represent the majority species at the given assay condition . As both unliganded and CD4-bound states are metastable , our assay is best viewed as providing snapshots of a shifting equilibrium between conformational states . By exposing the Env to a wide range of sCD4 concentrations , we demonstrate a simple strategy to promote successive receptor-induced Env states to form , allowing their detailed antigenic characterization . What advances our setup over any described thus far is the establishment of individual basal epitope exposure curves for each antibody tested . With this , we minimize the allosteric effects of antibody binding on Env that might obscure the true conformational preference of antibodies . Our study shows how this limitation can be overcome and how a simple Env binding setup can be turned into a powerful antigenic characterization tool . Whereas our study focused solely on the effects of CD4 engagement on antibody epitope exposure , extending these analyses in future studies to evaluate the allosteric effects of antibody binding on the trimer and to explore potential multistep binding of antibodies to the trimer will be of high interest . In the present study , we sought to limit the allosteric effects of antibodies by carefully titrating antibody doses and restricting the readout to low doses . Differential epitope exposure at high and low antibody doses that we note in these dose-finding experiments ( S3 Fig ) is an indication of antibody-induced allosteric effects at higher antibody doses , highlighting potential avenues for further investigations . Our results complement recent investigations on the dynamic conformational rearrangements of the HIV-1 trimer [5 , 41] . The on-cell sCD4-triggering assay allows for fine antigenic mapping of the conformational dynamics of the full-length Env trimer and to resolve dynamic changes in the antigenic landscape upon receptor triggering . By assessing the response of Env trimers to sCD4 , triggering the assay further delivers a measure of trimer stability of the CD4-unbound state by disclosing how refractory a respective trimer to receptor-induced conformational changes is . This recommends our assay for screening of stable Env trimers considered for immunogen development . Binding efficacy of nAbs to native HIV-1 Env is a known predictor of in vitro neutralization potency [4 , 15 , 16 , 78] . Beyond confirming the strong link between neutralization and trimer binding [4 , 15 , 16 , 78] , our analyses provide the first survey , to our knowledge , of these relationships that covers the probed Env from the native state through the full range of CD4-triggered conformations . Our results further expand on prior findings by highlighting that exceptions exist . Certain antibodies can bind the unliganded Env with high affinity but fail to completely neutralize the corresponding virus as shown for the V2 apex bnAbs and 2G12 bnAb ( Fig 5 ) . A high rate of antibody dissociation can result in reversible neutralization as previously shown for 2G12 [22 , 56] , and a high off-rate of bnAb binding has been suggested to result in incomplete neutralization [55] . Here , we observe incomplete neutralization occurring predominantly for antibodies that display a high dissociation rate . Incomplete neutralization of HIV-1 is a commonly noted phenomenon in different assay systems , and other factors such as differential glycosylation and conformational heterogeneity of Env can affect its appearance [79–82] . In the present study , we show that the high off-rate in binding to the native Env and incapacity to recognize and neutralize CD4-triggered Env are two factors that may contribute strongly to the incomplete neutralization by V2 bnAbs . A key finding of our study is that we show several bnAbs to preferentially bind the CD4-triggered Env . This may indicate a capacity of these antibodies to neutralize both pre- and post-CD4 engagement , which needs to be explored . Inferring neutralization capacity post CD4 triggering directly from binding activity alone is , however , not straightforward . Accessibility of free virus and cell-bound virus may differ . Certain Env conformations that allow high epitope accessibility may be sampled beyond a relevant step in entry that can be blocked . In line with this , we observed an intriguing disparity for bnAb VRC34 . 01 , which showed enhanced binding upon CD4 triggering but decreased neutralization activity that will be interesting to tease apart in forthcoming studies . Overall , we found neutralization and binding activity to correlate well for native and partially triggered trimer but not for the fully opened trimer ( Table 1 ) . A strong correlation with native trimer fits with the notion that inhibition prior to receptor engagement is important [4 , 15 , 16 , 78] . Adding to this , our data suggest that neutralization in early stages of CD4 triggering may be possible for certain types of antibodies . Current assay systems predominantly record pre-CD4-attachment neutralization effects of antibodies [83] . Preincubation of antibodies and virions before addition to target cells favors the pre-CD4-attachment activity . Proof of post-CD4-attachment neutralization activity has long been established for MPER bnAbs but remained less studied for gp120 antibodies [22 , 59] . In light of our new findings , post-CD4-attachment activity should be systematically investigated to understand if our current screening systems capture this activity properly and to define its role in neutralization in vivo . Retaining neutralization activity beyond CD4 triggering may be key when antibody binding is reversible . Likewise , the capacity to neutralize post-CD4 engagement has been implicated in blocking cell–cell transmission of HIV-1 where engagement of CD4 is more rapid than in the setting of free virus infection [18 , 59] . Collectively , the findings we made using the on-cell sCD4 triggering assay contribute to a refined view of the process of HIV-1 entry and its inhibition . Our study highlights a continued need to resolve which Env conformations are neutralization-relevant . This will not only be interesting from a mechanistic point of view but will also provide guidance for immunogen development .
HEK 293T cells were obtained from the American Type Culture Collection and TZM-bl cells [84] through the NIH AIDS Reagent Program , Division of AIDS , NIAID , NIH . Cell lines were maintained in DMEM , high glucose , pyruvate ( Gibco , Thermo Fisher Scientific , Waltham , MA , USA ) supplemented with 10% heat-inactivated FBS ( Gibco , Thermo Fisher Scientific , Waltham , MA , USA ) , 100 U/ml penicillin , and 100 μg/ml streptomycin ( Gibco , Thermo Fisher Scientific , Waltham , MA , USA ) at 37 °C , 5% CO2 , and 80% relative humidity . The sources of inhibitors and plasmids used in this study are listed in S1 and S2 Tables , respectively . A codon-optimized sequence corresponding to 183 N-terminal amino acid residues of mature human CD4 followed by AviTag ( Avidity , Aurora , CO , USA ) , GSG linker , and 8xHis-tag was cloned into the pET-32a ( + ) ( Merck KGaA , Darmstadt , Germany ) expression vector so as to contain only one additional N-terminal methionine residue . The resulting plasmid was cotransformed together with the pBirAcm biotin ligase expression plasmid ( Avidity , Aurora , CO , USA ) into SHuffle T7 Express Competent Escherichia coli ( New England Biolabs , Ipswich , MA , USA ) . Bacteria were kept under antibiotic selection pressure in TYH medium ( 2% w/v tryptone , 1% w/v yeast extract , 1 . 1% w/v HEPES , 0 . 5% w/v NaCl , 0 . 1% [w/v] MgSO4 [pH = 7 . 3] ) to maintain both plasmids . Bacterial cultures , of which the optical density at 600 nm reached 0 . 7–0 . 8 , were induced with 50 μM Isopropyl β-D-1-thiogalactopyranoside in the presence of 50 μM D-biotin for 20 hours at 16 °C . Cells were mechanically disrupted in extraction buffer ( 50 mM phosphate , 250 mM NaCl , 20 mM imidazole , 20% v/v glycerol , 0 . 2% TWEEN-20 [pH = 7 . 4] ) . The recombinant protein in the soluble fraction was bound to Ni-NTA Superflow resin ( Qiagen , Venlo , the Netherlands ) . After washing with 60 resin volumes of wash buffer 1 ( 50 mM phosphate , 250 mM NaCl , 20 mM imidazole , 20% v/v glycerol [pH = 7 . 4] ) and 20 resin volumes of wash buffer 2 ( 50 mM phosphate , 20 mM imidazole , 10% v/v glycerol [pH = 7 . 4] ) , the protein was eluted with 8 resin volumes of elution buffer ( 50 mM phosphate , 250 mM imidazole , 10% v/v glycerol [pH = 7 . 4] ) . Finally , the protein was purified by size-exclusion chromatography in FPLC buffer ( 50 mM phosphate , 10% v/v glycerol [pH = 7 . 4] ) on a HiLoad 26/600 Superdex 200 column/Äktaprime plus FPLC system ( GE Healthcare , Uppsala , Sweden ) . A total of 1 . 25 × 105 HEK 293T cells per well were seeded in 1 ml of culture medium in 12-well tissue culture plates and incubated at 37 °C . Twenty-four hours later , cells in each well were transfected with a total of 1 μg DNA ( Env expression plasmid and pCMV-rev expression helper plasmid in 4:1 ratio ) mixed with 3 μg 25-kDa linear PEI or 40-kDa PEI MAX ( Polysciences , Warrington , PA , USA ) in 200 μl 150-mM NaCl . After settling the DNA-PEI complexes by a short spin ( 3 minutes , room temperature , 300g ) , the cells were incubated for 36 hours at 37 °C . All subsequent steps were carried out at room temperature . Cells were harvested , pooled , and distributed into 96-well round-bottom tissue culture plates for the individual staining reactions . For each staining reaction , cells were washed once with 200 μl staining buffer ( DPBS [Gibco , Thermo Fisher Scientific , Waltham , MA , USA] with 2% heat-inactivated FBS [Gibco , Thermo Fisher Scientific , Waltham , MA , USA] , and 2 mM EDTA ) and stained for 20 minutes ( unless indicated otherwise ) in 20 μl of staining buffer with 10 μg/ml of primary antibody ( unless indicated otherwise ) with or without the presence of sCD4 . After washing twice with 200 μl staining buffer , a secondary staining mix of 30 μl of staining buffer with 1:1 , 000 diluted APC-conjugated F ( ab' ) ₂ fragment goat anti-human IgG ( Jackson ImmunoResearch , West Grove , PA , USA ) or APC-conjugated streptavidin ( BioLegend , San Diego , CA , USA ) was added to the cells for 20 minutes . Following two washes with staining buffer , the cells were resuspended in 100 μl staining buffer with 0 . 1 μg/ml propidium iodide ( BD Biosciences , San Jose , CA , USA ) . Flow cytometry data were acquired on the FACSVerse system ( BD Biosciences , San Jose , CA , USA ) and analyzed using FlowJo 10 software ( FlowJo , Ashland , OR , USA ) . Arithmetic mean of APC fluorescence intensity ( MFI ) was calculated for the gated propidium iodide–negative single-cell population as a measure of primary antibody binding to the live cells in each staining reaction . For the experiments that required cell fixation , the protocol was conducted with the following modifications: A four times greater number of cells was used per staining reaction . Instead of propidium iodide in the final resuspension buffer , 1:1 , 000 Zombie Green Fixable Viability Dye ( BioLegend , San Diego , CA , USA ) was included in the primary staining mix of 30 μl total volume . After incubation with the primary staining mix , the cells were washed once with 200 μl DPBS , a further 200 μl of DPBS were added , and cells were put to 37 °C . At a specified time point , the DPBS was replaced with a fixing solution of 3% paraformaldehyde in DPBS . After 20-minute incubation at room temperature , the cells were washed twice with DPBS before adding the secondary staining mix . For postfixation cell staining , a simultaneous primary/secondary staining step was used with APC-conjugated F ( ab' ) ₂ fragment goat anti-human IgG ( Jackson ImmunoResearch , West Grove , PA , USA ) diluted 1:1 , 000 in staining buffer together with the primary mAb 2G12 at 10 μg/ml . In the on-cell sCD4 triggering assay , the MFI values for cells subjected to the same experimental conditions other than a different concentration of sCD4 were regarded as one data series , the MFI staining curve . To obtain a relative MFI staining curve , each of the MFI values within an MFI staining curve was divided by the maximum MFI value present within the same MFI staining curve . In the inhibitor dissociation assay , the relative MFI values for each inhibitor/control were calculated by dividing the MFI values for each time point by the MFI value for the time point t = 0 . In the on-cell sCD4 triggering assay , the basal epitope exposure curve was selected from the relative MFI staining curves of each individual antibody/inhibitor based on the following criteria: lowest concentration of the tested antibody/inhibitor that yields at least a 10-fold higher MFI signal over MuLV background at the curve maximum . We used CD4-IgG2 as an indicator for CD4bs saturation to set an upper bound on the CD4 occupancy in the assay . CD4-IgG2 has a higher affinity compared to sCD4 because of the capacity of multivalent binding . It further allows detection via the Fc . Employing CD4-IgG2 as a reference point allowed us to define if and which CD4-triggered Env stages occur below saturating conditions . A total of 2 . 25 × 106 HEK 293T cells were seeded in 20 ml of culture medium in a T75 flask and incubated at 37 °C . Twenty-four hours later , cells were transfected with a total of 20 μg DNA ( Env expression plasmid and pNL-lucAM HIV-1 backbone plasmid in 1:3 ratio ) mixed with 60 μg 25-kDa linear PEI ( Polysciences , Warrington , PA , USA ) in 4 ml 150-mM NaCl . At 6–18 hours post transfection , the culture medium was exchanged . Culture supernatant was harvested 48 hours post transfection , vacuum filtered through a 0 . 22-μm-pore-size membrane , and frozen as pseudovirus stock . The TZM-bl based pseudovirus neutralization assay was conducted essentially as previously described [85] . TZM-bl cells ( 1 × 104 ) in 100 μl of culture medium containing 20 μg/ml DEAE-Dextran ( Amersham Biosciences , Uppsala , Sweden ) were seeded in each well of white , 96-well , clear flat-bottom tissue culture plates ( Greiner Bio-One , Kremsmünster , Austria ) and incubated at 37 °C . Twenty-four hours later , pseudovirus was preincubated with serially diluted Env-directed inhibitors in culture medium for 1 hour at 37 °C or alternatively 20 minutes at room temperature for sCD4 coinhibition experiments . The pseudovirus-inhibitor mixture ( 100 μl ) was added to the TZM-bl cells . Luciferase reporter gene expression was assessed 48 hours post infection with Bright-Glo Luciferase Assay System ( Promega , Fitchburg , WI , USA ) on the Dynex MLX luminometer ( Dynex Technologies , Chantilly , VA , USA ) . Virus input was chosen to yield virus infectivity corresponding to 10 , 000–40 , 000 relative light units ( RLU ) on the medium sensitivity setting in the absence of inhibitors . Prism 7 software ( GraphPad Software , La Jolla , CA , USA ) was used to fit 4-parameter logistic curves to the data and to calculate the area under each neutralization curve ( AUC ) . To quantify the neutralization cooperativity of a nAb/inhibitor with sCD4 , the percent effect of a nAb in the presence of sCD4 was calculated relative to the residual infectivity measured with sCD4 at the respective concentration in the absence of nAb/inhibitor . The difference between the area under the percent effect curve of a nAb/inhibitor in the presence of 2 . 5 , 5 , 10 , 20 , or 40 μM sCD4 and the area under its percent effect curve in sCD4 absence ( i . e . , its neutralization curve ) was calculated and designated as ΔAUC . The neutralization cooperativity of a particular nAb/inhibitor was then calculated as the sum of the five ΔAUC values . Pearson’s r coefficients and respective P values were calculated using Prism 7 software ( GraphPad Software , La Jolla , CA , USA ) . The structure figure was prepared with PyMOL Molecular Graphics System 1 . 7 software ( Schrödinger , New York , NY , USA ) using the referenced protein structure database file .
|
The trimeric human immunodeficiency virus type 1 ( HIV-1 ) envelope glycoprotein ( Env ) mediates HIV-1 entry into its target cells . Entry is initiated by sequential triggering of Env upon interaction with its primary receptor CD4 and a coreceptor on target cells . The ensuing structural rearrangements of the Env trimer bring the viral membrane in close vicinity of the cellular membrane , enabling fusion . Resolving the structural differences between the consecutive conformations Env adopts during the entry process is of high interest , as different antigenic domains are exposed , which may affect the capacity of neutralizing antibodies to bind to Env and inhibit entry . Here , we compared the conformation of unliganded closed Env with the transitional CD4-bound Env forms by studying the antigenicity of cell surface–expressed Env with and without CD4 triggering . We show that incremental triggering by soluble CD4 allows the capture of the full continuum of conformational changes , including events that follow coreceptor interaction . Thus , the setup we introduce here turns a simple binding assay into a powerful tool to study transitional conformation changes in HIV-1 Env . Analyzing the capacity of Env-reactive antibodies to recognize the diverse Env stages , our study reveals novel aspects of the binding preferences of neutralizing antibodies that affect their inhibitory activity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] |
[
"cell",
"binding",
"flow",
"cytometry",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"rna",
"viruses",
"glycoproteins",
"antibodies",
"chemical",
"dissociation",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"immune",
"system",
"proteins",
"spectrum",
"analysis",
"techniques",
"proteins",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"chemistry",
"hiv-1",
"spectrophotometry",
"biochemistry",
"cytophotometry",
"cell",
"staining",
"cell",
"biology",
"virus",
"glycoproteins",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"lentivirus",
"glycobiology",
"organisms"
] |
2019
|
CD4 occupancy triggers sequential pre-fusion conformational states of the HIV-1 envelope trimer with relevance for broadly neutralizing antibody activity
|
Active DNA demethylation in plants occurs through base excision repair , beginning with removal of methylated cytosine by the ROS1/DME subfamily of 5-methylcytosine DNA glycosylases . Active DNA demethylation in animals requires the DNA glycosylase TDG or MBD4 , which functions after oxidation or deamination of 5-methylcytosine , respectively . However , little is known about the steps following DNA glycosylase action in the active DNA demethylation pathways in plants and animals . We show here that the Arabidopsis APE1L protein has apurinic/apyrimidinic endonuclease activities and functions downstream of ROS1 and DME . APE1L and ROS1 interact in vitro and co-localize in vivo . Whole genome bisulfite sequencing of ape1l mutant plants revealed widespread alterations in DNA methylation . We show that the ape1l/zdp double mutant displays embryonic lethality . Notably , the ape1l+/−zdp−/− mutant shows a maternal-effect lethality phenotype . APE1L and the DNA phosphatase ZDP are required for FWA and MEA gene imprinting in the endosperm and are important for seed development . Thus , APE1L is a new component of the active DNA demethylation pathway and , together with ZDP , regulates gene imprinting in Arabidopsis .
DNA methylation is a stable epigenetic mark that regulates numerous aspects of the genome , including transposon silencing and gene expression [1]–[7] . In plants , DNA methylation can occur within CG , CHG , and CHH motifs ( H represents A , T , or C ) . Genome-wide mapping of DNA methylation in Arabidopsis has revealed that methylation in gene bodies is predominantly at CG context whereas methylation in transposon- and other repeat-enriched heterochromatin regions can be within all three motifs [8] . Although the function of abundant CG methylation within genic regions remains unclear , DNA methylation generally correlates with histone modifications that repress transcription activities [1] , [9] , [10] . DNA methylation patterns are coordinately controlled by methylation and demethylation reactions . In Arabidopsis , symmetric CG and CHG methylation can be maintained by DNA METHYLTRANSFERASE 1 ( MET1 ) and CHROMOMETHYLASE 3 ( CMT3 ) , respectively , during DNA replication . In contrast , asymmetric CHH methylation cannot be maintained and is established de novo by DOMAINS REARRANGED METHYLASE 2 ( DRM2 ) , which can be targeted to specific sequences by the RNA-directed DNA methylation ( RdDM ) pathway [1] , [10] , [11] . DNA methylation is antagonized by an active DNA demethylation pathway that includes the DNA glycosylases REPRESSOR OF SILENCING1 ( ROS1 ) , DEMETER ( DME ) , DEMETER-LIKE2 ( DML2 ) and DEMETER-LIKE3 ( DML3 ) [12]–[14] . ROS1 , DME , DML2 and DML3 are all bifunctional DNA glycosylases that initiate active DNA demethylation by removing the 5-methylcytosine ( 5-meC ) base and subsequently cleaving the phosphodiester backbone by either β- or β , δ-elimination [12] , [14]–[16] . When β , δ-elimination occurs , a gap with a 3′-phosphate group is generated . Our previous work demonstrated that the 3′ DNA phosphatase ZDP catalyzes the conversion of 3′-phosphate group to a 3′-hydroxyl ( 3′-OH ) , enabling DNA polymerase and ligase activities to fill in the gap [17] . The β-elimination product , a gap with a blocking 3′-phosphor-α , β-unsaturated aldehyde ( 3′-PUA ) , also must be converted to a 3′-OH to allow completion of the demethylation process through single-nucleotide insertion or long patch DNA synthesis by DNA polymerase and ligase [18] . However , the enzymes that may function downstream of ROS1 and DME in the β-elimination pathway have not been identified . The mutation of ROS1 leads to hypermethylation and transcriptional silencing of a luciferase reporter gene driven by the RD29A promoter , as well as of the endogenous RD29A gene [13] . ROS1 dysfunction also causes DNA hypermethylation in thousands of endogenous genomic regions [19] . zdp mutants also show hypermethylation in the RD29A promoter and many endogenous loci . However , the hypermethylation in the RD29A promoter caused by zdp mutations is not as high as that caused by ros1 mutations , and there are many ROS1 targets that are not hypermethylated in zdp mutants [17] . These observations suggest that there may be an alternative , ZDP-independent branch of the DNA demethylation pathway downstream of ROS1 and other DNA glycosylases/lyases . Although ROS1 functions in almost all plant tissues [13] , DME is preferentially expressed in the central cell of the female gametophyte and is important for the regulation of gene imprinting in the endosperm [20]–[22] . In Arabidopsis , the imprinted protein-coding genes include FWA ( Flowering Wageningen ) , MEA ( MEDEA ) and FIS2 ( Fertilization-Independent Seed 2 ) and the list is expanding [21]–[25] . The loss-of-function mutation of DME results in aberrant endosperm and embryo development because of DNA hypermethylation and down-regulation of the maternal alleles of imprinted genes [26] . DME is also necessary for DNA demethylation in the companion cells in the male gametophyte [27]–[29] . SSRP1 , a chromatin remodeling protein , was identified as another factor required for gene imprinting and the mutation of SSRP1 gives rise to a maternal lethality phenotype similar to that caused by DME mutations [30] . Therefore , it is possible that ZDP and other protein ( s ) acting downstream of the 5-meC DNA glycosylases/lyases may also affect gene imprinting in Arabidopsis . Intriguingly though , neither ZDP mutants nor mutants in other DNA repair enzymes that may be downstream of DNA glycosylases/lyases show developmental phenotypes associated with defective gene imprinting . In this study , we characterized the functions of Arabidopsis APE-like proteins in the processing of 3′-blocking ends generated by ROS1 and examined methylome changes induced by ape mutations . We found that purified APE1L can process 3′-PUA termini to generate 3′-OH ends . APE1L also displays a weak activity in converting 3′-phosphate termini to 3′-OH ends . ape1l-1 mutants show altered methylation patterns in thousands of genomic regions . Interestingly , we found that the ape1l+/−zdp−/− mutant is maternally lethal , giving rise to a seed abortion phenotype resembling that of dme mutants . The maternal alleles of the imprinted genes FWA and MEA are hypermethylated , and their expression levels are reduced in the endosperm of such abnormal seeds of the double mutant . Thus , APE1L functions downstream of the ROS1/DME subfamily of DNA glycosylases/lyases in active DNA demethylation and genomic imprinting in Arabidopsis .
The Arabidopsis genome encodes three AP endonuclease-like proteins: APE1L , APE2 and ARP [31] . We purified recombinant full-length APE1L , APE2 and ARP proteins , and found that all three enzymes exhibit AP endonuclease activity in vitro . APE1L , but not APE2 or ARP , also displayed a 3′-phosphatase activity ( Fig . 1A ) . We next wanted to determine if these proteins can process the 3′-PUA termini generated by ROS1 after the β-elimination reaction . We first incubated ROS1 with a 51-mer duplex DNA substrate containing a 5-meC residue at position 29 in the 5′-end labeled strand ( Fig . 1B ) . As expected , the DNA glycosylase/lyase activity of ROS1 generated a mixture of β- and β , δ -elimination products , with either 3′-PUA or 3′-phosphate ends , respectively ( Fig . 1B , lane 1 ) . These products were then purified and combined with either APE1L , APE2 or ARP proteins . We found that APE1L efficiently processed the 3′-PUA to generate a 3′-OH terminus . In comparison , 10-fold higher amounts of APE2 or ARP proteins displayed either weak [32] or undetectable ( APE2 ) activity against 3′-PUA ends ( Fig . 1B ) . To confirm that APE1L is responsible for the detected enzymatic activity we generated an APE1L mutant , N224D . Residue N224 corresponds to N212 of human APE1 , which is essential for the enzymatic activity of the mammalian protein [33] . Substitution of N224 by aspartic acid almost completely abolished the activity of APE1L on the 3′-PUA termini ( Fig . 1C ) . The mutation also greatly reduced the AP endonuclease activity on a synthetic AP site and the 3′ phosphatase activity on 3′-phosphate ends . Altogether these results indicate that , in addition to its AP endonuclease activity , APE1L possesses a potent 3′-phosphodiesterase activity that can efficiently process the 3′-PUA blocking ends generated by ROS1 . ROS1 remains bound to its reaction products , which contributes at least partially to the highly distributive behavior of the enzyme in vitro [34] . To determine whether APE1L is able to process 3′-PUA and/or 3′-phosphate termini in the presence of ROS1 , we incubated ROS1 and a duplex DNA substrate containing a single 5-meC residue , with WT or N224D APE1L ( Fig . 2A ) . We found that a 3′-OH terminus is efficiently generated in the presence of WT but not mutant APE1L ( Fig . 2A ) . The emergence of the 3′-OH terminus is concomitant with the loss of both 3′-PUA and 3′-phosphate ends , suggesting that the 3′-OH terminus is produced by the 3′-phosphodiesterase activity of APE1L on the 3′-blocking ends generated by ROS1 . Quantification of the reaction products revealed that the total amount of strand incision is not increased in the presence of APE1L ( Fig . 2B ) . To assess whether APE1L modulates the DNA glycosylase/lyase activity of ROS1 , we performed the reaction in the absence of Mg2+ , which is required for APE1L but not ROS1 activity . We found that the enzymatic activity of ROS1 is not increased in the presence of APE1L ( S1 Fig . ) . Thus , APE1L is able to access the 3′-blocked termini generated by ROS1 but does not increase the turnover of this DNA glycosylase . These results suggest that APE1L does not displace ROS1 from DNA . We next used in vitro pull-down assays to test whether ROS1 and APE1L can physically interact ( Fig . 3A ) . His-tagged ROS1 ( His-ROS1 ) was incubated with either Maltose Binding Protein ( MBP ) or MBP-APE1L bound to an amylose column . We found that MBP-APE1L , but not MBP , associates with His-ROS1 , suggesting that APE1L and ROS1 directly interact in vitro . To gain insights into the transfer of DNA demethylation intermediates between ROS1 and APE1L , we performed electrophoretic mobility shift assays with a gapped DNA substrate ( Fig . 3B–C ) . MBP-APE1L alone is not able to form a stable complex with the substrate , judging by the smeared band next to the position of the free probe ( Fig . 3B , lanes 2 and 1 ) . A mobility shift was observed when the DNA substrate was incubated with His-ROS1 , consistent with complex formation ( Fig . 3B , lane 3 ) . As we have previously reported [17] , part of the labeled probe remained trapped in the wells , hinting at the formation of insoluble His-ROS1-DNA complexes . Next , we incubated the gapped DNA substrate and His-ROS1 with increasing concentrations of MBP-APE1L to assess complex formation . With increasing MBP-APE1L , the band corresponding to the ROS1-DNA complex and the labeled material in the well gradually disappeared , concomitant with the appearance of a discrete , new band ( Fig . 3B , lanes 4–7 ) . Importantly , this band was only detected when both ROS1 and APE1L were present in the binding reaction . These results suggest that ROS1 , APE1L , and gapped DNA form a ternary complex and that ROS1 is required for APE1L to stably associate with the DNA substrate . To further examine complex formation we performed supershift experiments using antibodies against MBP-APE1L and His-ROS1 ( Fig . 3C ) . We found that adding anti-MBP to a binding reaction containing MBP-APE1L , His-ROS1 and DNA generated an additional shift , thus confirming the presence of APE1L in the complex . However , a supershift was not observed in the presence of the anti-His antibody . We reasoned that access to the His epitope on His-ROS1 might be restricted in the complex . Therefore , as an alternative approach , we compared the mobility shifts generated from binding reactions containing the DNA gapped substrate , MBP-APE1L and either His-ROS1 or MPB-ROS1 ( Fig . 3D ) . We found that MBP-ROS1 , which has a higher molecular weight than His-ROS1 , gave rise to a higher molecular weight gel shift , thus confirming that ROS1 is also present in the complex . The most likely interpretation for these results is that ROS1 , APE1L , and the gapped DNA substrate form a ternary complex . To further confirm the interaction between APE1L and ROS1 , we performed a firefly luciferase complementation imaging assay [35] in tobacco leaves . We found that APE1L can interact with ROS1 in the tobacco leaves ( Fig . 3E ) . Our previous data show that ZDP , a component of the active DNA demethylation pathway , co-localizes with ROS1 in subnuclear foci [17] . To determine the subnuclear localization of APE1L protein , we generated antibodies specific to APE1L and used them for immunolocalization of APE1L in Arabidopsis leaf nuclei . As shown in Fig . 4A , APE1L is broadly distributed throughout the nucleus . In 62% of the cells examined , APE1L is enriched in the nucleolus whereas in 38% of the cells , APE1L localizes to small nucleoplasmic foci . Only very weak signals were observed when the antibodies were applied to nuclei preparations of ape1l-1 mutant plants , indicating that the staining patterns in wild type plants reflect APE1L localization rather than non-specific binding of the antibody ( Fig . 4A ) . To test whether APE1L co-localizes with ROS1 or ZDP , we performed co-immunofluorescence . In our experiments , FLAG-tagged ROS1 was expressed from its native promoter in ros1-1 mutants and visualized with anti-FLAG antibodies . We observed APE1L co-localization with ROS1 within both nucleoplasmic foci and also the nucleolus in about 10% of cells , as shown by the strong yellow signals ( Fig . 4B ) . In 54% of the cells , APE1L co-localizes with ROS1 in the nucleolus but not in nucleoplasmic foci , whereas in 36% of the cells , APE1L and ROS1 do not substantially co-localize ( Fig . 4B ) . APE1L and ZDP also co-localize in nucleoplasmic foci in approximately 28% of cells ( Fig . 4C ) . Thus , APE1L co-localizes with components of the DNA demethylase machinery in distinct subnuclear structures in a subset of cells . To evaluate possible roles of APE1L in active DNA demethylation initiated by the ROS1 subfamily of DNA glycosylases/lyases , two T-DNA insertion lines were isolated for APE1L ( S2A Fig . ) . RT-PCR analysis with APE1L-specific primers corresponding to the full-length open reading frame of the gene detected the expected product in wild-type plants in both the Ws and Col backgrounds , but not in ape1l-1 , which is in Ws . In contrast , ape1l-2 shows almost the same expression level as wild-type plants ( S2B Fig . ) . Since the endonuclease ARP shows weak activity against the 3′-PUA blocking ends generated by ROS1 in vitro , we also isolated two T-DNA insertion lines for ARP ( S2A Fig . ) and confirmed by RT-PCR that they have a complete loss of mRNA expression ( S2B Fig . ) . One of the mutants , arp-1 , was used for further experiments . To examine the general DNA methylation status in the ape1l-1 and arp-1 mutants , we compared the susceptibility of 5S rDNA and 180-bp centromeric repeat regions to the restriction enzymes HpaII and MspI . These enzymes recognize the same site ( CCGG ) , but HpaII cleavage is methylation-inhibited whereas methylation does not affect cleavage by MspI . DNA cleavage was assessed by Southern analysis . Similar to the zdp-1 and ros1-4 mutations , the ape1l-1 or arp-1 mutation does not affect the DNA methylation levels at the 5S rDNA or 180-bp centromeric repeats ( S3 Fig . ) , suggesting that the ape1l-1 and arp-1 mutants do not have changes in their overall DNA methylation patterns . We performed whole genome bisulfite sequencing using DNA from 14-day-old ape1l-1 , arp-1 , zdp-1 and their corresponding wild-type control plants . The CG methylation levels in wild type ( Col-0 ) and zdp-1 mutant are similar , but the CHG and CHH levels are mildly elevated in zdp-1 ( S4A Fig . ) . For ape1l-1 , its overall genome methylation level in CG , CHG and CHH contexts is slightly higher than that in Ws ( S4B Fig . ) . In total , we identified 6389 DMRs ( differentially methylated regions ) in ape1l-1 mutant plants , including 3497 hyper-DMRs that have a significant increase in methylation and 2892 hypo-DMRs that have a significant reduction in methylation ( S4C Fig . ; S3 Table ) . In contrast , arp-1 only affects methylation levels at 403 genomic regions , including 162 hyper-DMRs and 241 hypo-DMRs ( S4C Fig . ) . 1559 hyper-DMRs and 612 hypo-DMRs were identified from zdp-1 ( S4C Fig . ; S4 Table ) . The hyper-DMRs and hypo-DMRs identified in ape1l-1 , zdp-1 and ros1-4 are evenly distributed along the five chromosomes ( S4D Fig . ) . To determine whether APE1L and ZDP mutations affected DNA demethylation in specific genomic regions , we analyzed intergenic regions , transposable elements ( TEs ) outside of genes , TEs overlapping with genes and genic regions . Unlike zdp-1 , ros-1 mutants and ros1-3;dml2-1;dml3-1 ( rdd ) triple mutants , which have less than 43% of hypermethylated ( hyper- ) DMRs distributed in gene regions , in ape1l-1 and arp-1 more than 60% of the hyper-DMRs are distributed in gene regions ( S4E Fig . ) . In contrast , the percentages of hyper-DMRs distributed in TEs in ape1l-1 and arp-1 are lower than those in zdp-1 , ros-1 and rdd mutants ( S4E Fig . ) . These data indicate that the APE1L and ARP mutations preferentially impact DNA demethylation of gene regions while the ZDP and ROS1 mutations have a greater impact on TE regions . The distribution patterns of classified hypo-DMRs are different from those of hyper-DMRs . The percentages of hypo-DMRs in gene regions are higher than 70% in rdd , zdp-1 and ape1l-1 . The arp-1 has a low percentage of hypo-DMRs in gene regions but high percentage of hypo-DMRs in intergenic regions ( S4E Fig . ) . The ape1l-1 mutation affects CHG and CHH demethylation more profoundly than CG demethylation , both in gene regions or in TEs ( S5A Fig . ) . We also examined the effect of APE1L mutation on TEs of different lengths and found that the ape1l-1 mutation has a bigger impact on shorter genes but longer TEs ( S5B and S5C Figs . ) . Unlike ape1l-1 , zdp-1 shows almost the same DNA methylation pattern for both gene regions and TEs ( S5 Fig . ) . Compared to the high level of overlap ( 70 . 9% ) between zdp-1 and rdd hyper-DMRs , less than 50% of the hyper-DMRs in ape1l overlap with those in rdd ( S4C and S5D–S5E Figs . ) . One reason for this relatively low level of overlap may be the difference in genetic backgrounds; the ape1l-1 mutant is in the Ws background whereas the other mutants are in Col . When the hyper-DMRs in ros1-1 ( C24 background ) and ros1-4 ( Col-0 background ) were compared , the overlap was also quite low ( 52% ) . For the hyper-DMRs , the level of overlap between ape1l-1 and zdp-1 is also very low ( 14% ) ( S5F Fig . ) even though some loci do show hypermethylation in ape1l-1 as well as zdp-1 ( S5G–S5J Fig . ) . These results are consistent with the notion that APE1L and ZDP largely represent two different mechanisms ( AP endonuclease vs 3′-phosphatase ) downstream of the DNA glycosylases/lyases , despite their redundant functions ( as 3′-phosphatases ) . ROS1 and ZDP mRNA levels are decreased in RdDM pathway mutants [17] . We examined the expression of APE1L and ARP in the RdDM mutants nrpd1-3 and nrpe1-11 , and found no substantial decreases in the mRNA levels in the mutants compared to Col ( S6A Fig . ) . We also measured the expression levels of ROS1 and ZDP in ape1l-1 and arp-1 mutants , and found that the expression levels are similar in the mutants compared to those in the Col or Ws wild type control plants ( S6B Fig . ) . Also , unlike the zdp-1 mutant , which is hypersensitive to MMS induced DNA damage , the ape1l-1 and arp-1 mutants show a sensitivity level similar to that of wild-type plants ( S7A Fig . ) . To study the potential genetic interactions between APE1L and ZDP , we crossed ape1l-1 and zdp-1 mutant plants . Interestingly , we found that ape1l+/−zdp−/− and ape1l−/−zdp+/− plants produce many aborted seeds , suggesting that the double mutations of APE1L and ZDP are lethal ( Fig . 5A ) . We grew the viable seeds , genotyped the seedlings , and found no ape1l−/−zdp−/− plants ( Table 1 ) . The ratio of aborted seeds is 48 . 7% in self-pollinated ape1l+/−zdp−/− plants and 26 . 5% in self-pollinated ape1l−/−zdp+/− plants ( S5 Table ) . Approximately seven days after pollination , the seeds fated to abortion show white color and plump phenotypes ( S8A Fig . ) . The endosperm in those seeds fails to undergo cellularization and the growth of their embryos is arrested ( S8B Fig . ) . Later , those seeds accumulate brown pigments and collapse . The 48 . 7% seed abortion ratio of self-pollinated ape1l+/−zdp−/− plants suggests that the lethality of this mutant may be maternally regulated . Also , because APE1L and ZDP may act downstream of ROS1 and DME and some of the characteristics of seed abortion in ape1l+/−zdp−/− and ape1l−/−zdp+/− mutants resemble those in the dme+/− mutant , we examined whether double mutations of APE1L and ZDP , like dme mutation , are also maternally lethal . If so , all seeds derived from a female gametophyte with APE1L and ZDP double mutations will abort irrespective of the paternal allele . We crossed ape1l+/−zdp−/− ( ♀ ) with ape1l+/+zdp−/− ( ♂ ) and the cross resulted in about 50% aborted seeds and 50% viable seeds . When we crossed them in a reverse direction , we observed 100% viable seeds ( Fig . 5B and S5 Table ) . Furthermore , when we crossed ape1l+/−zdp−/− ( ♀ ) plants with wild type plants , approximately 50% of the seeds aborted . These data indicate that ape1+/−zdp−/− mutant is indeed maternally lethal ( Fig . 5B and S5 Table ) . However , the ape1l−/−zdp+/− mutant is not maternally lethal , based on the fact that few seeds aborted when we crossed ape1l−/−zdp+/− to Col or ape1l−/−zdp+/+ in either directions ( Fig . 5B and S5 Table ) . This is consistent with its seed abortion ratio ( 26 . 5% ) ( S5 Table ) and segregation ratio ( ape1l−/−zdp+/+∶ape1l−/−zdp+/−∶ape1l−/−zdp−/− = 0 . 98∶2∶0 ) ( Table 1 ) when self pollinated . We examined the morphology of aborting seeds from ape1l+/−zdp−/− and ape1l−/−zdp+/− mutants using differential interference contrast microscopy . The major defects of aborting seeds are arrested embryo growth at the heart stage or earlier ( S8B Fig . ) and abnormal sizes of endosperm nuclei ( S8C Fig . ) . In some aborting seeds , the embryos are invisible , indicating that the embryos are arrested very early in development . The aborting seeds of ape1l+/−zdp−/− and dme+/− both display arrested embryo growth . Unlike ape1l+/−zdp−/− mutant seeds , dme+/− mutant seeds display clumps of unknown structures but there were no aberrant endosperm nuclei ( S8B–C Fig . ) . We noticed that the ape1l+/−zdp−/− mutant has abnormal segregation ratio ( 4 . 07∶1∶0 ) , which does not fit the expected segregation ratio of maternally lethal plant ( 1∶1∶0 ) . Alexander staining and in vitro germination assay were carried out to examine the pollen development in different mutants . The ape1l+/−zdp−/− mutant showed defects in pollen development and germination ( S9 Fig . ) , suggesting that the ape1l+/−zdp−/− mutation not only leads to maternal lethality but also gives rise to paternal defects . Maternal lethality phenotypes can be caused by aberrant expression of maternally imprinted genes and defects in the central cell or the endosperm [12] , [26] , [30] . FWA and FIS2 are two well-studied maternally imprinted genes , and their maternal expression in the endosperm relies on active DNA demethylation initiated by DME [12] , [23] . We investigated whether the methylation of the FWA and FIS2 promoters in endosperm tissues is affected by APE1L and ZDP double mutations ( Fig . 6A ) . The ape1l+/−zdp−/− plants were backcrossed to zdp−/− plants three times to minimize the Ws background . To examine the methylation levels of DME target genes in our mutants , we employed the method of Buzas et al . [36] where the DNA methylation specific restriction enzyme McrBC is used to digest DNA before doing q-PCR in seeds at 3 days post manual pollination . We found that after digestion with McrBC , the amount of DNA recovered from FWA and FIS2 promoter regions ( where is methylated in wild type leaf ) was reduced in both dme and ape1l−/−zdp−/− mutants compared with wild type , but there was no difference in the unmethylated FWA gene body region ( Fig . 6A ) . These results indicate that the ape1l−/−zdp−/− endosperm has hypermethylation in FWA and FIS2 promoter regions . In order to measure the mRNA levels of FWA and MEA in Col and ape1l−/−zdp−/− endosperms , we carried out real-time PCR and found that the expression levels of FWA and MEA but not the DME and FIE mRNAs are down-regulated in the ape1l−/−zdp−/− mutant endosperm ( Fig . 6B ) . To confirm and further analyze the FWA expression change , we introduced a pFWA::ΔFWA-GFP reporter into the ape1l+/−zdp−/− and ape1l−/−zdp+/− mutants by crossing the mutants with a transgenic line expressing the reporter [23] . Both ape1l+/−zdp−/− and ape1l−/−zdp+/− plants produce about 50% seeds defective in pFWA::ΔFWA-GFP expression ( Fig . 6C–6D and S6 Table ) . To our surprise , ape1l−/−zdp+/−mutant also produced 50% GFP-off seeds even though it is not maternally lethal and it produces about 75% viable seeds ( Table 1 ) . It turns out that hypermethylation of pFWA::ΔFWA-GFP promoter and silencing of FWA-GFP can occur in mutants which do not show maternal lethality . In addition , it seems that GFP-off seeds can be viable , so 75% viable seeds may be comprised of 50% GFP-on seeds and 25% GFP-off seeds . Taken together , our data suggest that DNA hypermethylation and down-regulation of imprinted genes occur and may be the cause of defects in the ape1l−/−zdp−/− endosperm .
Active DNA demethylation in plants is initiated by the ROS1 subfamily of 5-meC DNA glycosylases/lyases and presumably completed through a base excision repair pathway [2] , [37] . Previous work has reported that the 3′-phosphatase ZDP and the scaffold DNA repair protein XRCC1 also function in active DNA demethylation in Arabidopsis [17] , [38] . AP endonucleases are known to catalyze post-excision events during base excision repair . Our study here demonstrates that APE1L , one of the Arabidopsis AP endonucleases , functions in active DNA demethylation by processing β-elimination products of the bifunctional 5-meC DNA glycosylases/lyases and generating a 3′-OH group . APE1L-mediated reaction comprises a new branch of the DNA demethylation pathway downstream of ROS1 , DME , DML2 and DML3 ( Fig . 7 ) . Our biochemical data show that APE1L has an additional , weak 3′-phosphatase activity , and thus may also function in the other branch , perhaps redundantly with ZDP , to process β , δ-elimination products . Interestingly , it has been recently reported that the wheat homolog of APE1L also possesses 3′-phosphatase and 3′-phosphodiesterase activities [39] . Our results suggest that APE1L not only functions downstream of ROS1 , DML2 and DML3 in vegetative tissues to prevent DNA hypermethylation but also functions together with ZDP downstream of DME to control DNA demethylation and gene imprinting in the central cell and endosperm and is thus important for seed development . Active DNA demethylation in mammals can be initiated through the deamination of 5meC by AID to generate thymine , or the oxidation of 5meC to generate 5-hydroxymethylcytosine ( 5hmC ) , and further to 5-formylcytosine ( 5fC ) and 5-carboxycytosine ( 5caC ) by the TET family of DNA dioxygenases [2] , [40]–[42] . 5fC and 5caC can be excised by the monofunctional DNA glycosylase TDG , whereas thymine can be removed by the monofunctional DNA glycosylase MBD4 . Thus , a base excision repair pathway is required for completing the DNA cleavage and cytosine insertion steps during active DNA demethylation in mammals . Little is known about the DNA repair factors involved in active DNA demethylation in mammals , but it is likely that mammalian APE functions in active DNA demethylation downstream of the DNA glycosylases . The ape1l-1 mutation leads to DNA hypermethylation in thousands of genomic regions , indicating that APE1L is required for DNA demethylation in these regions in Arabidopsis . Like mutations in 5-methylcytosine DNA glycosylases/lyases such as ROS1 , mutations in DNA repair enzymes downstream of these enzymes are expected to preclude active DNA demethylation and cause hypermethylation . Coordinating the DNA glycosylase/lyase and repair activity would be predicted to prevent an otherwise fatal accumulation of strand breaks throughout the genome [17] . APE1L and ROS1 physically interact in vitro and co-localize in vivo , strongly suggesting that these proteins form a complex which coordinates their activities . One may ask why the DNA demethylation pathway includes both lyase activity of ROS1 and AP endonuclease activity of APE1L . In a recent study , it was reported that Wheat APE1L has weak endonuclease activity but robust 3′-repair phosphodiesterase and 3′-phosphatase activities [43] . Even though we detected the endonuclease activity of Arabidopsis APE1L in vitro , it is possible that , like Wheat APE1L , Arabidopsis APE1L is weak in cleaving DNA backbone at AP sites when involved in DNA demethylation . In this case , the lysase activity of ROS1 is required for generating the DNA gap . The Arabidopsis ARP endonuclease also processes the 3′-PUA generated by ROS1 in vitro , although this activity is much weaker than APE1L . Our whole genome bisulfite sequencing data identified only a small number of DMRs in the arp-1 mutant . Therefore , ARP is unlikely to play a major role in DNA demethylation , at least under normal growth conditions . Interestingly , we detected many genomic regions that are hypomethylated in the ape1l-1 mutant . APE1L is a multifunctional enzyme; its APE and 3′-phosphatase activities may contribute to other DNA repair pathways in addition to active DNA demethylation , Thus , APE1L dysfunction may affect many DNA-related processes that directly or indirectly cause DNA hypomethylation . Compared to the ros1-4 and rdd mutations , ape1l and arp mutations induce higher percentages of hypermethylation in genic regions , whereas the zdp mutation induces a higher percentage of hypermethylation in TEs . The mechanisms underlying this genomic specificity are unclear , but it is possible that APE1L and ARP function redundantly in the demethylation of TEs , such that mutating either one individually does not cause hypermethylation . Unlike ZDP , which processes 3′-phosphate blocking ends and promotes the release of ROS1 from its products , APE1L converts both 3′-phosphate and 3′-PUA to 3′-OH , but does not increase the turnover of ROS1 . Although both ZDP and APE1L interact with ROS1 in vitro and co-localize with ROS1 in vivo , ZDP and APE1L do not show extensive co-localization . It is possible that ZDP and APE1L exist mostly in two different protein complexes ( Fig . 7 ) . ZDP dysfunction caused DNA hypermethylation and transcriptional silencing of a luciferase reporter driven by the RD29A promoter , although the mutant phenotype was less severe than ros1 mutants [17] . We hypothesized that at some DNA demethylation target regions , such as the RD29A promoter , the DNA glycosylases/lyases may use both β- and β , δ-elimination activities and thus require both APE and ZDP to process the intermediates and prevent transcriptional silencing . However , we found that the ape1l-1 and arp mutations did not affect expression of the reporter gene ( S7B Fig . ) . It is possible that APE1L may function redundantly with ARP and/or ZDP in demethylation of the RD29A promoter . zdp mutant showed sensitivity to MMS but ape1l and arp mutants are not sensitive to MMS probably because they carry out different reactions . In addition , APE1L , APE2 and ARP may play redundant roles in repairing MMS-induced DNA damage , such that the single mutation or double mutations are not sufficient to induce sensitivity to MMS . The choice between the APE branch and the ZDP branch of the active DNA demethylation pathway depends on the elimination mechanism used by the DNA glycosylases/lyase enzymes . It is unclear when and where a DNA glycosylases/lyase employs β-elimination , β , δ-elimination , or both . Knowing which genomic regions depend on APE1L and which depend on ZDP for demethylation would be helpful . However , because the zdp-1 and ape1l-1 mutants are in different ecotypes , it is not ideal to compare the genomic regions targeted by the two different branches of the demethylation pathway . The double mutations of APE1L and APE2 are embryonic lethal , but not paternally or maternally lethal based on our results and the segregation ratio of selfed ape1l+/−ape2−/− reported previously [31] . It is possible that the lethal phenotype caused by APE1L and APE2 double mutations reflect deficiencies of DNA repair . Interestingly , we found that the ape1l+/−zdp−/− mutant shows a maternal lethality phenotype , which has been shown to occur in other mutants that are defective in DNA demethylation , such as the dme and ssrp1 mutants [26] , [30] . Unexpectedly , only the ape1l+/−zdp−/− mutant shows maternal lethality but the ape1l−/−zdp+/− mutant is not maternally lethal . As a result of maternal lethality , about 50% of seeds abort in dme+/− and ape1l+/−zdp−/− mutants . In contrast , about 25% seeds abort in the ape1l−/−zdp+/− mutant . All of the aborting seeds display embryos arrested at early growth stages presumably because an abnormal endosperm cannot support normal growth of the embryo . The morphology of aborted seeds in the ape1l+/−zdp−/− and ape1l−/−zdp+/− mutants is almost the same as that in the ape1l+/−ape2−/− mutant , which is not maternally lethal and gives about 25% aborted seeds [31] . It is likely that the ape1l+/−ape2−/− mutant is also defective in DNA demethylation . Alternatively , this type of morphology ( arrested embryo and aberrant endosperm ) may reflect deficiencies of base excision repair . It is likely that APE1L and ZDP function downstream of DME in the active DNA demethylation pathway that controls seed development . However , the aborting seeds in ape1l+/−zdp−/− mutants have varied sizes of endosperm nuclei but the aborting seeds in dme+/− mutants have endosperm nuclei of uniform sizes . This phenotypic difference may arise because APE1L and ZDP have multiple functions in DNA demethylation and repair , whereas DME only participates in DNA demethylation . As in dme+/− mutants , the seed abortion phenotype in ape1l+/−zdp−/− mutants is associated with the hypermethylation of the FWA promoter and the MEA ISR , and reduced FWA and MEA expression . Similar to the ape1l+/−zdp−/− mutant , the ape1l−/−zdp+/− mutant also produces about 50% GFP-off seeds , suggesting that these two types of mutants are similarly defective in DNA demethylation of imprinted genes . The phenotype of the ape1l−/−zdp+/− mutant in pFWA-GFP silencing ( 50% GFP-off ) and seeds viability ( 25% aborted and 75% viable ) resembles that of the recently discovered atdre2 mutant [44] . Some other factors beyond DNA demethylation or some dosage effects must be differentially involved in different types of mutants , leading to maternally lethality in some mutants but not in others , even though they are all defective in the expression of imprinted genes . In summary , our results show that APE1L and ZDP are important regulators of gene imprinting in plants , and suggest that DME-initiated active DNA demethylation in the central cell and endosperm employs both APE- and ZDP-dependent mechanisms .
Full-length APE1L and APE2 cDNAs were subloned into pMAL-c2X ( New England Biolabs ) to generate MBP-APE1L and MBP-APE2 fusion proteins . The full-length ARP cDNA was subcloned into pET28a ( Novagen ) to generate a His-ARP fusion protein . Expression was induced in Escherichia coli BL21 ( DE3 ) dcm− Codon Plus cells ( Stratagene ) . MBP-APE1L and MBP-APE2 were purified by amylose affinity chromatography ( New England Biolabs ) and His-ARP was purified by affinity chromatography on a Ni2+-nitrilotriacetic acid column ( Amersham Biosciences ) . His-ROS1 and MBP-ROS1 were expressed and purified as previously described [15] , [34] . Site directed mutagenesis of APE1L was performed using the Quick-Change II XL kit ( Stratagene ) according to the manufacturer's instructions . The N224D mutation was introduced into pMal-APE1L by using the oligonucleotides APE1LN212D_F4 and APE1LN212D_R4 ( see S1 Table ) . The mutant sequence was confirmed by DNA sequencing , and the construct was used to transform E . coli strain BL21 ( DE3 ) dcm− Codon Plus cells ( Stratagene ) . Mutant protein was expressed and purified as described above for APE1L . Oligonucleotides used to prepare DNA substrates ( see S2 Table ) were synthesized by Integrated DNA Technologies [45] and purified by PAGE before use . Double-stranded DNA substrates were prepared by mixing a 5 µM solution of a 5′-fluorescein-labeled or 5′-Alexa Fluor-labeled oligonucleotide ( upper strand ) with a 10 µM solution of an unlabeled oligomer ( lower strand ) . For preparation of 1-nt gapped DNA , a 5 µM solution of the corresponding 5′-labelled oligonucleotide was mixed with 10 µM solutions of unlabelled 5′-phosphorylated oligonucleotides P30_51 and CGR . Annealing reactions were performed at 95°C for 5 min , followed by slow cooling to room temperature . To detect 5-meC DNA glycosylase/lyase activity , purified His-ROS1 ( 35 nM ) was incubated at 30°C for 4 h with a Alexa Fluor-labeled DNA duplex ( 20 nM ) , containing a single 5-meC , in a reaction mixture containing 50 mM Tris–HCl pH 8 . 0 , 1 mM DTT , 0 . 1 mg/ml BSA . In reactions containing APE1L , the mixture also included 200 mM NaCl and 1 mM MgCl2 . Reactions were stopped by adding 20 mM EDTA , 0 . 6% sodium dodecyl sulfate , and 0 . 5 mg/ml proteinase K , and the mixtures were incubated at 37°C for 30 min . DNA was extracted with phenol/chloroform/isoamyl alcohol ( 25∶24∶1 ) and ethanol precipitated at −20°C in the presence of 0 . 3 mM NaCl and 16 µg/ml glycogen . When the ROS1 reaction products were used as purified substrates for AP endonucleases ( see below ) , samples were resuspended in 5 µl of distilled water . Otherwise , they were resuspended in 10 µl of 90% formamide , heated at 95°C for 5 min , and separated in a 12% denaturing polyacrylamide gel containing 7 M urea . Alexa Fluor-labeled DNA was visualized using the blue fluorescence mode of the FLA-5100 imager and analyzed using Multigauge software ( Fujifilm ) . The AP endonuclease activity was detected using a DNA substrate containing a synthetic AP site ( tetrahydrofuran , THF ) opposite G . The 3′-phosphatase activity was assayed on a 1-nt gapped substrate containing 3′-phosphate and 5′-phosphate ends . The 3′-phosphodiesterase activity was tested on purified ROS1 products , which contain a mixture of fragments with 3′-PUA and 3′-phosphate termini . In all assays , purified AP endonucleases were incubated with DNA substrates ( 20 or 40 nM ) at 30°C for the indicated times in a reaction mixture containing 50 mM Tris–HCl pH 8 . 0 , 200 mM NaCl , 1 mM DTT , 0 . 1 mg/ml BSA and 1 mM MgCl2 . Reactions were stopped and products analyzed as indicated above . Purified MBP alone or MBP-APE1L ( 200 pmol ) in 100 µl of Column Buffer ( 20 mM Tris , pH 7 . 4 , 1 mM EDTA , 1 mM DTT , 0 . 5% Triton X-100 ) was added to 100 µl of amylose resin ( New England Biolabs ) and incubated for 1 h at 4°C . The resin was washed twice with 600 µl of Binding Buffer ( 10 mM Tris , pH 8 . 0 , 1 mM DTT , 0 . 01 mg/ml BSA ) . Purified His-ROS1 ( 15 pmol ) was incubated at 25°C for 1 h with either MBP or MBP-APE1L bound to resin . The resin was washed twice with Binding Buffer . Bound proteins were analyzed by Western blot using antibodies against His6 tag ( Novagen ) . EMSAs were performed using an Alexa Fluor-labeled duplex containing a gap flanked by 3′-phosphate and 5′-phosphate termini prepared as described above . The labeled duplex substrate ( 10 nM ) was incubated with MBP-APE1L and/or His-ROS1 at the indicated concentrations in DNA-binding reaction mixtures ( 10 µl ) containing 10 mM Tris HCl , pH 8 . 0 , 1 mM DTT , 10 µg/ml BSA . After 15 min incubation at 25°C , reactions were immediately loaded onto 0 . 2% agarose gels in 1× Tris acetate/EDTA . Electrophoresis was carried out in 1× Tris acetate/EDTA for 40 min at 80 V at room temperature . Alexa Fluor-labeled DNA was visualized in a FLA-5100 imager and analyzed using MultiGauge software ( Fujifilm ) . To investigate the interaction between APE1L and ROS1 , two constructs was generated: APE1L-Cluc and ROS1-Nluc . The BamHI and SalI sites were used for cloning APE1L genomic DNA into pCAMBIA1300-CLUC vector . ROS1 was introduced to NLUC by In-Fusion HD Cloning Kit ( Clontech ) . For protein interaction analysis , two combinatory constructs were transformed simultaneously into Nicotiana benthamiana leaves . To prevent the silencing of those genes , a virus p19 protein gene containing construct was transformed at the same time . After 3 d , 1 mM luciferin was sprayed onto the lower epidermis and kept in the dark for 5 min , then a CCD camera ( 1300B; Roper ) was used to capture the fluorescence signal at 21°C . Two T-DNA insertion mutants of the APE1L gene ( At3g48425 ) , INRA Flag240B06 and Salk_024194C , were used and they were referred to as ape1l-1 and ape1l-2 respectively . T-DNA insertions are present in the fifth exon and fourth intron of ARP in arp-1 ( SALK_021478 ) and arp-2 ( SAIL_866_H10 ) respectively . For all plants , seeds were sown on 1/2 MS plates containing 2% sucrose and 0 . 7% agar , stratified for 48 hours at 4°C and grown under long day conditions at 22°C . They were collected at 14 days or transplanted to soil . Immunofluorescence was performed in 2- to 3-week-old leaves as described by Pontes et al . , [46] . Nuclei preparations were incubated overnight at room temperature with rabbit anti-APE1L ( anti-APE1L antibodies were generated by injecting rabbits with a recombinant full length APE1L protein that was purified by affinity chromatography ) , anti-ZDP [17] and mouse anti-Flag ( F3165 , Sigma ) . Primary antibodies were visualized using mouse Alexa 488-conjugated and rabbit Alexa-594 secondary antibody at 1∶200 dilution ( Molecular Probes ) for 2 h at 37°C . DNA was counterstained using DAPI in Prolong Gold ( Invitrogen ) . Nuclei were examined with a Nikon Eclipse E800i epifluorescence microscope equipped with a Photometrics Coolsnap ES . Total RNA was extracted from 2-week-old seedlings using the RNeasy Plant Mini Kit ( QIAGEN ) . 2-µg RNA was used for the first-strand cDNA synthesis with the Super script III First-Strand Synthesis System ( Invitrogen ) for RT-PCR following the manufacturer's instructions . The cDNA synthesis reaction was then diluted five times , and 1 µl was used as template in a 20-µl PCR reaction with iQ SYBR Green Supermix ( Bio-Rad ) . All reactions were carried out on the iQ5 Multicolour Real-Time PCR Detection System ( Bio-Rad ) . The comparative threshold cycle ( Ct ) method was used for determining relative transcript levels ( Bulletin 5279 , Real-Time PCR Applications Guide , Bio-Rad ) , with TUB8 as an internal control . DNA was extracted from 2 g of 12-day-old seedlings grown in a growth chamber and sent to BGI ( Shenzhen , China ) for bisulfite treatment , library preparation , and sequencing . Images of seed phenotypes were captured using an Olympus SZX7 microscope equipped with a Canon Powershot A640 camera . For cleared whole-mount observation , immature seeds , that are 8 days after pollination , were cleared using chloral hydrate , glycerol , and water ( 8 g: 1 ml: 2 ml ) and photographed using a Leica DM6000 B differential interference contrast microscope equipped with a Leica DFC 425 camera . Fluorescence was detected with an Olympus BX53 fluorescence microscope equipped with an Olympus DP80 digital camera . The McrBC assay was performed according to Buzas et al [36] . Briefly , wild type and the apel1+/−zdp−/− mutant were pollinated with Ler pollen . 3 days after pollination , pools of GFP-on and GFP-off seeds were selected under a dissecting fluorescence microscope and more than 300 seeds were used for DNA extraction . Genomic DNA concentration was measured by Nanodrop . Approximately 1 µg of DNA was digested with 1 µL of McrBC overnight at 37°C . After digestion , DNA methylation levels at the specific loci were determined by real-time PCR using absolute quantification against a 1∶1 mixture of genomic DNA extracted from Col-0 and Ler leaves . Primers are listed in S1 Table . Female ape1l+/−zdp−/− plants ( Col-0 ) were crossed with male wild type plants . The endosperm plus seed coat fraction was collected for RNA purification using the Trizol method . DNAase treatment and LiCl precipitation were applied to remove DNA and polysaccharide contaminations , respectively . RNA was reverse transcribed into cDNA by the SuperScript III First-Strand Synthesis System ( Invitrogen ) with an oligo dT primer . Real-time PCR analysis was performed using SYBR Premix Ex Taq ( TaKaRa ) and CFX96 real-time system ( Bio-Rad ) . ACT11 was used as the internal control . We used whole-genome bisulfite sequencing to analyze the methylomes of Ws , ape1l-1 , arp-1 and zdp-1 mutant plants . The data set was deposited at NCBI ( GSE52983 ) .
|
DNA cytosine methylation ( 5-methylcytosine , 5-meC ) is an important epigenetic mark , and methylation patterns are coordinately controlled by methylation and demethylation reactions during development and reproduction . In plants , REPRESSOR OF SILENCING ( ROS1 ) is one of the well characterized 5-meC DNA glycosylases that initiate active DNA demethylation by 5-meC excision . Our previous work showed that a 3′-DNA phosphatase , ZDP , functions downstream of ROS1 during active DNA demethylation in Arabidopsis . Here we found that the apurinic/apyrimidinic endonuclease APE1L functions downstream of ROS1 in a ZDP-independent branch of the active DNA demethylation pathway in Arabidopsis . In plants , gene imprinting requires the 5-meC DNA glycosylase Demeter ( DME ) that has been proposed to initiate a base excision repair pathway for active DNA demethylation in the central cell in female gametophyte . However , besides DME , no other base excision repair enzymes have been found to be important for gene imprinting . Our results show that APE1L and ZDP act jointly downstream of DME to regulate gene imprinting in plants , and suggest that DME-initiated active DNA demethylation in the central cell and endosperm uses both APE- and ZDP-dependent mechanisms .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"developmental",
"biology",
"base",
"excision",
"repair",
"genomic",
"imprinting",
"gene",
"expression",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"dna",
"repair",
"dna",
"modification",
"epigenetics",
"dna",
"methylation"
] |
2015
|
An AP Endonuclease Functions in Active DNA Demethylation and Gene Imprinting in Arabidopsis
|
The brain's decoding of fast sensory streams is currently impossible to emulate , even approximately , with artificial agents . For example , robust speech recognition is relatively easy for humans but exceptionally difficult for artificial speech recognition systems . In this paper , we propose that recognition can be simplified with an internal model of how sensory input is generated , when formulated in a Bayesian framework . We show that a plausible candidate for an internal or generative model is a hierarchy of ‘stable heteroclinic channels’ . This model describes continuous dynamics in the environment as a hierarchy of sequences , where slower sequences cause faster sequences . Under this model , online recognition corresponds to the dynamic decoding of causal sequences , giving a representation of the environment with predictive power on several timescales . We illustrate the ensuing decoding or recognition scheme using synthetic sequences of syllables , where syllables are sequences of phonemes and phonemes are sequences of sound-wave modulations . By presenting anomalous stimuli , we find that the resulting recognition dynamics disclose inference at multiple time scales and are reminiscent of neuronal dynamics seen in the real brain .
Many aspects of our sensory environment can be described as dynamic sequences . For example , in the auditory domain , speech and music are sequences of sound-waves [1] , [2] , where speech can be described as a sequence of phonemes . Similarly , in the visual domain , speaking generates sequences of facial cues with biological motion [3] , [4] . These auditory and visual sequences have an important characteristic: the transitions between the elements are continuous; i . e . , it is often impossible to identify a temporal boundary between two consecutive elements . For example , phonemes ( speech sounds ) in a syllable are not discrete entities that follow each other like beads on a string but rather show graded transitions to the next phoneme . These transitions make artificial speech recognition notoriously difficult [5] . Similarly , in the visual domain , when we observe someone speaking , it is extremely difficult to determine exactly where the movements related to a phoneme start or finish . These dynamic sequences , with brief transitions periods between elements , are an inherent part of our environment , because sensory input is often generated by the fluent and continuous movements of other people , or indeed oneself . Dynamic sequences are generated on various time-scales . For example , in speech , formants form phonemes and phonemes form syllables . Sequences , which exist at different time-scales , are often structured hierarchically , where sequence elements on one time-scale constrain the expression of sequences on a finer time-scale; e . g . a syllable comprises a specific sequence of phonemes . This functional hierarchy of time-scales may be reflected in the hierarchical , anatomical organisation of the brain [6] . For example , in avian brains , there is anatomical and functional evidence that birdsong is generated and perceived by a hierarchical system , where low levels represent transient acoustic details and high levels encode song structure at slower time-scales [7] , [8] . An equivalent temporal hierarchy might also exist in the human brain for representing auditory information , such as speech [1] , [9]–[12] . Here we ask the following question: How does the brain recognize the dynamic and ambiguous causes of noisy sensory input ? Based on experimental and theoretical evidence [13]–[18] we assume the brain is a recognition system that uses an internal model of its environment . The structure of this model is critical: On one hand , the form of the model must capture the essential architecture of the process generating sensory data . On the other hand , it must also support robust inference . We propose that a candidate that fulfils both criteria is a model based on a hierarchy of stable heteroclinic channels ( SHCs ) . SHCs have been introduced recently as a model of neuronal dynamics per se [19] . Here , we use SHCs as the basis of neuronal recognition , using an established Bayesian scheme for modelling perception [20] . This brings together two recent developments in computational approaches to perception: Namely , winnerless competition in stable heteroclinic channels and the hypothesis that the brain performs Bayesian inference . This is important because it connects a dynamic systems perspective on neuronal dynamics [19] , [21] , [22] with the large body of work on the brain as an inference machine [13]–[18] . To demonstrate this we generate artificial speech input ( sequences of syllables ) and describe a system that can recognize these syllables , online from incoming sound waves . We show that the resulting recognition dynamics display functional characteristics that are reminiscent of psychophysical and neuronal responses .
SHCs are attractors formed by artificial neuronal networks , which prescribe sequences of transient dynamics [22]–[25] . The key aspect of these dynamical systems is that their equations of motion describe a manifold with a series of saddle points . At each saddle point , trajectories are attracted from nearly all directions but are expelled in the direction of another saddle point . If the saddle points are linked up to form a chain , the neuronal state follows a trajectory that passes through all these points , thereby forming a sequence . These sequences are exhibited robustly , even in the presence of high levels of noise . In addition , the dynamics of the SHCs are itinerant due to dynamical instability in the equations of motion and noise on the states . This noise also induces a variation in the exact times that sequence elements are visited . This can be exploited during recognition , where the SHC places prior constraints on the sequence that elements ( repelling fixed-points ) are visited but does not constrain the exact timing of these visits . The combination of these two features , robustness of sequence order but flexibility in sequence timing , makes the SHC a good candidate for the neuronal encoding of trajectories [19] , [26] . Rabinovich et al . have used SHCs to explain how spatiotemporal neuronal dynamics observed in odour perception , or motor control of a marine mollusc , can be expressed in terms of a dynamic system [22] , [27] . Varona et al . used Lotka-Volterra-type dynamics to model a network of six neurons in a marine mollusc [27]: With particular lateral inhibition between pairs of neurons and input to each neuron , the network displayed sequences of activity . Following a specific order , each neuron became active for a short time and became inactive again , while the next neuron became active , and so on . Stable heteroclinic channels rest on a particular form of attractor manifold that supports itinerant dynamics . This itinerancy can result from deterministic chaos in the absence of noise , which implies the presence of heteroclinic cycles . When noise is added , itinerancy can be assured , even if the original system has stable fixed-points . However , our motivation for considering stochastic differential equations is to construct a probabilistic model , where assumptions about the distribution of noise provide a formal generative model of sensory dynamics . As reviewed in [22] , Lotka-Volterra dynamics can be derived from simple neural mass models of mean membrane potential and mean firing rate [21] . Here , we use a different neural mass model , where the state-vector x can take positive or negative values: ( 1 ) where the motion of a hidden-state vector ( e . g . , mean membrane potentials ) x is a nonlinear function of itself with scalar parameters , , and a connectivity matrix . The hidden state-vector enters a nonlinear function S to generate outcomes ( e . g . , neuronal firing rates ) y . Each element determines the strength of lateral inhibition from state j to i . Both the state and observation equations above include additive normally distributed noise vectors w and z . When choosing specific parameter values ( see below ) , the states display stereotyped sequences of activity [28] . Rabinovich et al . [19] termed these dynamics ‘stable heteroclinic channels’ ( SHCs ) . If the channel forms a ring , once a state is attracted to a saddle point , it will remain in the SHC . SHCs represent a form of itinerant dynamics [26] , [29] , [30] and may represent a substrate for neuronal computations [31] . Remarkably , the formation of SHCs seems to depend largely on the lateral inhibition matrix and not on the type of neuronal model; see Ivanchenko et al . [32] for an example using a complex two-compartment spiking neuron model . In this paper , we propose to use SHCs not as a model for neuronal dynamics per se but as a generative model of how sensory input is generated . This means that we interpret x as hidden states in the environment , which generate sensory input y . The neuronal response to sampling sensory input y are described by recognition dynamics , which decode or deconvolve the causes x from that input . These recognition dynamics are described below . This re-interpretation of Eq . 1 is easy to motivate: sensory input is usually generated by our own body and other organisms . This means input is often generated by neuronal dynamics of the sort described in Eq . 1 . A SHC can generate repetitive , stereotyped sequences . For example , in a system with four saddle points , an SHC forces trajectories through the saddle points in a sequence , e . g . ‘1-2-3-4-1-2-3-4-1…’ . In contrast , a SHC cannot generate ‘1-2-3-4-3-4-2-1…’ , because the sequence is not repetitive . However , to model sensory input , for example speech , one must be able to recombine basic sequence-elements like phonemes in ever-changing sequences . One solution would be to represent each possible sequence of phonemes ( e . g . each syllable ) with a specific SHC . A more plausible and parsimonious solution is to construct a hierarchy of SHCs , which can encode sequences generated by SHCs whose attractor topology ( e . g . the channels linking the saddle points ) is changed by a supraordinate SHC . This can be achieved by making the connectivity matrix at a subordinate level a function of the output states of the supra-ordinate level . This enables the hierarchy to generate sequences of sequences to any hierarchical depth required . Following a recent account of how macroscopic cortical anatomy might relate to time-scales in our environment [6] , we can construct a hierarchy by setting the rate constant of the j-th level to a rate that is slower than its subordinate level , . As a result , the states of subordinate levels change faster than the states of the level above . This means the control parameters at any level change more slowly than its states , ; because the slow change in the attractor manifold is controlled by the supraordinate states: ( 2 ) where the superscript indexes level j ( level 1 being the lowest level ) , are ‘hidden states’ , and are outputs to the subordinate level , which we will call ‘causal states’ . As before , at the first level , is the sensory stream . In this paper , we consider hierarchies with relative time-scales of around four . This means that the time spent in the vicinity of a saddle point at a supraordinate level is long enough for the subordinate level to go through several saddle points . As before , all levels are subject to noise on the motion of the hidden states and the causal states . At the highest level , the control parameters , are constant over time . At all other levels , the causal states of the supraordinate level , , enter the subordinate level by changing the control parameters , the connectivity matrix : ( 3 ) Here , is a linear mixture of ‘template’ control matrices , weighted by the causal states at level . Each of these templates is chosen to generate a SHC . Below , we will show examples of how these templates can be constructed to generate various sequential phenomena . The key point about this construction is that states from the supraordinate level select which template controls the dynamics of the lower level . By induction , the states at each level follow a SHC because the states at the supraordinate level follow a SHC . This means only one state is active at any time and only one template is selected for the lower level . An exception to this is the transition from one state to another , which leads to a transient superposition of two SHC-inducing templates ( see below ) . Effectively , the transition transient at a specific level gives rise to brief spells of non-SHC dynamics at the subordinate levels ( see results ) . These transition periods are characterized by dissipative dynamics , due to the largely inhibitory connectivity matrices , inhibition controlled by parameter ( Eq . 2 ) and the saturating nonlinearity S . In summary , a hierarchy of SHCs generates the sensory stream at the lowest ( fastest ) level , which forms a sequence of sequences expressed in terms of first-level states . In these models , the lower level follows a SHC , i . e . the states follow an itinerant trajectory through a sequence of saddle points . This SHC will change whenever the supraordinate level , which follows itself a SHC , moves from one saddle point to another . Effectively , we have constructed a system that can generate a stable pattern of transients like an oscillator; however , as shown below , the pattern can have deep or hierarchical structure . Next , we describe how the causes can be recognized or deconvolved from sensory input y . We have described how SHCs can , in principle , generate sequences of sequences that , we assume , are observed by an agent as its input y . To recognise the causes of the sensory stream the agent must infer the hidden states online , i . e . the system does not look into the future but recognizes the current states and of the environment , at all levels of the hierarchy , by the fusion of current sensory input and internal dynamics elicited by past input . An online recognition scheme can be derived from the ‘free-energy principle’ , which states that an agent will minimize its surprise about its sensory input , under a model it entertains about the environment; or , equivalently maximise the evidence for that model [18] . This requires the agent to have a dynamic model , which relates environmental states to sensory input . In this context , recognition is the Bayesian inversion of a generative model . This inversion corresponds to mapping sensory input to the posterior or conditional distribution of hidden states . In general , Bayesian accounts of perception rest on a generative model . Given such a model , one can use the ensuing recognition schemes in artificial perception and furthermore compare simulated recognition dynamics ( in response to sensory input ) , with evoked responses in the brain . The generative model in this paper is dynamical and based on the nonlinear equations 1 and 2 . More precisely , these stochastic differential equations play the role of empirical priors on the dynamics of hidden states causing sensory data . In the following , we review briefly , the Bayesian model inversion described in [20] for stochastic , hierarchical systems and apply it , in the next section , to hierarchical SHCs . Given some sensory data vector y , the general inference problem is to compute the model evidence or marginal likelihood of y , given a model m: ( 4 ) where the generative model is defined in terms of a likelihood and prior on hidden states . In Equation 4 , the state vector subsumes the hidden and causal states at all levels of a hierarchy ( Eq . 2 ) . The model evidence can be estimated by converting this difficult integration problem ( Eq . 4 ) into an easier optimization problem by optimising a free-energy bound on the log-evidence [33] . This bound is constructed using Jensen's inequality and is a function of an arbitrary recognition density , : ( 5 ) The free-energy comprises an energy term and an entropy term and is defined uniquely , given a generative model . The free-energy is an upper bound on the surprise or negative log-evidence , because the Kullback-Leibler divergence , between the recognition and conditional density , is always positive . Minimising the free-energy minimises the divergence , rendering the recognition density an approximate conditional density . When using this approach , one usually employs a parameterized fixed-form recognition density , [20] . Inference corresponds to optimising the free-energy with respect to the sufficient statistics , of the recognition density: ( 6 ) The optimal statistics are sufficient to describe the approximate posterior density; i . e . the agent's belief about ( or representation of ) the trajectory of the hidden and causal states . We refer the interested reader to Friston et al . [34] for technical details about this variational Bayesian treatment of dynamical systems . Intuitively , this scheme can be thought of as augmented gradient descent on a free-energy bound on the model's log-evidence . Critically , it outperforms conventional Bayesian filtering ( e . g . , Extended Kalman filtering ) and eschews the computation of probability transition matrices . This means it can be implemented in a simple and neuronally plausible fashion [20] . In short , this recognition scheme operates online and recognizes current states of the environment by combining current sensory input with internal recognition dynamics , elicited by past input . A recognition system that minimizes its free-energy efficiently will come to represent the environmental dynamics in terms of the sufficient statistics of recognition density; e . g . the conditional expectations and variances of . We assume that the conditional moments are encoded by neuronal activity; i . e . , Equation 6 prescribes neuronal recognition dynamics . These dynamics implement Bayesian inversion of the generative model , under the approximations entailed by the form of the recognition density . Neuronally , Equation 6 can be implemented using a message passing scheme , which , in the context of hierarchical models , involves passing prediction errors up and passing predictions down , from one level to the next . These prediction errors are the difference between the causal states ( Equation 2 ) ; ( 7 ) at any level j , and their prediction from the level above , evaluated at the conditional expectations [18] , [35] . In addition , there are prediction errors that mediate dynamical priors on the motion of hidden states within each level ( Equation 2 ) ; ( 8 ) This means that neuronal populations encode two types of dynamics: the conditional expectations of states of the world and the prediction errors . The dynamics of the first are given by Equation 6 , which can be formulated as a function of prediction error . These dynamics effectively suppress or explain away prediction error; see [34] for details . This inversion scheme is a generic recognition process that receives dynamic sensory input and can , given an appropriate generative model , rapidly identify and track environmental states that are generating current input . More precisely , the recognition dynamics resemble the environmental ( hidden ) states they track ( to which they are indirectly coupled ) , but differ from the latter because they are driven by a gradient descent on free-energy; Eq . 6 ( i . e . minimize prediction errors: Eqs . 7 and 8 ) . This is important , because we want to use SHCs as a generative model , not as a model of neuronal encoding per se . This means that the neuronal dynamics will only recapitulate the dynamics entitled by SHCs in the environment , if the recognition scheme can suppress prediction errors efficiently in the face of sensory noise and potential beliefs about the world . We are now in a position to formulate hierarchies of SHCs as generative models , use them to generate sensory input and simulate recognition of the causal states generating that input . In terms of low-level speech processing , this means that any given phoneme will predict the next phoneme . At the same time , as phonemes are recognized , there is also a prediction about which syllable is the most likely context for generating these phonemes . This prediction arises due to the learnt regularities in speech . In turn , the most likely syllable predicts the next phoneme . This means that speech recognition can be described as a dynamic process , on multiple time-scales , with recurrently evolving representations and predictions , all driven by the sensory input . In the auditory system , higher cortical levels appear to represent features that are expressed at slower temporal scales [36] . Wang et al . [37] present evidence from single-neuron recordings that there is a ‘slowing down’ of representational trajectories from human auditory sensory thalamus ( a ‘relay’ to the primary auditory cortex ) , the medial geniculate body ( MGB ) to primary auditory cortex ( AI ) . In humans , it has been found that the sensory thalamus responds preferentially to faster temporal modulations of sensory signals , whereas primary cortex prefers slower modulations [10] . These findings indicate that neuronal populations , at lower levels of the auditory system ( e . g . MGB ) , represent faster environmental trajectories than higher levels ( e . g . , A1 ) . Specifically , the , MGB responds preferentially to temporal modulations of ∼20 Hz ( ∼50 ms ) , whereas AI prefers modulations at ∼6 Hz ( ∼150 ms ) [10] . Such a temporal hierarchy would be optimal for speech recognition , in which information over longer time-scales provides predictions for processing at shorter time scales . In accord with this conjecture , optimal encoding of fast ( rapidly modulated ) dynamics by top-down predictions has been found to be critical for communication [1] , [12] , [38] . We model this ‘slowing down’ with a hierarchical generative model based on SHCs . This model generates sequences of syllables , where each syllable is a sequence of phonemes . Phonemes are the smallest speech sounds that distinguishes meaning and a syllable is a unit of organization for a sequence of phonemes . Each phoneme prescribes a sequence of sound-wave modulations which correspond to sensory data . We generated data in this fashion and simulated online recognition ( see Figure 1 ) . By recognizing speech-like phoneme-sequences , we provide a proof-of-principle that a hierarchical system can use sensory streams to infer sequences . This not only models the slowing down of representations in the auditory system [10] , [12] , [37] , [38] , but may point to computational approaches to speech recognition . In summary , the recognition dynamics following Equation 6 are coupled to a generative model based on SHCs via sensory input . The systems generating and recognising states in Fig . 1 are both dynamic systems , where a non-autonomous recognition system is coupled to an autonomous system generating speech . All our simulations used hierarchies with two levels ( Figure 2 ) . The first ( phonemic ) level produces a sequence of phonemes , and the second ( syllabic ) level encodes sequences of syllables . We used Equation 2 to produce phoneme sequences , where the generating parameters are listed in Table 3 . The template matrices ( Equation 3 ) were produced in the following way: We first specified the sequence each template should induce; e . g . , sequence 1-2-3 for three neuronal populations . We then set elements on the main diagonal to 1 , the elements ( 2 , 1 ) , ( 3 , 2 ) , ( 1 , 3 ) to value 0 . 5 , and all other elements to 5 [28] . More generally for sequence ( 9 ) Note that SHC hierarchies can be used to create a variety of different behaviours , using different connectivity matrices . Here we explore only a subset of possible sequential dynamics . When generating sensory data y , we added noise and to both the hidden and causal states . At the first and second levels , this was normally distributed zero-mean noise with log-precisions of ten and sixteen , respectively . These noise levels were chosen to introduce noisy dynamics but not to the extent that the recognition became difficult to visualise . We repeated all the simulations reported below with higher noise levels and found that the findings remained qualitatively the same ( results not shown ) . Synthetic stimuli were generated by taking a linear mixture of sound waves extracted from sound files , in which a single speaker pronounced each of four vowel-phonemes: [a] , [e] , [i] , [o] . These extracts W were sampled at 22050 Hz and about 14 ms long . The mixture was weighted by the causal states of the phonemic level; . This resulted in a concatenated sound wave file w . When this sound file is played , one perceives a sequence of vowels with smooth , overlapping transitions ( audio file S1 ) . These transitions are driven by the SHCs guiding the expression of the phonemes and syllables at both levels of the generative hierarchy . For computational simplicity , we circumvented a detailed generative model of the acoustic level . For simulated recognition , the acoustic input ( the sound wave ) was transformed to phonemic input by inverting the linear mixing described above every seven ms of simulated time ( one time bin ) . This means that our recognition scheme at the acoustic level assumes forward processing only ( Fig . 1 ) . However , in principle , given an appropriate generative model [39] , [40] , one could invert a full acoustic model , using forward and backward message passing between the acoustic and phonemic levels .
To create synthetic stimuli we generated syllable sequences consisting of four phonemes or states; [a] , [e] , [i] , and [o] , over 11 . 25 seconds ( 800 time points ) , using a two-level SHC model ( Fig . 2 ) . To simulate word-like stimuli , we imposed silence at the beginning and the end by windowing the phoneme sequence ( Fig . 3A , top left ) . At the syllabic level , we used three syllables or states to form the second-level sequence ( 1–2–3 ) ( 2 ) ; where the numbers denote the sequence and the superscript indicates the sequence level . The three causal states of the syllabic level entered the phonemic level as control parameters to induce their template matrices as in Equation 3 . This means that each of the three syllable states at the second level causes a phoneme sequence at the first: , , and , see Fig . 2 and listen to the audio file S1 . In Fig . 3A we show the causal and hidden states , at both levels , generated by this model . The remaining parameters , for both levels , are listed in Table 3 . Note that the rate constant of the syllabic level is four times slower than at the phonemic level . As expected , the phoneme sequence at the first level changes as a function of the active syllable at the second level . The transients caused by transitions between syllables manifest at the first level as temporary changes in the amplitude or duration of the active phoneme . We then simulated recognition of these sequences . Fig . 3B shows that our recognition model successfully tracks the true states at both levels . Note the recognition dynamics rapidly ‘lock onto’ the causal states from the onset of the first phoneme of the first syllable ( time point 50 ) . Interestingly , the system did not recognize the first syllable ( true: syllable 3 ( red line ) , recognized: syllable 2 ( green line ) between time points 50 to 80 ( see red arrow in Fig . 3B ) , but corrected itself fairly quickly , when the sensory stream indicated a new phoneme that could only be explained by the third syllable . This initial transient at the syllabic level shows that recognition dynamics can show small but revealing deviations from the true state dynamics . In principle , these deviations could be used to test whether the real auditory system uses a recognition algorithm similar to the one proposed; in particular , the simulated recognition dynamics could be used to explain empirical neurophysiological responses . What happens if the stimuli deviate from learned expectations ( e . g . violation of phonotactic rules ) ? In other words , what happens if we presented known phonemes that form unknown syllables ? This question is interesting for two reasons . First , our artificial recognition scheme should do what we expect real brains to do when listening to a foreign language: they should be able to recognize the phonemes but should not derive high-order ‘meaning’ from them; i . e . should not recognize any syllable . Secondly , there are well-characterised brain responses to phonotactic violations , e . g . [41]–[43] . These are usually event-related responses that contain specific waveform components late in peristimulus time , such as the N400 . The N400 is an event-related potential ( ERP ) component typically elicited by unexpected linguistic stimuli . It is characterized as a negative deflection ( topologically distributed over central-parietal sites on the scalp ) , peaking approximately 400 ms after the presentation of an unexpected stimulus . To model phonotactic violations , we generated data with the two-level model presented above . However , we used syllables , i . e . sequences of phonemes , that the recognition scheme was not informed about and consequently could not recognise ( it has three syllables in its repertoire: , , and ) . Thus the recognition scheme knows all four phonemes but is unable to predict the sequences heard . Fig . 4A shows that the recognition system cannot track the syllables; the recognized syllables are very different from the true syllable dynamics . At the phonemic level , the prediction error deviates from zero whenever a new ( unexpected ) phoneme is encountered ( Fig . 4B ) . The prediction error at the syllabic level is sometimes spike-like and can reach high amplitudes , relative to the typical amplitudes of the true states ( see Fig . 4A and B ) . This means that the prediction error signals violation of phonotactic rules . In Fig . 4C , we zoom in onto time points 440 to 470 to show how the prediction error evolves when evidence of a phonotactic violation emerges: At the phoneme level , prediction error builds up because an unexpected phoneme appears . After time point 450 , the prediction error grows quickly , up to the point that the system resolves the prediction error . This is done by ‘switching’ to a new syllable , which can explain the transition to the emerging phoneme . The switching creates a large amplitude prediction error at time point 460 . In other words , in face of emerging evidence that its current representation of syllables and phonemes cannot explain sensory input , the system switches rapidly to a new syllable representation , giving rise to a new prediction error . It may be that these prediction errors are related to electrophysiological responses to violations of phonotactic rules , [44] , [45] . This is because the largest contributors to non-invasive electromagnetic signals are thought to be superficial pyramidal cells . In biological implementations of the recognition scheme used here [20] , these cells encode prediction error . In summary , these simulations show that a recognition system cannot represent trajectories or sequences that are not part of its generative model . In these circumstances , recognition experiences intermittent high-amplitude prediction errors because the internal predictions do not match the sensory input . There is a clear formal analogy between the expression of prediction error in these simulations and mismatch or prediction violation responses observed empirically . The literature that examines event-related brain potentials ( ERPs ) and novelty processing “reveals that the orienting response engendered by deviant or unexpected events consists of a characteristic ERP pattern , comprised sequentially of the mismatch negativity ( MMN ) and the novelty P3 or P3a” [46] . Human speech recognition is robust to the speed of speech [47] , [48] . How do our brains recognize speech at different rates ? There are two possible mechanisms in our model that can deal with ‘speaker speed’ parameters online . First , one could make the rate constants and free parameters and optimise them during inversion . Adjusting to different speaker parameters is probably an essential faculty , because people speak at different speeds [49] . The second mechanism is that the recognition itself might be robust to deviations from the expected rate of phonemic transitions; i . e . , even though the recognition uses the rate parameters appropriate for much slower speech , it still can recognize fast speech . This might explain why human listeners can understand speech at rates that they have never experienced previously [47] . In the following , we show that our scheme has this robustness . To simulate speed differences we used the same two-level model as in the simulations above with for the generation of phonemes , but with for recognition so that the stimulus stream was 50% faster than expected . As can be seen in Fig . 5A , the recognition can successfully track the syllables . This was because the second level supported the adaption to the fast sensory input by changing its recognition dynamics in responses to prediction error ( see Fig . 5B: note the amplitude difference in Fig . 5A between the true and recognized ) . The prediction errors at both levels , and , are shown in Fig . 5C . In particular , the second-level error displayed spike-like corrections around second-level transitions . These are small in amplitude compared to both the amplitude of the hidden states and the prediction errors of the previous simulation ( Fig . 4B ) . These results show that the system can track the true syllables veridically , where the prediction error accommodates the effects caused by speed differences . This robustness to variations in the speed of phoneme transitions might be a feature shared with the auditory system [50] .
There is emerging evidence in several areas of neuroscience that temporal hierarchies play a critical role in brain function [6] . The three areas where this is most evident are auditory processing [12] , [37] , [54]–[56] , cognitive control [57]–[59] , and motor control [60] . Our conclusions are based on a generic recognition scheme [20] and are therefore a consequence of our specific generative model , a temporal hierarchy of SHCs . This hierarchy of time-scales agrees well with the temporal anatomy of the hierarchical auditory system , where populations close to the periphery encode the fast acoustics , while higher areas form slower representations [9] , [10] , [37] , [38] , [61] , [62] . In particular , our model is consistent with findings that phonological ( high ) levels have strong expectations about the relevance of acoustic ( low ) dynamics [38] . Neurobiological treatments of the present framework suppose that superficial pyramidal cell populations encode prediction error; it is these cells that contribute most to evoked responses as observed in magneto/electroencephalography ( M/EEG ) [63] . There is an analogy between the expression of prediction error in our simulations and mismatch or prediction violation responses observed empirically . In our simulations , prediction error due to a deviation from expectations is resolved by all levels ( Fig . 4B ) . This might be an explanation for prominent responses to prediction violations to be spatially distributed , e . g . , the mismatch negativity , the P300 , and the N400 all seem to involve various brain sources in temporal and frontal regions [45] , [46] , [64]–[66] . Inference on predictable auditory streams has been studied and modelled in several ways , in an attempt to explain the rapid recognition of words in the context of sentences , e . g . , [38] , [67]–[70] . Our simulations show how , in principle , these accounts might be implemented in terms of neuronal population dynamics . Learning , storing , inferring and executing sequences is a key topic in experimental [71]–[79] , and theoretical neurosciences [80]–[82]; and robotics [83]–[86] . An early approach to modelling sequence processing focussed on feed-forward architectures . However , it was realised quickly that these networks could not store long sequences , because new input overwrote the internal representation of past states . The solution was to introduce explicit memory into recurrent networks , in various forms; e . g . as contextual nodes or ‘short-term memory’ [87] , [88] . Although framed in different terms , these approaches can be seen as an approximation to temporal hierarchies , where different units encode representations at different time-scales . A central issue in modelling perception is how sequences are not just recalled but used as predictions for incoming sensory input . This requires the ‘dynamic fusion’ of bottom-up sensory input and top-down predictions , Several authors e . g . , [83] , [89]–[92] use recurrent networks to implement this fusion . Exact Bayesian schemes based on discrete hierarchical hidden Markov models , specified as a temporal hierarchy , have been used to implement memory and recognition [93] . Here , we have used the free-energy principle ( i . e . variational Bayesian inference on continuous hierarchical dynamical systems ) to show how the ensuing recognition process leads naturally to a scheme which can deal with fast sequential inputs . In conclusion , we have described a scheme for inferring the causes of sensory sequences with hierarchical structure . The key features of this scheme are: ( i ) the ability to describe natural sensory input as hierarchical and dynamic sequences , ( ii ) modeling this input using generative models , ( iii ) using dynamic systems theory to create plausible models , and ( iv ) online Bayesian inversion of the resulting models . This scheme is theoretically principled but is also accountable to the empirical evidence available from the auditory system; furthermore , the ensuing recognition dynamics are reminiscent of real brain responses .
|
Despite tremendous advances in neuroscience , we cannot yet build machines that recognize the world as effortlessly as we do . One reason might be that there are computational approaches to recognition that have not yet been exploited . Here , we demonstrate that the ability to recognize temporal sequences might play an important part . We show that an artificial decoding device can extract natural speech sounds from sound waves if speech is generated as dynamic and transient sequences of sequences . In principle , this means that artificial recognition can be implemented robustly and online using dynamic systems theory and Bayesian inference .
|
[
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] |
[
"neuroscience/theoretical",
"neuroscience",
"neuroscience/sensory",
"systems",
"computational",
"biology/computational",
"neuroscience"
] |
2009
|
Recognizing Sequences of Sequences
|
We need to find ways of enhancing the potency of existing antibiotics , and , with this in mind , we begin with an unusual question: how low can antibiotic dosages be and yet bacterial clearance still be observed ? Seeking to optimise the simultaneous use of two antibiotics , we use the minimal dose at which clearance is observed in an in vitro experimental model of antibiotic treatment as a criterion to distinguish the best and worst treatments of a bacterium , Escherichia coli . Our aim is to compare a combination treatment consisting of two synergistic antibiotics to so-called sequential treatments in which the choice of antibiotic to administer can change with each round of treatment . Using mathematical predictions validated by the E . coli treatment model , we show that clearance of the bacterium can be achieved using sequential treatments at antibiotic dosages so low that the equivalent two-drug combination treatments are ineffective . Seeking to treat the bacterium in testing circumstances , we purposefully study an E . coli strain that has a multidrug pump encoded in its chromosome that effluxes both antibiotics . Genomic amplifications that increase the number of pumps expressed per cell can cause the failure of high-dose combination treatments , yet , as we show , sequentially treated populations can still collapse . However , dual resistance due to the pump means that the antibiotics must be carefully deployed and not all sublethal sequential treatments succeed . A screen of 136 96-h-long sequential treatments determined five of these that could clear the bacterium at sublethal dosages in all replicate populations , even though none had done so by 24 h . These successes can be attributed to a collateral sensitivity whereby cross-resistance due to the duplicated pump proves insufficient to stop a reduction in E . coli growth rate following drug exchanges , a reduction that proves large enough for appropriately chosen drug switches to clear the bacterium .
Bacteria have a remarkable capacity to adapt and evolve . It is probably unsurprising in retrospect that resistance has developed to every antibiotic in clinical use [1] , with the genes responsible disseminated globally [2 , 3] . Antibiotic resistance , therefore , has the potential to become a very grave problem . Bacteria evolve so rapidly , in fact , that whole-genome sequencing studies have been able to elucidate dozens of de novo drug-resistance mutations occurring at high frequency within a clinical patient’s infection during a 12-wk treatment [4] . Given this , the following seems an important question: what ways of combining antibiotics might be used to combat infection even when the bacterial species in question exhibits rapid decreases in drug susceptibility during treatment ? Or , to put it differently , how can we enlarge the “optimisation space” of antibiotic combinations and search within those for novel , effective treatments ? One possibility may lie with so-called sequential treatments . They have been the subject of several recent laboratory studies [5–7] and clinical trials [8 , 9] in which the idea is to alternate the use of different antibiotic classes through time . Thus , if , for example , two antibiotics are available and n rounds of treatment are to be given , then there are 2n different ways of administering the drugs . Our hypothesis states that this exponentially large optimisation space can contain more effective treatments than the equivalent two-drug combination treatment when the same dosages of each antibiotic are applied . We demonstrate the veracity of this claim in one particular in vitro laboratory model that mimics something of the gravity of the situation we now face by using a bacterium that possesses a scalable drug efflux mechanism that quickly reduces the efficacy of the antibiotics at our disposal . Despite this mechanism , we show that sequential treatments can clear the bacterium when the equivalent combination treatment fails to , provided , that is , that the drugs are deployed in a suitably optimised , sequential manner . To demonstrate this , we use the following laboratory system . Escherichia coli K12 ( AG100 ) is targeted with two antibiotics , erythromycin ( a macrolide , ERY ) and doxycycline ( a tetracycline , DOX ) , that bind to different ribosomal RNA subunits , thereby inhibiting translation . While this is a nonclinical drug pairing , the commercial drug Synercid ( comprising quinupristin and dalfopristin ) also targets ribosomal RNA combinatorially [10] . Moreover , some clinical combinations have ambiguous pharmacological interactions that can appear antagonistic in vitro [11 , 12] , whereas the ERY—DOX pairing has an established synergy [13 , 14] . Before continuing , we need to declare a standard notational device that we will use throughout . It defines how antibiotic efficacy is measured , independently of the drug under study . Thus , ICx will denote the antibiotic concentration that reduces the density of the ancestral bacterial strain ( AG100 ) , rather than ( for example ) any other fitness measure , exponential growth rate , or area under a growth curve , by a factor x% relative to that produced without antibiotic in any single period of bacterial growth . Now , E . coli is known to decrease susceptibility to ERY and DOX by amplifying a genomic region that contains the operon acrRAB because a multidrug pump is formed from the products of acrRAB and tolC [13 , 15] . Selection for amplification mutations occurs even when the drugs are combined at high concentrations whereupon pump duplications and triplications are observed [16] . The triplications permit bacteria subjected to 5 d of combination treatment at twice IC95 dosages , and thus at very low population densities , to eventually restore their growth rates and population densities to almost untreated levels [16] . In these circumstances , the successful clearance of E . coli using sublethal dosages of ERY and DOX appears implausible . Low-dose monotherapies are unlikely to work [17] , and combining the antibiotics into a synergistic IC50 cocktail ( that achieves IC90 overall because of the synergy ) is known to be futile because of resistance increases provided by the pump duplications [13] . We therefore turn to sequential treatments , an approach that has been used to treat cancers [18–20] and some clinical infections [9] . These might also appear predestined to fail; after all , cross drug collateral sensitivities are believed to be the basis of successful sequential treatments [7] , whereas our model system , by contrast , has a scalable multidrug pump at its disposal . Nevertheless , to evaluate the impact of extended antibiotic treatments , we propagated populations of E . coli in 96-well microtitre plates containing liquid medium supplemented with antibiotics based on 12-h cycles , aka seasons , of growth . Thus , two drug treatments per day were administered . At the end of each season , 1% of the spent liquid media , containing biomass , was transferred to a plate containing fresh medium and antibiotics , where growth could resume . The media was supplemented with enough glucose that this protocol would not clear the bacterium in the absence of drug but would instead establish a near-constant , season-by-season total observed population density of about 108 cells per ml in stationary phase ( as can be discerned from Fig S1 and Fig S7 in S1 Text ) . Given this model , we sought antibiotic treatments capable of clearing the bacterium .
By the term sequential treatment , we mean the following protocol: one of the two drugs is used in season 1 , and , whether ERY or DOX , it may be re-used in season 2 , or , alternatively , the other drug may be deployed instead . This process then continues each season until treatment ends . For a treatment of eight seasons , there are 28–2 = 254 possible sequential protocols ( minus the two monotherapies ) . However , seeking to understand whether drug switches per se reduce population growth , only balanced sequential treatments that use four seasons of both drugs were trialled ( Fig S6 in S1 Text , section 1 ) . Seeking evidence of successful low-dose treatments , we first treated E . coli with ERY and DOX for eight seasons at dosages corresponding to the IC50 of each drug , implementing the following treatments: two monotherapies , one 50/50 combination using a half dose of both drugs ( achieving approximately IC90 , Fig S3 and Fig S4 in S1 Text , section 1 ) in addition to 70 sequential treatments ( three replicates each ) . An analogous screen of sequential treatments was then implemented at IC70 dosages ( but only 66 of these sequential treatments were implemented ) . Fig . 1 summarises the IC50 data . In Fig . 1A , the 50/50 combination treatment achieves greater single-season inhibition than each monotherapy , as expected from prior reports of synergy ( p<10-7 , test as indicated in Fig S3 in S1 Text ) . However , by 36 h the combination therapy no longer produces the lowest bacterial densities , and by 96 h it produces high final densities ( Fig . 1A and Fig . 1B ) , higher than the mean of the family of sequential treatments ( p<10-8 , F ( 1 , 69 ) ≈47 . 1 , one-way ANOVA ) . Although a sequential treatment has the lowest final density of all those trialled ( Fig . 1A ) , no IC50 treatment provided any evidence of eliminating the bacteria by 96 h . After increasing dosages to their IC70 values , the following evidence of bacterial clearance by 96 h was observed . Sixteen sequential treatments that produced some of the lowest population densities after 96 h of treatment ( treatments marked with boxes in Fig . 2A ) were examined , and , using spot tests , we could isolate no live cells for five of these treatments in all three replicates . The 11 remaining treatments lead to a zero cell count in some replicates but not in all ( Fig S15 in S1 Text , section 3 ) . We then replicated all 16 treatments an additional three times , and the five previously successful treatments again produced a zero cell count by 96 h , although the remaining 11 treatments showed substantial between-replicate variability in their population dynamics ( Fig S15 ) . By contrast , Fig . 2B shows that the 50/50 combination treatment ( with a greater inhibition than IC70 due to the synergy ) and both monotherapies yielded recovering ( i . e . , increasing ) mean population densities beyond 48 h at these dosages . ( In addition , we recall that twice IC95 combinations of these drugs can fail in this treatment model too [16] . ) However , these observations serve to illustrate that appropriately optimised , sequential therapies at IC70 can clear a bacterium even when synergistic combination treatments with greater one-season inhibition do not . In order to determine genetic changes due to the differential stresses found in drug-free conditions and in the sequential and combination treatments , two treatments at IC50 that produced comparable densities at 96 h were subjected to a whole-genome sequencing analysis and compared to the drug-free populations ( S1 Text , section 4 ) . Writing “E” for a season of ERY and “D” for DOX , when metagenomes from the EDEDEDED and 50/50 combination treatments were sequenced , known resistance mutations were observed in both . Fig . 3B highlights a 412 Kb genomic region containing the acrRAB operon whose duplication was observed more frequently in both the combination and sequential treatments at 96 h ( namely , eight seasons ) than at 24 h ( or two seasons; Fisher exact test for both , p = 0 . 05; Fig S17 in S1 Text , section 4 ) . Treating sequentially does not , therefore , avert selection for duplications of the acrRAB operon . We sought evidence for triplications of acrRAB by asking whether the ratio of coverage depths between amplified and nonamplified genomic regions was above a value of 2 , the latter being the maximum value possible of this statistic if no triplications were present . However , at 96 h , in neither the sequential treatment ( one-sided t test , p≈0 . 12 , T≈1 . 68 , n = 3 ) nor the 50/50 combination treatment ( one-sided t-test , p≈0 . 061 , T≈2 . 60 , n = 3 ) was this value significantly above 2 . Finally , single nucleotide polymorphisms ( SNPs ) were observed in the putative drug transporter gene mdtG ( yceE [21] ) in all conditions ( Table S3 in S1 Text , section 4 ) ; this member of the marA-soxS-rob stress regulon mediates expression of the acrAB-tolC pump [22] . We expected the rate of adaption ( defined as a rate of increase in growth rate [14] ) to correlate positively with dose . Instead , we observed that adaptation can be just as rapid in the absence as in the presence of antibiotics ( Fig . 3A ) , and our culture conditions may explain this . Slow growing cells , like persister phenotypes [23] and small colony variants [24] , are cleared by our protocol , whereas cells that achieve rapid growth above approximately 6 . 6 generations every 12 h can survive . Rapid bacterial growth is associated with physiological changes that include negative DNA supercoiling and multiple DNA replication forks per cell [25] , increased cell size [26] , and heightened ribosomal demand [27] . The latter likely induced a stringent response in the fastest growth conditions ( the absence of drug ) . In these conditions , SNPs associated with fatty acid degradation , lipid peroxidation stress , and sulphur transportation ( tauA ) were observed , the latter at high frequency ( Table S3 in S1 Text , section 4 ) . As tauA is expressed in our growth media only during cysteine limitation [28] , overcoming α-amino acid starvation is a likely mechanism supporting the SNPs detected in all seven 23S ribosomal RNA operons ( rrn ) of E . coli by 96 h in the absence of drugs ( Table S1 in S1 Text , section 4 ) . Although mutations in the same rrn loci were observed at low frequency at 96 h in the slower-growing populations treated sequentially with drugs , none of these operons were mutated in populations treated with the drug combination . We hypothesise , therefore , that the antibiotics have slowed the rise and sweep of adaptive mutations needed for optimal growth in our culture conditions ( Table S1 in S1 Text , section 4 ) . Finally , we found no significant evidence of SNPs within drug targets in any conditions ( S1 Text , section 4 ) . Antibiotic combinations are used to slow drug-resistance adaptation because they enhance their antibacterial effect through inhibitory synergisms [29] and because they reduce the number of potential resistance mutations . Here , consistent with prior studies of ERY—DOX combinations [13] , growth rate adaptation is so rapid when using ERY and DOX in a synergistic IC70 combination that the bacterium is not cleared ( Fig . 1A , Fig . 2B , Fig . 3A ) , and an analogous observation has already been made at double IC95 dosages [16] . Collateral sensitivities on the other hand , in which the prior use of one antibiotic sensitises the bacterium to the use of another , have few recognised mechanisms [30] , but they too have been proposed as a possible basis for successful sequential treatments [7 , 31] because the change of environment hampers adaptation . Promisingly , the rate of adaptation is demonstrably lower here for sequential treatments than for combinations ( Fig . 3A; p<10-4 , F ( 2 , 21 ) ≈16 . 8 , one-way ANOVA with Bonferroni correction ) . However , it has also been suggested that the antibiotic sequences should follow optimised pathways through networks of drug choices that maximally sensitise the bacterium to treatment [7] . However , E . coli AG100 has pumps for both ERY and DOX , and we might therefore expect to observe cross-resistance , not collateral sensitivity , for this drug pair and this bacterium [6] . It therefore appears we do not have enough drugs for drug cycling to work in this model , but it will turn out , in fact , that we do . This is because at least two cross sensitivity properties are observed . The first of these , which was noted recently for doxycycline [6] , we term nonreciprocal collateral sensitivity ( NCS ) , and it is defined as follows . Label two drugs “A” and “B” and choose equivalent dosages for both , meaning ICx for some x , and let D ( T1 , T2 ) denote density of the population when treatment T2 follows treatment T1 . We will also use the notation An and Bn to denote monotherapies with n rounds of treatment . Now , suppose we begin with a clonal population and treat with A for n+1 time units so that D ( An , A ) denotes the population density after the ( n+1 ) -th treatment . Then , in a separate experiment , we treat with A for n time units followed by B for one time unit so that D ( An , B ) denotes the final population density . A nonreciprocal collateral sensitivity between A and B is said to occur when the switch from A to B results in a density decrease so that D ( An , B ) <D ( An , A ) , whereas an analogous switch from B to A results in a density increase , meaning D ( Bn , A ) >D ( Bn , B ) . When satisfied , this definition means A-adapted populations appear sensitised to drug B , whereas B-adapted populations have increased resistance to A . If present , an NCS demonstrates that single-season inhibitory values cannot be used to infer the later inhibitory effect of antibiotics as the treatment proceeds; thus , ICX values and rate of adaptation measures capture very different properties of the bacterium . For example , despite both drugs having equivalent inhibitory effects on a wild-type population after one dose , drug-resistance mutations could sweep more rapidly for one drug than the other , and this could result in an NCS . Nevertheless , if the observed collateral sensitivity is much larger than the cross-resistance within a dataset that indicates the presence of an NCS , appropriately chosen sequential regimens may still be sufficiently potent to eliminate the bacterium . We first sought collateral sensitivities within the entire dataset shown in Fig . 2A but found no significant evidence ( Fig S12 in S1 Text , section 3 ) that a switch from ERY to DOX had a different effect on population density than switching from DOX to ERY . We therefore tested for the presence of an NCS using a simpler “ ( n+1 ) -protocol”: n seasons of culture with one drug , followed by a switch to the other drug for just one season’s duration . This protocol ( Fig . 4 ) shows that when AG100 is treated with ERY for n seasons ( of 24 h duration ) and DOX on the ( n+1 ) -th season , both at IC70 , the continued increase in bacterial density on the last season is consistent with cross-resistance ( see Fig . 4B for p-values ) . However , when treating with DOX for n seasons and then ERY on the ( n+1 ) -th , a density reduction is observed on the last treatment , consistent with a collateral sensitivity ( Fig . 4B ) . This drug pair therefore possesses an NCS: although both inhibit growth of wild-type E . coli equally , they report different levels of inhibition on drug-adapted populations . This observation accords with predictions of the following theoretical model [13]: ddtb1=G ( S , D1 , E1 ) b1−δ ( b1− ( 1+Δ ) b2 ) , ( 1 ) ddtbj=G ( S , Dj , Ej ) bj−δ ( ( 2+Δ ) bj−bj−1− ( 1+Δ ) bj+1 ) , ( 2 ) ddtbN=G ( S , DN , EN ) bN−δ ( ( 1+Δ ) bN−bN−1 ) , ( 3 ) ddtS=−VSK+S∑j=1Nbj , ( 4 ) ddtDext=−dDDext−∑j=1Nbj ( φd ( Dext−Dj ) −vdpjkd+pjDj ) , ( 5 ) ddtDj=−dDDj+bj ( φd ( Dext−Dj ) −vdpjkd+pjDj ) , ( 6 ) ddtEext=−dEEext−∑j=1Nbj ( φe ( Eext−Ej ) −vepjke+pjEj ) , ( 7 ) ddtEj=−dEEj+bj ( φe ( Eext−Ej ) −vepjke+pjEj ) , ( 8 ) where j = 2 , … , N-1 is a parameter that controls the number of efflux genes each cell can express . Equations 1–8 capture the densities , bj , of bacteria with duplications of a gene that exports drugs from within the cell . At time t , S is the concentration of a limiting carbon source , Dext and Eext are extracellular concentrations of each drug , DOX and ERY respectively , Dj and Ej are the intracellular drug concentrations , and drugs degrade at rates dD and dE . The variable pj represents the expected number of efflux pumps expressed by a cell with j-1 efflux genes , for each j≥2 . We also assume that p1 = 0 so that it is possible for cells to encode the pump without expressing it ( meaning genotypes for which j = 1 have the efflux gene but do not express it ) . More generally , the nonzero quantity pj+1 is defined , for j≥1 , by j/ ( 1+γ∙j ) ) , and then pj/ ( ke+pj ) is the probability that a given drug molecule is bound to an efflux pump . This model is a simplification of the competition each transcription unit has for each efflux operon ( the acrRAB promoter ) , whereby a diminishing return is present in the number of pumps expressed as the number of efflux genes increases; the rate of diminishing returns is controlled by γ>0 . Increases and decreases in the number of efflux genes in a cell are assumed to be a Poisson process with parameter δ per cell per hour . Other variables in Equations 1–8 have the following meaning: φe , φd are antibiotic diffusion rates across the cell membrane , νe , vd are maximal drug efflux rates , and ke , kd are half-saturation constants associated with pump-antibiotic binding; V and K are maximal uptake rate and half-saturation constants associated with a Michaelis-Menten uptake model of the limiting carbon source ( S ) ; growth rate G ( S , D , E ) = cVS/ ( ( 1+keE+κdD+κedED ) ( K+S ) ) is proportional to uptake rate via a per-sugar biomass yield constant c , and G ( S , D , E ) is reduced in value synergistically by the drugs ( where κe , κd and κed are parameters that control drug efficacy and strength of synergy ) ; δ is the rate of amplification of the efflux gene , and δ ( 1+Δ ) , a value necessarily greater than δ , is the rate of loss of the gene . N-1 is the maximum number of copies of the efflux gene . We set N = 3 to represent three different cell phenotypes: an unexpressed pump gene ( a wild type ) , a single expressed pump gene , and one additional copy of that gene in which both copies are expressed . Finally , the model is simulated with several seasons so as to mimic the in vitro protocol , with the loss of 99% of all cells implemented at the end of each season . Equations 1–8 were solved numerically using a parameterisation determined from a prior training dataset ( S1 Text , section 5 ) [13] . Although our theory does not capture all aspects of our data , computations show that , like E . coli , the model possesses an NCS ( Fig . 5 ) . The model predicts that a pump asymmetry due to different efflux efficiencies of ERY and DOX produces populations with differential susceptibility to each drug resulting from having different frequencies of drug-susceptible wild-type cells existing in mutation-selection equilibrium with less susceptible mutants ( Fig . 5A ) . Supporting the hypothesis of different efflux efficiencies of ERY and DOX , data from the E . coli acr efflux knockout strain AG100A ( Δacr ) ( Table S2 ) [13] shows that the loss of acrB reduces the IC50 of ERY to approximately 5% of the wild-type AG100 value but reduces it to just 23% in the case of DOX . The model captures others features of the data , particularly that appropriately chosen sequential treatments produce fewer bacteria than the combination , yet some sequential treatments produce more ( Fig S18 in S1 Text , section 5 ) . A second cross sensitivity property of the ERY—DOX was also established , as follows . Having found a mechanism for an NCS with respect to population densities , we hypothesised that the ( n+1 ) -protocol data could exhibit cross sensitivities with respect to other measures of bacterial fitness . To demonstrate this , we fitted the logistic growth model dxdt = R ⋅ x ( 1 − x/K ) to bacterial density time series , where the parameter R is per hour per capita growth rate and K is the population carrying capacity . The resulting data exhibits collateral sensitivities irrespective of the order in which the drugs were exchanged because a reduction of R was observed following a change of drug for every n tested ( from 3 to 6 ) , although not all reductions were significant ( Fig . 6 ) .
Despite the ability of bacteria to adapt to an antibiotic challenge , our laboratory model shows that one can exploit a sensitising property of fluctuating environments to eliminate a bacterium eventually at dosages that only inhibit growth by 70% initially . However , ours is a very simple treatment model inspired by bacterial infections , and we do not wish to overstate its predictive power in relation to the treatment of humans . In particular , the loss of slow-growing cells from our microcosm is not representative of the in vivo conditions in which slow-growing , antibiotic-tolerant phenotypes can be responsible for recalcitrance to treatment [32] . Our model is also limited because it lacks an immune response or any of the environmental complexity found in the human body . The clinical practise of how antibiotics are used to treat bacterial infections begins with the minimal inhibitory concentration ( MIC ) , the minimal drug concentration at which no visible growth of a bacterium is observed after overnight culture in vitro [33 , 34] . After determining the MIC , antibiotics are deployed at high enough dosages so that peak concentrations are achieved in vivo well in excess of this number [35] . Experience has shown that super-MIC dosages are necessary for successful recovery from infection when using combination treatment and monotherapy [33–39] , although there is recent evidence of effective lower-dose antifungal , anti-MRSA ( methicillin-resistant Staphylococcus aureus ) , and antimalarial treatments in vivo [40–42] . However , so few sequential treatments have been trialled in the clinic that there is little accumulated data on or understanding of what dosing or scheduling criteria might be needed when using sequential treatments in vivo . Furthermore , not all IC70 sequential treatments lead to clearance , and some drug sequences produced higher population densities than the equivalent combination treatment ( Fig . 2A ) . The reasons for this are not clear , although both data and the theoretical efflux model do exhibit large between-treatment variation ( Fig . 2A , Fig S18 in S1 Text , section 5 ) . A prior hypothesis states that greater antibiotic heterogeneities in the bacterial environment should diminish the rate of drug-resistance adaptation [43] . This would mean , for example , that bacteria adapt more readily to an EEEEDDDD treatment ( one drug switch ) than to EDEDEDED ( seven switches ) . We therefore sought a relationship between bacterial population densities and the number of drug switches implemented in different sequential treatments but found no evidence of the predicted correlation ( Fig S11 in S1 Text , section 3 ) . It is well known that the use of low dosages can select for resistant strains when they are competed in co-culture with susceptible strains [17] . However , that mid-dose clearance is still possible has a simple explanation in principle: the antibiotic sequences eventually reduce , and then maintain , Malthusian fitness of the evolving population below zero . To examine this in the simplest of theoretical contexts , suppose Bn represents bacterial population density after n rounds of treatment where 1-I , with 0<I<1 , is an expected fraction of cells that are not cleared by the treatment but instead lost because of other effects ( for example , host immunity ) each treatment . Indeed , granulocyte-mediated clearance has been shown to achieve a two-log10 reduction in bacterial load in a 24 h period using a murine model [44] , giving a value of I≈1/100 . At any sublethal dosage , whereby exponential population growth ( at rate r ) occurs between treatments spaced T time units apart ( cf . Fig . 1A ) , it follows for a bacteriostatic antibiotic at a dose of “A units” that Bn+1 = I∙Bne ( r-A ) T . Bacterial clearance is assured when the population decays eventually; this happens when Bn+1/Bn<1 , which is equivalent to the condition A>r- ( log ( I-1 ) ) /T . Note that this value is less than r by an amount that depends on I . If we now define an analogy of the MIC as , say , the IC99 in this toy model , which is the antibiotic dose that reduces bacterial growth by 99% after one treatment , the condition on A to achieve IC99 is e ( r-A ) T<1/100 , or A>r+ ( log 100 ) /T . This value is greater than r and therefore effective therapeutically , but it is not representative of the critical minimal dose needed to clear the bacterium [45] . The absence of visual eradication overnight in vitro should not , according to this argument , itself be used as a rationale to preclude the practical use of an antibiotic drug . Indeed , the antifungal azole drugs are used to treat Candida albicans clinically at dosages that do not always eliminate population growth after overnight culture in vitro [45–47] . In our in vitro study , we needed to keep the rate of adaptation low for the above theoretical rationale to work , and only certain sequential treatments were able to do this ( Fig . 3A and 3B ) . The requirement for low rates of adaptation likely needs the mutations and physiological changes that arise early during treatment , when population sizes are large , to provide no benefit , or even be deleterious later during treatment and so prevent recovery when population size , and mutational supply , is small . There have been clinical successes for one particular sequential treatment: Helicobacter pylori infection has improved eradication rates for a sequential treatment relative to a combination therapy at the same dose [8 , 9] , although geographical variations in successes have been observed and attributed to pathogen strain differences [48] . We hypothesise that the treatment of other clinical pathogens may be possible using sequential antibiotic treatments . We note that low dosing is used to treat some bacterial infections . The ability of antibiotics to act as modulators of gene expression at low doses [49] can be exploited , for example when certain drug classes ( including macrolides ) are used to control the expression of virulence factors in MRSA [50] . Moreover , the use of β-lactam antibiotics as a low-dose adjuvant is a novel strategy in the treatment of recalcitrant MRSA infection [51 , 52] , even though MRSA is resistant to most of these drugs . The β-lactam does not target the cell directly; rather , it enhances the activity of host peptides that are not antimicrobial per se but which modulate the host immune response [53] . However , it is not our intention to advocate for the indiscriminate clinical use of low-dose regimens . Rather , we are claiming that sequential dosing strategies exist for administering antibiotics that are sufficiently potent , and which prevent adaptation enough , to clear a bacterium when the equivalent dose combination treatment fails to do so . That this can be done even though the bacterium has a scaleable multidrug resistance mechanism in its chromosome gives us cause to hypothesise that new ways of optimising antibiotic use in vivo can be found by alternating them as part of treatment .
We used E . coli AG100 ( a gift from Stuart B . Levy ) and M9 minimal media ( 0 . 2% glucose and 0 . 1% casamino acids ) . Stock solutions of DOX and ERY were made from powder stocks ( Sigma-Aldrich ) at 5 mg/ml in water for DOX and 100 mg/ml in ethanol for ERY and stored at -20°C . All subsequent dilutions were made from these stocks and kept at 4°C . A microtitre plate reader measured optical densities every 20 min at 600nm as a proxy for population densities in different environments ( R2>0 . 99 , Fig S1 in S1 Text , section 1 ) . 96-well plates containing 150 μLof liquid per well incubated at 30°C were used to culture bacteria; these were shaken in a linear manner before each measurement was taken . For prolonged exposure to antibiotics , inoculating bacteria were taken from one colony and cultured overnight in M9 minimal media ( 0 . 2% glucose , 0 . 1% casamino acids ) at 30°C in a shaker-incubator . At the end of each season , a 96-pin replicator sampled the liquid volume , which was then transferred to a new plate containing fresh growth medium and antibiotics; the same environment for each replicate population was maintained . Every subsequent transfer was performed using the 96-pin replicator; the volume transferred was approximately 1 . 5μL . OD time series were imported into Matlab R2013b to subtract the background ( blank wells containing only medium ) and generate all other statistics . No claim is made on the basis of optical density data alone that a zero population density had resulted from treatment . Zero densities were determined by observing an OD value below 10-2 units , whereafter the presence of cells was determined by spot tests . Serial dilutions were then used to determine live cell numbers in colony-forming units , if any were detected ( Fig S15 in S1 Text , section 3 ) . Whole-genome sequence data with 18 samples and an annotated draft genome is available from the European Nucleotide Archive ( ENA ) with study accession number PRJEB7832 . This data can be downloaded from http://www . ebi . ac . uk/ena/data/view/PRJEB7832 .
|
So-called “cocktail” treatments are often proposed as a way of enhancing the potency of antibiotics , based on the idea that multiple drugs can synergise when used together as part of a single combined therapy . We investigated whether any other multidrug deployment strategies are as effective as—or perhaps even better than—synergistic antibiotic combinations at reducing bacterial densities . “Collateral sensitivities” between antibiotics are frequently observed; this is when measures taken by a bacterium to counter the presence of one antibiotic sensitise it to the subsequent use of another . Our approach was to see if we could exploit these sensitivities by first deploying one drug , then removing it and instead deploying another , and then repeating this process . This is not an entirely new idea , and there is a precedence for this form of treatment that has been trialled in the clinic for Helicobacter pylori infection . The idea we pursued here is an extension of “sequential treatment”; we investigated whether with two antibiotics and n rounds of treatment , if we search within the set of all possible 2n “sequential treatments”—including the two single-drug monotherapies—there might be treatments within that set that are more effective than the equivalent two-drug cocktail . Using a simple in vitro treatment model , we show that some sequential-in-time antibiotic treatments are successful under conditions that cause the failure of the cocktail treatment when implemented at the equivalent dosage .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Using a Sequential Regimen to Eliminate Bacteria at Sublethal Antibiotic Dosages
|
The common non-steroidal anti-inflammatory drug ibuprofen has been associated with a reduced risk of some age-related pathologies . However , a general pro-longevity role for ibuprofen and its mechanistic basis remains unclear . Here we show that ibuprofen increased the lifespan of Saccharomyces cerevisiae , Caenorhabditis elegans and Drosophila melanogaster , indicative of conserved eukaryotic longevity effects . Studies in yeast indicate that ibuprofen destabilizes the Tat2p permease and inhibits tryptophan uptake . Loss of Tat2p increased replicative lifespan ( RLS ) , but ibuprofen did not increase RLS when Tat2p was stabilized or in an already long-lived strain background impaired for aromatic amino acid uptake . Concomitant with lifespan extension , ibuprofen moderately reduced cell size at birth , leading to a delay in the G1 phase of the cell cycle . Similar changes in cell cycle progression were evident in a large dataset of replicatively long-lived yeast deletion strains . These results point to fundamental cell cycle signatures linked with longevity , implicate aromatic amino acid import in aging and identify a largely safe drug that extends lifespan across different kingdoms of life .
Levels of cellular and organismal dysfunction increase dramatically with old age . Aging is the greatest risk factor for numerous pathologies , including most forms of cancer , stroke , neurodegenerative disorders , heart disease and diabetes [1] . Hence , delaying aging therapeutically promises immense benefits to human health [2] . However , even with short-lived model organisms , the labor and time associated with unbiased screens for compounds that extend lifespan is a major obstacle [3] , [4] . To overcome this drawback , many studies focus on compounds that target pathways already implicated in aging , such as TOR signaling [5] , [6] , AMP kinase [7] , and Sirtuins [8] , [9] . Alternatively , phenotypes associated with aging are used as a proxy in screens for potential anti-aging therapeutics [3] . These phenotypes usually include resistance to various types of stress and mitochondrial degeneration , as well as maintenance of proteostasis and genomic stability [10] . The ultimate goal of all these approaches is to identify drugs that will delay the onset of age-related dysfunction and/or provide novel therapeutics to the diseases of aging [2] . Studies of replicative and chronological lifespan in the budding yeast Saccharomyces cerevisiae have been driving forces in the identification of conserved genetic pathways that extend lifespan [11] , [12] . In this organism , individual cells can be tracked from birth to death [13] , with the number of divisions a cell can undergo defining its replicative lifespan ( RLS ) [14] . The pathways controlling yeast RLS and C . elegans lifespan exhibit significant overlap [15] . Hence , aging studies in yeast and other model systems represent invaluable platforms for the discovery of therapeutics that affect aging and a mechanistic dissection of their mode of action . However , translating leads from aging screens in yeast and other model organisms to drugs that are efficacious and safe in humans represents a significant hurdle [2] . Alternatively , emphasis could be placed on relatively safe compounds that are already used in humans for some indication . One could then ask whether such compounds could extend the lifespan of model organisms . If successful , these drugs would represent excellent candidates for testing in humans for outcomes on healthspan parameters and biomarkers of longevity . They would also serve as invaluable tools to probe conserved longevity pathways , expanding and deepening our understanding of the basic biology of aging . Here we show that ibuprofen , a common and relatively safe non-steroidal anti-inflammatory drug , extends the lifespan of S . cerevisiae , C . elegans and D . melanogaster . We find that ibuprofen extends the replicative lifespan of yeast cells by destabilizing the high-affinity tryptophan transporter . We also show that ibuprofen causes a small size at birth and a moderate delay in initiation of cell division . Mirroring the effects of ibuprofen , we found that most long-lived yeast mutants were also moderately delayed in initiation of cell division , primarily due to a smaller size at birth . These results point to fundamental cellular properties associated with longevity , and identify a relatively safe drug that alters these properties and extends the lifespan of different species .
We decided to focus on ibuprofen for three reasons: First , it is a relatively safe over-the-counter medication . Second , ibuprofen may be associated with reduced risk of some age-related pathologies . Third , ibuprofen has not been reported to target any of the known aging pathways ( e . g . , the TOR or Sirtuin pathways ) , offering the possibility of novel insights into aging mechanisms . Ibuprofen was invented over 50 years ago . It is the prototypical 2-aryl-propionic acid NSAID . Relative to other NSAIDs , ibuprofen is arguably one of the safest [16]–[19] , and is in the World Health Organization's “model list of essential medicines” ( 18th edition , 2013 ) . As other NSAIDs , ibuprofen has analgesic and anti-pyretic indications . However , these indications stem from ibuprofen's well-established role as a cyclooxygenase inhibitor , interfering with prostaglandin biosynthesis [20] . With regard to age-related pathologies , long-term ibuprofen use reduced the risk of Alzheimer [21] and Parkinson [22] , [23] diseases by more than 30% . However , it is unlikely that these beneficial outcomes were solely due to ibuprofen's anti-inflammatory roles because they were not necessarily shared by other NSAIDs . For example , among the NSAIDs examined , ibuprofen showed the most profound reduction in Alzheimer risk ( 40% ) , while others , such as celecoxib , had no effect [21] . Similarly , ibuprofen alone , but not other NSAIDs tested , reduced the risk of Parkinson disease [22] . To our knowledge , despite the vast number of studies dealing with ibuprofen , there are no direct measurements of ibuprofen's effects on the lifespan of organisms . Consequently , we decided to measure the effects of ibuprofen on yeast replicative lifespan . Added at 0 . 2 mM , we found that ibuprofen significantly extended the RLS of the standard BY4742 strain background ( ≈17% , p<0 . 0001 , see Fig . 1A ) . To test if ibuprofen extends the lifespan of organisms other than yeast , we turned to C . elegans for three reasons: First , C . elegans is a well-established metazoan aging model , allowing us to gauge the ability of ibuprofen to extend the lifespan of organisms from different kingdoms of life [24] . Second , as in yeast , in C . elegans we could probe ibuprofen's effects independently of its role as a cyclooxygenase inhibitor because this organism lacks cyclooxygenase enzymes [25] , which are targeted by ibuprofen in mammals [20] . Third , in C . elegans ibuprofen has been shown to suppress a phenotype associated with aging , inhibiting the deposition of amyloid β peptide , a marker for Alzheimer disease [26] . We found that animals exposed continuously to varying doses of ibuprofen ( 0 . 010–0 . 4 mM ) from hatching until death had a longer lifespan ( S1 Table ) . Note that we used UV-killed bacteria in these experiments , so it is unlikely that these effects are due to indirect effects through the action of ibuprofen on bacterial metabolism ( see Materials and Methods ) . The concentration of ibuprofen at which the lifespan extension was maximal was 0 . 1 mM ( Fig . 1B and S1 Table ) . To further test the conservation of the pro-longevity effects of ibuprofen , we asked if the drug could extend the lifespan of D . melanogaster , another aging model system . Although the COX gene is absent in Drosophila , cyclooxygenase-like activity and inflammatory responses are thought to be present [27]–[29] . We found that ibuprofen ( at 0 . 5 µM ) extended the mean and the maximum lifespan of female flies ( Fig . 1D and S2 Table ) . In males , although mean lifespan may also be extended , this was accompanied by a reduction of the maximum lifespan at all doses tested ( Fig . 1C and S2 Table ) . The reasons for the sex-dependent differences in the longevity effects of ibuprofen are not clear . Although the effect in flies is influenced by the sex of the animal , it is nonetheless remarkable that ibuprofen promotes longevity in organisms as divergent as yeast , worms and flies . Overall , these results suggest that ibuprofen extends lifespan across different kingdoms of life . At least in the case of yeast and worms , the pro-longevity function of ibuprofen is through non-cyclooxygenase-related activity . To understand how ibuprofen might extend lifespan , we focused on the yeast system for the remainder of this report . A previous study interrogated systematically single gene deletions that were sensitive to ibuprofen [30] . Seven out of the eight genes encoding enzymes for de novo synthesis of tryptophan were among the 28 gene deletions that sensitized cells specifically to ibuprofen [30] . To account for these observations , we hypothesized that ibuprofen may interfere with import of aromatic amino acids , including tryptophan . In this scenario , mutants lacking the ability to make tryptophan would rely exclusively on mechanisms responsible for importing tryptophan . Such mutants would be sensitive to any drugs that may impair tryptophan import , perhaps explaining their sensitivity to ibuprofen . To test this model , we then measured directly import of tryptophan in cells treated with ibuprofen . Indeed , ibuprofen inhibited the import of [14C]-tryptophan ( Fig . 2A ) . Impaired import of amino acids , including tryptophan , could alter their intracellular pools . Consequently , we measured intracellular levels of amino acids after cells were exposed for 1 hr to 0 . 2 mM ibuprofen ( Fig . 2B ) , the same dose that extended RLS ( see Fig . 1A ) . We found that the levels of several amino acids were moderately affected by ibuprofen , either increasing , or decreasing , compared to untreated cells ( Fig . 2B and S3 Table ) . Ibuprofen lowered the levels of all aromatic amino acids ( Fig . 2B and S3 Table ) . Overall , these results support the notion that exposing cells to a concentration of ibuprofen that extends RLS inhibits import of aromatic amino acids and lowers their intracellular levels . Tat2p and Tat1p are the high- and low-affinity tryptophan permeases , respectively [31] , [32] . Tat1p is also the high affinity tyrosine permease [31] , and may also be involved in the transport of valine and threonine [33] . In addition to tryptophan , Tat2p can transport tyrosine , phenylalanine and , to a lesser extent , alanine and glycine [33] . In agreement with earlier reports [31] , tryptophan uptake was moderately inhibited in cells lacking Tat1p , more so in cells lacking Tat2p , and completely blocked in cells lacking both Tat1p and Tat2p ( S1 Figure ) . To test whether inhibition of aromatic amino acid import is sufficient to extend lifespan , we measured the RLS of tat1Δ and tat2Δ cells ( Fig . 2C ) . Loss of Tat2p extended RLS significantly ( Fig . 2C ) . We next asked if the ability of ibuprofen to extend RLS depends on aromatic amino acid transport . We found that RLS extension by ibuprofen was attenuated in tat2Δ cells and eliminated in cells lacking both Tat1p and Tat2p ( Fig . 2D ) , which cannot import any tryptophan ( S1 Figure and [31] ) . Cyclooxygenase enzymes are not present in yeast [25] . Therefore , ibuprofen must affect yeast cells via unknown off-target mechanisms . Among possible novel mechanisms , our results point to regulation of tryptophan import through Tat2p as a primary conduit by which ibuprofen extends yeast lifespan . One of the earliest discovered outputs of the TOR pathway in yeast involves control of tryptophan import and regulation of Tat2p stability [34] , [35] . When the Tor1p kinase is active , its downstream effector kinase Npr1p is hyperphosphorylated and inactive . However , inhibiting TOR with rapamycin leads to the dephosphorylation and activation of the Npr1 kinase , which triggers the degradation of Tat2p [36] , [37] . Inhibition of TOR activity is a well-characterized , conserved mechanism that delays aging [5] , [6] . Furthermore , tryptophan auxotrophs are more sensitive to rapamycin [34] and to ibuprofen [30] . Interestingly , we found that ibuprofen re-sensitized to rapamycin otherwise rapamycin-resistant mutants in the TOR pathway , such as TOR1-1 , TOR2-1 and npr1Δ , suggesting interactions between ibuprofen and the TOR pathway ( S2 Figure ) . For the above reasons , we examined whether ibuprofen functions through the TOR pathway to inhibit the import of aromatic amino acids and extend RLS . Since the TOR pathway controls Tat2p stability and sorting , we examined Tat2p levels in cells treated with ibuprofen . For protein surveillance experiments , we used strains carrying a single allele of the gene of interest , encoding a C-terminal TAP-tagged version of the otherwise wild type ORF , expressed from its native chromosomal location [38] . As reported previously [36] , inhibition of TOR by rapamycin reduced Tat2p levels ( Fig . 3A ) . Within 30–60 min after exposure to 0 . 2 mM ibuprofen , steady-state levels of Tat2p-TAP were reduced to a degree similar to that upon exposure to rapamycin ( Fig . 3A ) . Next , we asked if the reduction in steady-state Tat2p-TAP levels were attributable to destabilization of the protein . We monitored Tat2p-TAP levels after the cells were exposed to cycloheximide , to block new protein synthesis . From these experiments , we estimated that Tat2p-TAP had a half-life of 28 min ( Fig . 3B , top panels ) , consistent with data from a genome-wide study that evaluated with the same approach the stability of TAP-tagged proteins [39] . However , upon addition of 0 . 2 mM ibuprofen , the half-life of Tat2p-TAP was reduced to about 10 min ( Fig . 3B , bottom panels ) . Next , we measured by qPCR steady-state mRNA levels of TAT2 and TAT1 after exposure to ibuprofen or rapamycin . We did not observe a significant difference in the steady-state levels of these mRNAs upon exposure to ibuprofen ( Fig . 3C , top ) . We conclude that in cells treated with ibuprofen the drop in Tat2p levels is likely the result of destabilization of this permease . Since both rapamycin and ibuprofen destabilize Tat2p , we next asked if ibuprofen affects other TOR-mediated responses . We evaluated known molecular outputs of the TOR pathway after treatment with ibuprofen , in comparison to cells treated with rapamycin . First , we looked at transcriptional outputs ( Fig . 3C ) . Inhibition of TOR by rapamycin is known to trigger expression of mRNAs under the control of the Gln3p and Gcn4p transcription factors [35] . Gcn4p is activated downstream of the Gcn2p kinase . There are also some mRNAs whose transcription is activated in a manner that is Gcn2p-dependent , but Gcn4p-independent [35] , [40] . We confirmed all these responses to rapamycin ( Fig . 3C ) . However , exposure to the ibuprofen dose ( 0 . 2 mM ) that extends RLS did not elicit any of the above TOR-dependent gene expression changes ( Fig . 3C ) . Since rapamycin addition mimics amino acid starvation , triggering increased translation of GCN4 [41] , [42] , we also asked if translation of GCN4 is affected by ibuprofen . For these experiments , we used cells carrying a reporter plasmid with GCN4 upstream regulatory sequences known to mediate translational control of β-galactosidase expression [43] . Although rapamycin increased β-galactosidase expression significantly , the increase in β-galactosidase expression upon ibuprofen addition was minimal ( S3 Figure ) . These results indicate that ibuprofen at doses that increase RLS does not lead to a general amino acid limitation , consistent with our measurements of intracellular amino acid pools ( Fig . 2B and S3 Table ) . We also measured different readouts of TOR-dependent signaling in cells treated with ibuprofen at the dose that extends RLS . A downstream effector of TOR is Npr1p , which is dephosphorylated upon rapamycin addition [37] , [44] . Whereas Npr1p-TAP appeared to migrate as a single species in rapamycin-treated cells , it migrated as a doublet on SDS-PAGE in both untreated and ibuprofen-treated cells ( Fig . 3D ) . These findings are consistent with TOR-dependent regulation of Npr1p [37] , and indicate that ibuprofen likely does not regulate Tat2p stability in an Npr1p-dependent manner . We then examined another TOR effector , the transcription factor Gln3p . Rapamycin triggers the dephosphorylation of Gln3p [35] , leading to a fast-migrating species of Gln3p on SDS-PAGE [44] , [45] . Such a mobility shift was very pronounced in cells treated with rapamycin but not in cells treated with ibuprofen ( Fig . 3D ) . Next , we asked if destabilization of Tat2p is required for ibuprofen's ability to extend RLS . Degradation of Tat2p is mediated by ubiquitination at N-terminal sites [36] . TAT2 alleles encoding Lys→Arg substitutions at five N-terminal Lys residues yield stable Tat2p variants [36] . In contrast to the extended RLS of cells lacking Tat2p ( Fig . 2C ) , cells expressing stabilized Tat2p-5KR from the endogenous chromosomal location had reduced RLS ( Fig . 3E ) . Furthermore , ibuprofen did not extend the RLS of cells expressing stabilized Tat2p-5KR ( Fig . 3E ) . These results suggest that destabilization of Tat2p is required for lifespan extension by ibuprofen . All the results described above indicate that although ibuprofen destabilizes Tat2p , a known target of TOR , it does so without significantly altering additional outputs of the TOR pathway . Consequently , we next tested if ibuprofen extends RLS in the context of TOR pathway mutants . Even though tor1Δ mutants are long-lived [46] , their RLS were increased further by ibuprofen ( Fig . 4 ) . Ibuprofen also extended the RLS of npr1Δ , gln3Δ and gcn4Δ cells , demonstrating that the corresponding gene products are not required for its longevity-promoting effects ( Fig . 4 ) . Taken together , these results suggest that the mechanism of RLS extension we described for ibuprofen converges on the stability of the Tat2p permease , which is also targeted by the TOR pathway ( Fig . 4E ) . Since preservation of a proliferative state is at the core of cellular replicative lifespan , we decided to examine if exposure to ibuprofen alters cell cycle kinetics . We used centrifugal elutriation to obtain highly synchronous , unbudded , early G1 cell populations . We scored these cultures over time microscopically and with a channelyzer , a particle counter that directly measures the volume of cells . Note that in yeast initiation of DNA replication is coupled to the formation of a bud [47] . Thus , one can monitor the timing of initiation of division by phase microscopy . We calculated the specific rate of size increase ( which we call here “growth rate” ) and critical size ( the size at which half of the cells budded ) . We then incorporated measurements of birth size , as we described previously [48] , [49] . Knowing how small the cells are when they are born , how big they have to get before they can divide , and how fast they grow from their birth size to their critical size , determines the absolute length of the G1 phase ( Fig . 5A ) . With this methodology , we examined in synchronous cultures G1 progression upon treatment with varying doses of ibuprofen ( Fig . 5B and S5 Figure ) . At lower doses , ibuprofen caused a dose-dependent delay in G1 progression , mostly due to reduction in birth size ( Fig . 5 and S4 Figure ) . Ibuprofen also significantly compromised growth rate at 0 . 4 mM or higher ( Fig . 5C and S5 Figure ) . At even higher doses , ibuprofen was toxic to yeast cells in this medium ( not shown ) . These results suggest that at the dose that ibuprofen extends RLS , it also reduces cell size at birth and moderately delays the G1 phase of the cell cycle . We next asked whether the cell cycle alterations caused by ibuprofen reflect more general links between cell cycle progression and RLS . To answer this question , we first queried systematic genome-wide datasets that report on these processes . Soon after systematic panels of yeast deletion mutants were generated , with each strain lacking a nonessential gene , they were assayed for their competitive fitness [50] , and for the mean cell size of asynchronously dividing cell populations [51] . Recently , we also measured by flow cytometry the DNA content of these mutants [48] and calculated the birth size of the newborn daughter cells [49] . Together these studies provide phenotypes that are associated with cell cycle progression ( i . e . , fitness , mean cell size , DNA content , and birth size ) for 3 , 979 single-gene deletion mutants . We have also initiated a systematic measurement of the RLS of all these mutants [46] , but this effort is still ongoing . Nonetheless , in the Saccharomyces Genome Database ( http://www . yeastgenome . org/ ) , 137 deletion mutants in the same background have already been studied and classified as having increased RLS , based on published data from our group and others ( see S1 Dataset , sheet “variables” , for a list of these ORFs and the corresponding variables ) . Collectively , these datasets permit a much broader evaluation of possible links between cell cycle progression and RLS . We compared the cell cycle related phenotypes of the 137 long-lived ( LL ) mutants with those of the remaining 3 , 842 not long-lived ( NLL ) strains ( S1 Dataset ) . We found that there was not a highly significant difference in the mean cell size of the two groups ( S6 Figure and S1 Dataset ) . However , there were small but significant differences in the birth size of newborn cells ( S1 Dataset , based on both the parametric Student's t test and the non-parametric Mann-Whitney test ) . Overall , it appears that LL mutants have a smaller birth size and reduced fitness , which likely accounts for the moderate increase in the relative duration of the G1 phase ( S1 Dataset and S6 Figure ) . To better visualize these differences , we plotted in a density scatter plot the birth size values against the corresponding fitness for each mutant , for the LL and NLL groups ( Fig . 6 ) . From these data , we conclude that although LL mutants have a slightly smaller birth size and reduced fitness , these relationships are constrained and not proportional . Compared to NLL mutants , LL mutants were neither the smallest , nor the least fit . To evaluate the extent that each of the above cell cycle parameters could serve as predictors for long RLS , we also performed binary logistic regression analysis ( see S1 Dataset , sheet “Statistics” ) . In the context of all the variables we analyzed , the best predictor for long RLS was daughter birth size ( p = 0 . 000613 ) , followed by fitness ( p = 0 . 026 ) . The phenotypes we examined above were from cells dividing asynchronously . Next , we evaluated synchronous cell populations , to measure parameters that determine the absolute length of the G1 phase of the cell cycle and the timing of initiation of a new round of cell division . For this analysis , we compared a group of 14 LL strains to 13 NLL strains ( see S4 Table ) . The mutants we chose lack ORFs that function in diverse cellular processes , including metabolism , protein synthesis , growth signaling or transcription ( see S4 Table ) . From these experiments with synchronous cultures , we conclude the following: Most LL mutants had a smaller birth size ( p = 0 . 026 , based on a Mann-Whitney test ) , but they appear to occupy a “sweet-spot” since strains with much reduced birth size were not long-lived ( Fig . 7A and S4 Table ) . However , at least for the LL mutants we examined here , the size at which they initiated division was normal ( Fig . 7C and S4 Table ) . Many LL mutants also had a reduced growth rate ( Fig . 7B and S4 Table ) but this difference was not statistically significant ( Mann-Whitney test ) . Overall , these observations support our conclusion from the genome wide datasets of parameters from asynchronous cultures that smaller daughter birth size is a phenotype associated with increased RLS ( Fig . 6 and S1 Dataset ) . Furthermore , the fact that critical size is not significantly altered in the long-lived mutants we evaluated probably explains why the mean cell size of asynchronous cultures is not a good predictor of long RLS . Next , we asked about the efficiency of size control mechanisms in the LL mutants . Plotting the logarithm of birth size against the relative growth in the G1 phase of the cell cycle is a measure of the efficiency of cell size control mechanisms [52] , [53] . In such plots , a slope of zero indicates no size control . From single-cell analysis , wild type budding yeast daughter cells display a negative slope of 0 . 7 [52] , [53] . We applied this methodology to all the synchronous daughter cell populations of the LL strains shown in S4 Table and also obtained a slope of −0 . 7 ( Fig . 7D ) . We conclude that the LL mutants we analyzed displayed cell size control that appeared to be as efficient as that of the wild type strain ( Fig . 7D ) . In the Discussion , we comment on the implications of all these data , in the context of recent models of cellular aging . To follow our other findings , we also interrogated possible connections among tryptophan levels , birth size , Tat2p stability and replicative lifespan . We found that although the effects of exogenous tryptophan on cell size were minimal ( S7A Figure ) , exogenous tryptophan suppressed ibuprofen's pro-longevity effects ( S7B Figure ) . Furthermore , steady-state levels of epitope-tagged Tat2p-TAP were not reduced in the hxk2Δ and rpl20bΔ mutants with small birth size and increased RLS , and appeared instead to be increased by ≈2 fold ( S8 Figure ) . We also noted that there was no disproportionate decrease in tryptophan levels in hxk2Δ and sch9Δ mutants ( S5 Table ) , which have a small birth size , and they are long-lived ( S4 Table ) . Instead , these mutants have lower levels ( ≈20%–60% ) of nearly all amino acids , including tryptophan ( S5 Table ) , resembling growth-limited cells in that regard [54] . However , stabilization of Tat2p increased birth size ( Fig . 8A , B ) and suppressed the long lifespan of mutants with small birth size ( Fig . 8C , D ) . Taken together , these results suggest that a link between small birth size and tryptophan levels is not straightforward . Replicative longevity in yeast is not always accompanied with a small birth size , lower Tat2p abundance and lower tryptophan levels . Nonetheless , stabilization of Tat2p both increases the size of cells at birth , and attenuates the pro-longevity effects upon loss of Hxk2p or Rpl20Bp . Lastly , we asked whether interfering with the expression of aromatic amino acid transporters could increase the lifespan of animals . To this end , we tested whether using RNAi to suppress expression of either one of two putative orthologs of Tat2p could extend the lifespan of C . elegans , resembling the long lifespan of yeast tat2Δ mutants . These worm orthologs are encoded by genes C50D2 . 2 and F23F1 . 6 , and the corresponding gene products are 24% , and 20% , identical to Tat2p , respectively . We found that suppressing expression of C50D2 . 2 significantly increased both the mean and the maximal lifespan of the RNAi-treated animals ( mean lifespan increased ≈15% , p<0 . 0001 ) , compared to the control animals ( S9A Figure ) . Animals treated with RNAi against F23F1 . 6 also lived longer , albeit the effect was less pronounced ( mean lifespan increased ≈5% , p = 0 . 0006 , see S9B Figure ) . These results suggest that targeting amino acid transport mechanisms may have general pro-longevity effects .
Although ibuprofen had not been tested for its effects on lifespan , other common NSAIDs have been examined . Aspirin slightly extended the lifespan of genetically heterogeneous male mice [54] . In the same study , a nitrosylated flurbiprofen analog had no effect on the lifespan of mice [54] . Even against cyclooxygenases , NSAIDs often display different modes of inhibition and specificity against specific isoforms . For example , flurbiprofen causes irreversible inhibition of cyclooxygenase activity while ibuprofen does not [20] . Another NSAID that has been reported to extend the lifespan of C . elegans is celecoxib [55] . Similar to our results with ibuprofen , celecoxib extended lifespan when added from hatching until death [55] . We also noted that the effective pro-longevity concentrations of ibuprofen were much lower in flies than in worms or yeast ( 0 . 5 µM vs . 100–200 µM , respectively; see Fig . 1 ) . The reason for this difference is unclear at present . In healthy humans who took a 600 mg ibuprofen dose up to four times daily , the peak plasma concentration was around 50 µg/ml , corresponding to 240 µM [56] . In another study , a single 400 mg dose of ibuprofen results in a plasma concentration of 8 . 4 µg/ml , or 40 µM [57] . Therefore , the levels of ibuprofen that extend the lifespan of worms and yeast are in the range of ibuprofen levels reached in people taking the drug at typical doses . Overall , our results add to the growing role of NSAIDs , and ibuprofen in particular . These compounds are relatively safe therapeutics that may combat age-related pathologies and extend the lifespan of divergent organisms , from yeast to invertebrates and possibly mammals . We discovered that ibuprofen inhibits tryptophan import , cells lacking the high affinity tryptophan transporter are long-lived , and ibuprofen's longevity effects depend on its ability to destabilize the high affinity tryptophan transporter ( Figs . 2 , 3 ) . How important is tryptophan uptake in meeting the needs of cells for this amino acid ? Metazoans cannot make tryptophan and rely exclusively on tryptophan uptake . Plants and microbes , including yeast , can synthesize tryptophan through the shikimate and chorismate pathways [58] . However , tryptophan is by far the costliest amino acid to synthesize , consuming 78 mol of ATP for 1 mol of tryptophan [58] . Hence , given a choice , it is likely that tryptophan uptake will be preferable to synthesis , even in tryptophan prototrophs . Consistent with a role for tryptophan uptake in aging , low tryptophan diets extend the lifespan of rodents [59] . Perhaps in line with our finding that worms with suppressed expression of the yeast tryptophan permease orthologs live longer ( S9 Figure ) , tryptophan analogs that may inhibit the tryptophan transport system have also been reported to increase the lifespan of flies [60] . Conversely , an increase in degradation of tryptophan through the kynuverine pathway has been associated with accelerated aging in animals ( reviewed in [61] ) . Tryptophan's roles are not limited to protein synthesis . For example , in animals , kynuverine metabolites serve as immune and neuronal modulators [61] , [62] and affect cell viability in tissue culture [63] . However , it is not clear if tryptophan degradation and its involvement in the above pathways are a cause or consequence of aging . Our results expand the role of aromatic amino acids in aging . We demonstrate that interfering with aromatic amino acid uptake , genetically or pharmacologically , can extend lifespan . Furthermore , to our knowledge , the extended lifespan of tat2Δ cells ( Fig . 2 ) is the first report of an amino-acid permease deletion that extends RLS in yeast . Perhaps this is a reflection of the incomplete systematic evaluation of the RLS of permease mutants . Alternatively , it may be due to the low abundance of tryptophan in the cells , coupled to the energetic cost of synthesizing tryptophan . Tryptophan levels are 5–10 times lower than the levels of the next low-abundance amino acids ( tyrosine and methionine , see S3 Table ) . Finally , we would like to note that when ibuprofen was added at the dose that extended RLS , the drop in tryptophan levels was 15–20% ( Fig . 2B and S3 Table ) . Hence , the cells were not severely limited for tryptophan , or any other amino acid ( Fig . 2B and S3 Table ) , consistent with the lack of significant GCN4 de-repression ( S3 Figure ) . As we comment later , perhaps this is yet another manifestation of hormesis , underpinning ibuprofen's effects in RLS . Our results strongly suggest not only that ibuprofen destabilizes the Tat2p permease ( Fig . 3 ) , but also that this is the critical function of ibuprofen in mediating RLS extension since stabilization of Tat2p neutralized the ability of ibuprofen to extend lifespan ( Fig . 3E ) . Stabilization of Tat2p also suppressed replicative lifespan extension by loss of Hxk2p or Rpl20Bp ( Fig . 8 ) . The TOR pathway is known to control the stability of Tat2p . A few years after the discovery of the TOR genes in yeast [64] , it was reported that increased levels of TAT2 conferred resistance to the macrolide FK506 [31] , [34] . Then , TOR activity was shown to inhibit turnover of Tat2p , through Npr1p , a downstream effector kinase of TOR [36] , [37] . Interestingly , however , loss of Npr1p does not stabilize Tat2p , presumably because in the absence of Npr1p other kinases substitute for Npr1p's role [37] . In any case , in light of all this information , it seemed plausible that ibuprofen increased RLS through the TOR pathway . Indeed , ibuprofen conferred rapamycin sensitivity to TOR gain-of-function mutants or cells lacking Npr1p ( S2 Figure ) . Such strains are normally resistant to rapamycin and have increased Tat2p activity . We noted however that at the dose that extended RLS , ibuprofen did not elicit molecular responses consistent with TOR pathway inhibition ( Fig . 3 and S3 Figure ) . Furthermore , ibuprofen extended the RLS of both tor1Δ and npr1Δ strains ( Fig . 4 ) , suggesting that it acts through a mechanism that is at least partially distinct from reduced TOR signaling . It is possible that ibuprofen , through its destabilization of Tat2p and inhibition of aromatic amino acid import , may further sensitize the TOR pathway and exacerbate the longevity effects of TOR pathway inhibition . Alternatively , ibuprofen may be acting in parallel with the TOR pathway , to destabilize Tat2p , inhibit import of aromatic amino acids , and extend RLS ( Fig . 4E ) . As we detail next , our results argue against the hypertrophy model of cellular aging . Before initiating a new round of cell division , newborn daughter cells reach a critical size threshold . Critical size is characteristic of the yeast strain and medium used . Yeast mother cells also increase in size with every successive division [65] , [66] . The hypertrophy model of aging proposed that mother cells reach their replicative potential once they attain a fixed maximal size , beyond which they cannot divide any more [67] , [68] . This terminal and very large cell size may represent the point at which cells have become so large that they cannot sustain functions necessary for cell division , perhaps due to a very low surface to volume ratio or for other reasons . The hypertrophy model makes a clear prediction: For small cells , it would take extra divisions to reach the terminal size , resulting in longer RLS . Conversely , large cells will reach the terminal size after fewer divisions , having a shorter RLS . Furthermore , changes in RLS and cell size ought to be proportional . Specifically , it was proposed that RLS is simply a quotient of the difference between maximal and critical size values , and the increase in cell size per cell division [67] . The hypertrophy model received a significant boost from a recent report of changes in RLS that were indeed strongly proportional to observed changes in size [69] . For example , a plot of cell diameter at birth against RLS displayed a linear fit with a coefficient of determination ( R2 ) equal to 0 . 96 [69] . However , it was pointed out that the above conclusions supporting a role for hypertrophy in RLS were drawn from a small sample of mutants ( <15 ) , of which no more than a handful were long-lived [70] , [71] . Short RLS can be due to many causes , with a significant portion unlinked to the mechanistic events driving RLS [70] , [72] . We based our conclusions on genome-wide datasets , parsed in two groups: mutants with a longer lifespan against all the rest ( S1 Dataset ) . We found that the mean cell size of the two groups was similar ( S1 Dataset and S6 Figure ) . This finding argues against a key prediction of the hypertrophy model: LL mutants would have a small overall cell size , which would in turn enable these cells to divide more times until they reach the terminal size . Yang et al also reported that the birth size of long-lived cells was significantly and proportionally smaller than that of short-lived cells [69] . Here , we also identified significant differences in birth size between LL and NLL mutants , but these differences were slight and they were not linearly proportional ( S1 Dataset , S4 Table and Figs . 6 , 7 ) . There are at least two reasons that may account for these discrepancies: First , the sample sizes were vastly different . Second , birth size was measured with different methodologies . Yang et al used photomicroscopy after micromanipulation to measure cell diameters and then extrapolate to calculate the birth size of those cells [69] . Instead , we relied on channelyzer measurements , which report directly on volume , to obtain the size of the smallest cells in dividing populations [49] . Longer RLS has been associated with reduced fitness [73] and slower growth rate [69] . We also noticed that LL mutants have reduced fitness ( S1 Dataset and Fig . 6 ) and growth rate ( Fig . 7 and S4 Table ) . However , these differences were again modest and constrained . The majority of the smallest and/or slowest growing mutants were not long-lived ( Fig . 6 , and as an example see sfp1Δ cells , S4 Table ) . We would also like to note that none of the 14 LL mutants we analyzed with detailed synchronous cell cycle profiles had significantly altered critical size ( Fig . 7 and S4 Table ) . Based on our data , it seems that LL mutants are born smaller and/or grow slightly slower , but they reach a normal critical size before initiating division . This is consistent with the observation that LL and NLL mutants have similar mean size ( S1 Dataset and S6 Figure ) . In fact , their size control appears intact and indistinguishable from wild type cells ( Fig . 7D ) . Again , these observations do not support the hypertrophy model . The cell cycle patterns we described above argue that in most cases long lifespan is associated with a moderate delay in cell cycle progression early in life . This delay arises from a small birth size and/or slower growth rate . As a result of these changes , LL mutants have a G1 delay ( S4 Table and S6 Figure ) . As was noted previously , these observations are perhaps in line with the antagonistic pleiotropy model [73] . In that scenario , mutations that increase lifespan have opposite effects at different ages . A small delay in cell cycle progression would have adverse effects in young cells , decreasing their rates of division , but be beneficial in older cells , enabling them to divide more times . It is also possible that our results can be explained from the viewpoint of hormesis . In hormetic situations , a treatment at low intensity or dose can be beneficial , but at higher levels , the same treatment is harmful . Hormetic responses to various types of stress are often associated with lifespan extension [74] . Perhaps a moderate delay in G1 progression produces such beneficial effects in lifespan . At the same time , it is not difficult to see why a more severe delay would be detrimental . Hormetic considerations may explain why the cell cycle differences we observed in LL mutants are moderate and constrained , but not severe and proportional . This is illustrated by the effects of ibuprofen: At low doses , ibuprofen causes a moderate cell cycle delay , mimicking the profile of LL mutants ( Fig . 5 ) , and extends lifespan ( Fig . 1A ) . At higher doses , however , ibuprofen delays cell proliferation more severely ( Fig . 5 ) . Regardless of the models invoked , our data suggest that using cell cycle parameters may be a promising and readily scalable approach to identify interventions that extend RLS . In conclusion , the results we report reveal unexpected cellular properties associated with longevity and demonstrate that novel functions of existing safe therapeutics can extend the longevity of organisms from different kingdoms of life .
The strains we used are shown in S6 Table . Single gene deletion mutants in the BY4741 , BY4742 , or BY4743 strains were generated by the Yeast Deletion Project [50] . Double mutants ( e . g . , strain CHY01 ) were constructed from crosses of the corresponding single mutants , each in the background of opposing mating types ( BY4741 and BY4742 ) , followed by tetrad dissection , growth on selective media lacking lysine and methionine , and genotyping by PCR . The 2HA-TAT2 and 2HA-TAT2-5KR plasmids pAS55 and pTB355 , respectively , were gifts from Dr . Michael Hall [36] . To generate strains carrying these alleles integrated in the chromosome , we performed the following: First , with one-step PCR replacement [75] , in strain BY4742 we replaced the TAT2 allele with URA3 , yielding a tat2Δ::URA strain ( CHY02 ) . Then , we PCR-amplified the TAT2 ORF and flanking sequences from plasmids pAS55 and pTB355 , using primers that correspond to upstream ( Forward primer: 5′- CCTTCTGAGTGACGCTTAAACCATCTGCAAGTCTCTTCCGCGGTGATGACGGTGAAAACC-3′ ) and downstream ( Reverse primer: 5′- GACGCGAATTGTTTCACACGGTAGGATAAGAGAAATTGCGGACGTTGTAAAACGACGGCC-3′ ) flanking sequences . These PCR products were then used to transform strain CHY02 , counter-selecting for the loss of the URA3 marker on plates containing 5-Fluoroorotic Acid ( 5-FOA ) . The resulting transformants , CHY03 and CHY04 , were genotyped by PCR to confirm the presence of the 2HA-TAT2 and 2HA-TAT2-5KR alleles , respectively , expressed from the endogenous TAT2 chromosomal location . Note that the epitope tag did not alter the lifespan-related function of Tat2p , because the 2HA-TAT2 strain had the same RLS as the untagged but otherwise identical strain and its lifespan was further extended by ibuprofen ( Fig . 3E ) . With one-step PCR replacement [75] , in strain CHY04 ( carrying the 2HA-TAT2-5KR allele ) , we replaced the HXK2 , or RPL20B , ORFs with URA3 , yielding strains CHY05 , or CHY06 , respectively . Similarly , using strain 202233243 ( carrying the TAT2-TAP allele ) we generated strains CHY07 , CHY08 , CHY09 , lacking HXK2 , RPL20B , or SCH9 , respectively . Unless indicated otherwise , the medium we used in most experiments , including cell cycle and RLS measurements , was YPD ( 1% w/v yeast extract , 2% w/v peptone , 2% w/v dextrose ) . We used the ibuprofen sodium salt ( Sigma , Cat#: I1892 ) dissolved in water to a 0 . 1 M stock solution , from which it was added to autoclaved media as indicated in each case . Rapamycin ( Sigma , Cat#: R0395 ) was dissolved in ethanol to a 1 mg/ml stock , before it was added to autoclaved media as indicated . We have described in detail elsewhere the methodology for elutriation experiments [48] , birth size measurements [49] and RLS assays [13] . The smoothened cell size histograms we show are the splines of the corresponding raw data , generated with the R software package “lattice” . All RLS experiments were carried out on standard YPD plates . Cells from an early logarithmic culture in YPD medium were divided in three . Drugs were added as indicated and the cultures were incubated at 30°C for another 3 hrs . The three culture fractions were treated as indicated ( mock , 0 . 2 mM ibuprofen or 50 ng/ml rapamycin ) , and they were incubated for an additional 3 h at 30°C . Cultures were harvested by centrifugation and washed twice in 10 mM sodium citrate , pH 5 . 5 . The cell pellets were resuspended in ice-cold uptake buffer ( 10 mM sodium citrate ( pH 5 . 5 ) ; 20 mM ( NH4 ) 2SO4; 2% glucose ) . The cell densities were measured and normalized to 5×107 cells/ml . While samples were on ice , they were treated again as before ( mock , ibuprofen at 0 . 2 mM , or rapamycin at 50 ng/ml ) . We then added radiolabelled tryptophan solution ( L-Tryptophan , [side chain-3-14C] , 40–60 mCi/mmol , Moravek Biochemicals , Cat#: MC402 ) . The cell cultures were then incubated at 30°C for the times shown . At the indicated time points , 0 . 1 ml aliquots were transferred to microcentrifuge tubes containing 1 ml ice-cold uptake solution to stop the uptake . The samples were centrifuged for 10 sec and the cell pellets were washed three times ( centrifuging for 10 sec in between ) with ice-cold uptake buffer . Finally , the cell pellets were resuspended in 0 . 1 ml of uptake buffer and transferred to a vial containing 5 ml of scintillation mixture . The retained radioactivity was then quantified by liquid scintillation using a Beckman LS6500 Multipurpose Scintillation Counter . Cells were grown in rich YPD medium until they reached a density of 1–5×106 cells/ml . For drug treatment , the culture was then divided in half and one part was treated with ibuprofen at 0 . 2 mM for 1 hr . After quenching with sodium azide ( at 0 . 1% ) and cycloheximide ( at 50 µg/ml ) , the cells were collected by centrifugation and washed with water ( 1∶20 volume , compared to the original culture volume ) . The cells were collected again by a brief spin ( 10 sec in a microfuge ) , and resuspended in 1∶100 volume of water ( compared to the original culture volume ) . Then , the cell suspension was boiled for 5 min and centrifuged to collect the supernatant , representing the metabolite extract . This extract was then analyzed by standard PTH-derivatization and HPLC analysis [76] at the Texas A&M University Protein Chemistry Facility , to quantify the nmoles of each amino acid present in the extract . These values were normalized for the starting cell density and reported in S3 Table . Protein extracts for immunoblots were made with the NaOH extraction method [77] . The extracts were run on 4–12% gradient SDS-PAGE gels . For detection of proteins of interest on immunoblots we used a rabbit polyclonal anti-Cdc28 antibody ( SantaCruz , Cat#: sc28550 ) at a 1∶500 dilution to detect Cdc28p , a mouse monoclonal anti-PSTAIR antibody ( Abcam , Cat#: ab10345 ) at a 1∶1000 dilution to detect Cdk , and the peroxidase-anti-peroxidase ( PAP ) soluble complex ( Sigma , Cat#: P1291 ) at a 1∶1000 dilution to detect TAP-tagged proteins . Conjugated anti-rabbit and anti-mouse secondary antibodies and chemiluminescence reagents were from Thermo Scientific , and used at the dilutions recommended by the manufacturer . Cells were grown to early logarithmic phase in YPD medium and treated with 0 . 2 mM ibuprofen or 50 ng/ml rapamycin for 0 min , 30 min , 60 min and 90 min , as indicated . Cells were harvested and cell extracts were prepared using the glass bead lysis method [75] . Total RNA was purified using the QIAGEN RNeasy kit ( Cat#: 74106 ) according to the manufacturer's instructions . 1 mg RNA was reverse-transcribed using Bio-Rad's iScript cDNA synthesis kit ( Cat#: 170-8890 ) containing oligo ( dT ) and random hexamer primers . Lysate preparation , RNA purification and reverse transcription were performed on multiple biological samples in parallel . cDNA products were amplified with a Roche LightCycler 480 using SYBRGreen I Master ( Roche , Cat#: 04887352001 ) for detection according to manufacturer's recommendations . Primer sequences are listed in S7 Table . cDNA of 0 min treatment was used as standard for normalization . The p180 reporter plasmid driving expression of β-galactosidase was a gift from Dr . Alan Hinnebusch [43] . The plasmid was transformed into BY4741 cells , and transformants were selected on plates lacking uracil . Overnight cultures were diluted 1∶100 in synthetic complete medium lacking uracil . After incubation for 3 h at 30°C , cell cultures were treated as indicated ( mock , 0 . 2 mM ibuprofen or 50 ng/ml rapamycin ) and incubated for an additional 3-4 h at 30°C . We examined three independent cultures for each treatment . Cells were harvested and cell extracts were prepared in Z buffer ( 60 mM Na2HPO4 . 7H2O , 40 mM NaH2PO4 . H2O , 10 mM KCl , 1 mM MgSO4 . 7H2O , 50 mM β-mercaptoethanol ) with glass bead lysis . Protein concentrations were measured using Bradford assay . Cell extracts were diluted in Z buffer after normalizing for the same amount of protein . We then added 0 . 1 ml of cell extract into 96-well plates . To each well , we then added 20 µl of ο-nitrophenyl-β-D-galactopyranoside ( ONPG , Sigma Cat#: 73660 ) solution ( 4 mg/ml in Z buffer ) . Plates were incubated at 30°C for 40–60 min until the color of the samples became pale yellow . OD450 nm for each sample was measured using a SpectraMax 190 Absorbance Microplate Reader . All the experiments were done as described elsewhere [78] . From the single-cell egg stage , lifespans were monitored on plates seeded with UV-killed bacteria ( strain OP-50 ) . Other than scoring times , during which the animals were moved to room temperature , at all stages of these experiments the animals were kept at 20°C . RNA interference ( RNAi ) was delivered to worms as described previously [79] . RNAi feeding bacteria were kind gifts from Dr . Gordon Lithgow . Control animals were fed bacteria carrying an empty vector ( strain pAD12 ) . Lifespan experiments were performed in the second generation of animals grown on RNAi bacteria . Wild-type strain Canton-S were obtained from the Bloomington Drosophila Stock Center ( Indiana University , Bloomington , Indiana , USA ) and used in lifespan experiments . Flies were kept under standard conditions , at 25°C , in a 12∶12 hour light-dark regime , on an agar/semolina/sugar/yeast medium [80] . 25 pairs of parents with synchronized 24 h egg laying , were used to obtain the experimental flies . The flies were extracted from vials immediately after imago eclosion . 150–200 flies were collected ( approximately 30 adult flies per 50 ml vial ) for each experimental variant . Males and non-virgin females were kept separately . Flies were put in vials with medium contained ibuprofen ( Sigma Cat#: I110 ) at concentration of 0 . 3 , 0 . 5 and 1 µM , and transferred to a fresh medium with ibuprofen twice weekly . Dead flies were scored daily during Drosophila lifetime . The data was used to plot survival curves and to calculate the mean , median , minimum and maximum lifespan and the age of 90% mortality calculated with the open source R software package . In order to estimate the significant statistical differences between experimental and control groups , log-rank tests were used . The significance of differences in maximum lifespan was evaluated using the Wang-Allison test . Following the Wang-Allison test , each animal in each experiment was categorized into one of two groups: either lifespan above the 90th percentile or lifespan below the 90th percentile . A two by two contingency table was used to record data . An ordinary χ2-test was used for independent testing of two groups [81] .
|
Aging is the greatest risk factor for many diseases , which together account for the majority of global deaths and healthcare costs . Here we show that the common drug ibuprofen increases the lifespan of yeast , worms and flies , indicative of conserved longevity effects . In budding yeast , an excellent model of cellular longevity mechanisms , ibuprofen's pro-longevity action is independent of its known anti-inflammatory role . We show that the critical function of ibuprofen in longevity is to inhibit the uptake of aromatic amino acids , by destabilizing the high-affinity tryptophan permease . We further show that ibuprofen alters cell cycle progression . Mirroring the effects of ibuprofen , we found that most yeast long-lived mutants were also similarly affected in cell cycle progression . These findings identify a safe drug that extends the lifespan of divergent organisms and reveal fundamental cellular properties associated with longevity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"cell",
"physiology",
"biochemistry",
"cell",
"biology",
"fungal",
"genetics",
"genetics",
"microbial",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"genetics",
"molecular",
"cell",
"biology"
] |
2014
|
Enhanced Longevity by Ibuprofen, Conserved in Multiple Species, Occurs in Yeast through Inhibition of Tryptophan Import
|
Snakebite envenomation is a neglected condition that constitutes a public health problem in tropical and subtropical countries , including Brazil . Interestingly , some animals are resistant to snake envenomation due to the presence of inhibitory glycoproteins in their serum that target toxic venom components . DM64 is an acidic glycoprotein isolated from Didelphis aurita ( opossum ) serum that has been characterized as an inhibitor of the myotoxicity induced by bothropic toxins bearing phospholipase A2 ( PLA2 ) structures . This antitoxic protein can serve as an excellent starting template for the design of novel therapeutics against snakebite envenomation , particularly venom-induced local tissue damage . Therefore , the aim of this work was to produce a recombinant DM64 ( rDM64 ) in the methylotrophic yeast Pichia pastoris and to compare its biological properties with those of native DM64 . Yeast fermentation in the presence of Pefabloc , a serine protease inhibitor , stimulated cell growth ( ~1 . 5-fold ) , increased the rDM64 production yield approximately 10-fold and significantly reduced the susceptibility of rDM64 to proteolytic degradation . P . pastoris fermentation products were identified by mass spectrometry and Western blotting . The heterologous protein was efficiently purified from the culture medium by affinity chromatography ( with immobilized PLA2 myotoxin ) and/or an ion exchange column . Although both native and recombinant DM64 exhibit different glycosylation patterns , they show very similar electrophoretic mobilities after PNGase F treatment . rDM64 formed a noncovalent complex with myotoxin II ( Lys49-PLA2 ) from Bothrops asper and displayed biological activity that was similar to that of native DM64 , inhibiting the cytotoxicity of myotoxin II by 92% at a 1:1 molar ratio .
Accidents involving venomous snakes are medical emergencies that are often neglected in many tropical and subtropical countries [1] . Several epidemiological studies have tried to estimate the true burden of snakebite envenoming in the world . Overall , they have reported up to 5 million snake bites/envenoming per year , including tens of thousands of deaths and a much larger number of victims that are left with permanent sequelae [2–6] . The number of cases of snakebite envenomation is highest in rural regions and in cities that border on forests . Brazil also has a high level of snakebite accidents , most of which involve four predominant genera; Bothrops is the genus that is held accountable for the highest number of accidents [7] . According to the Brazilian Ministry of Health’s Notifiable Disease Information System ( SINAN ) , 53 , 068 ( provisional ) Bothrops snakebite accidents occurred between 2013 and 2015 in the country [8] . Bothrops venoms contain complex mixtures of toxins that can cause the degradation of vascular basement membrane components and myonecrosis , resulting in local bleeding and tissue damage . In severe cases of envenoming , systemic bleeding , shock , hypotension and/or kidney injury may be observed , leading to high morbidity and mortality [9–11] . For longer than one century , snake envenomation has been treated using antivenoms that are based on horse antibodies . However , this remedy does not prevent local damage caused by some venomous snakes , and the antibodies can induce early or late adverse reactions [12] . The application of biochemical methods to the study of venoms has associated their pathological activities with proteins and peptides . Therefore , the study of venom proteomes ( i . e . , venomes ) is regarded as one of the best approaches for characterizing snake venom compositions [13] . Several studies have identified and determined the relative abundances of certain classes of toxins in different snake venoms . It has also been reported that protein and peptide groups exhibit several kinds of activity on different targets , such as the cardiovascular and nervous system , blood components , and muscular and endothelial tissue [14–18] . Venomics studies of Bothrops have identified two of the most abundant protein groups: metalloproteases and phospholipases A2 ( PLA2s ) . These protein groups are responsible for the most severe local clinical manifestations that are produced by these venoms , such as hemorrhage and muscular and endothelial damage [14 , 19–22] . Phospholipases A2 from snake venom have evolved into potent toxins that exhibit diverse activities such as neurotoxic , myotoxic , anticoagulant , hypotensive , cardiotoxic and edema-inducing [23] . PLA2s belong to one of the five principal groups that catalyze the Ca2+-dependent hydrolysis of the acyl ester at the sn-2 position of glycerophospholipids . Basic myotoxic phospholipases A2 are responsible for tissue degradation and necrosis at the bite site . These myotoxins bind to the plasma membrane of skeletal muscle cells , generating muscle necrosis . Some PLA2s contain a critical substitution at the calcium-binding site ( Asp49 to Lys49 ) that renders them non-catalytic , yet they conserve their myotoxic activity [24–29] . Many studies have sought alternative sources of natural venom inhibitors to complement the action of antivenoms , particularly to help neutralize local tissue damage . Some reports have identified some natural components that inhibit PLA2 myotoxins from several snake venom species [30–33] . Additionally , several studies have investigated myotoxic inhibitors that have been isolated from reptile or mammalian sera [34–36] . DM64 is a myotoxin-specific inhibitor isolated from D . aurita ( opossum ) serum and has been characterized as an acidic glycoprotein of 64 kDa in size . DM64 has been structurally classified as a member of the immunoglobulin supergene family , showing five immunoglobulin-type domains that are similar to α1B-glycoprotein [35 , 37] . DM64 forms a noncovalent soluble complex and efficiently inhibits the myotoxic activity of both inactive Lys49-PLA2 and active Asp49-PLA2 but does not inhibit the catalytic activity of the latter [35] . Currently , many alternative treatments are being developed to supplement antivenom serum treatments and prevent local tissue damage . Several newly discovered natural antiophidic molecules from plant extracts have been reported [30–33] . However , their pharmacological reliability has not yet been demonstrated because these inhibitors have different functional groups that could interact with different molecular targets[30 , 32 , 33] . Myotoxin inhibitors isolated from snake and mammalian sera seem to be more specific , making their recombinant expression a promising strategy for new therapeutic developments [35–38] . DM64 is an efficient myotoxin inhibitor that was isolated from D . aurita serum , and the heterologous expression of this glycoprotein could be a breakthrough in the development of Bothrops envenomation treatment . Thus , the aim of this study was to express rDM64 using Pichia pastoris . The biological activity of the recombinant inhibitor on the myotoxin Lys49-PLA2 from B . asper was analyzed .
Yeast extract , peptone , biotin , dextrose , agar , peroxidase-conjugated anti-rabbit secondary antibody , Pefabloc SC ( 4- ( 2-aminoethyl ) -benzenesulfonyl fluoride ) , sorbitol , EDTA ( ethylenediaminetetraacetic acid ) and DAB ( 3 , 3’-diaminobenzidine ) substrate kit were purchased from Sigma ( Missouri , USA ) . Glycerol and methanol were supplied by VETEC ( Rio de Janeiro , Brazil ) . Pichia pastoris X-33 , pPICZαA , and Zeocin were purchased from Invitrogen ( California , USA ) . Restriction enzymes and PNGase F were furnished by New England Biolabs ( Massachusetts , USA ) . Electroporation cuvettes , TMB ( 3 , 3’ , 5 , 5’-tetramethylbenzidine ) EIA substrate kit and low-range SDS-PAGE standards were obtained from Bio-Rad Laboratories ( California , USA ) . Trypsin was purchased from Promega ( California , USA ) . Pierce Glycoprotein Staining kit , DMEM ( Dubelcco’s Modified Eagle Medium ) , and fetal bovine serum were purchased from Thermo Fisher Scientific ( Massachusetts , USA ) . The DM64 gene sequence [35] was synthesized by Epoch Life Science ( Missouri , USA ) . The gene was cloned without a signal peptide into the expression vector pPICZαA , which contained an alcohol oxidase 1 ( AOX1 ) promoter . The gene sequence was cloned in-frame with the S . cerevisiae α-factor secretion sequence that is present in pPICZαA . A stop codon was inserted before the c-myc epitope and the polyhistidine ( 6x His ) tag . The resulting plasmid ( named pPICZαA-DM64 ) was linearized with SacI ( a site present in the AOX1 promoter ) prior to its transformation into P . pastoris X-33 cells via electroporation . Transformed cell suspensions were diluted to guarantee the growth of the well-separated colonies on the surface of a solid medium . The colonies were grown in YPDS ( 1% yeast extract , 2% peptone , 2% dextrose , 1 M sorbitol , and 2% agar ) medium containing 0 . 1 mg/mL Zeocin at 30°C for 3 to 5 days . A second selection was then performed in liquid YPD medium containing 0 . 1 mg/mL Zeocin at 30°C , 250 rpm for 24 hours . Fourteen clones that grew under these last conditions were submitted to a second screening by using increasing concentrations of antibiotics ( 0 . 2 , 0 . 4 , 0 . 6 , 0 . 8 , and 1 mg/mL Zeocin ) . The cells were grown at 30°C and 900 rpm , for 48 hours . Pichia pastoris with the multicopy expression vector pPICZαA-DM64 was initially grown in BMGY medium , which contains glycerol as a carbon source ( 1% yeast extract , 2% peptone , 1 . 34% YNB , 4 x 10−5% biotin , 100 mM potassium phosphate , pH 6 , and 1% glycerol ) at 30°C and 250 rpm . The culture grew until 50-fold diluted aliquots reached an OD600nm of 0 . 36 , which may have taken up to 24 hours . The culture medium was then changed to BMMY , which contains methanol for the purposes of induction and to be used as a carbon source ( 1% yeast extract , 2% peptone , 1 . 34% YNB , 4 x 10−5% biotin , 100 mM potassium phosphate , pH 6 , and 1% methanol ) . The culture flask was then incubated at 30°C and 250 rpm , for 144 and 264 hours . One percent methanol was added every 24 hours to maintain protein expression . Yeast culture with 0 . 2 mM Pefabloc was added to the BMMY medium every 24 hours . The culture was centrifuged at 5 , 000xg for 20 minutes at 4°C , and the supernatant was collected and stored at -20°C . Aliquots of the culture were analyzed by using 12% SDS-PAGE gels [39] that were stained with silver nitrate [40] . The following low-range SDS-PAGE standards were used: phosphorylase b ( 97 . 4 kDa ) , serum albumin ( 66 . 2 kDa ) , ovalbumin ( 45 kDa ) , carbonic anhydrase ( 31 kDa ) , trypsin inhibitor ( 21 . 5 kDa ) , and lysozyme ( 14 . 4 kDa ) . Molecular mass estimates were calculated using Image Master 2D Elite software ( GE Healthcare , version 4 . 01 ) . The expression medium ( 50 μL ) was alkalized by adding 1 M sodium carbonate buffer ( 10 μL ) , and each aliquot that was taken at a different time after induction ( 0–264 hours ) was incubated at 37°C for 2 hours in a 96-well polystyrene plate . The wells were washed three times with wash buffer ( PBS buffer and 0 . 1% ( v/v ) Tween 20 ) . Then , the wells were blocked with PBS buffer containing 5% ( w/v ) non-fat dry milk and incubated at 37°C for 2 hours . The wells were washed three times with wash buffer containing 5% ( w/v ) non-fat dry milk and then incubated for 1 hour at 37°C with fresh wash buffer containing 5% ( w/v ) non-fat dry milk and 100 μL of polyclonal anti-DM64 antibodies ( 0 . 23 mg/mL ) , prepared as previously described [41] , except for the use of DM64 as antigen , instead of anti-hemorrhagic proteins . After this incubation , the wells were washed three times with wash buffer and then incubated at 37°C for 1 hour with 100 μL of peroxidase-conjugated secondary anti-rabbit antibodies ( 1:30 , 000 ) in wash buffer that contained 5% ( w/v ) non-fat dry milk . Finally , the wells were washed three times with wash buffer and then incubated for 20 minutes at room temperature with 100 μL of the Single Component TMB ( 3 , 3’ , 5 , 5’-tetramethylbenzidine ) EIA Substrate kit . The reaction was stopped by the addition of 1 N H2SO4 , and the absorbance at 450 nm was recorded . Native DM64 ( 0 . 1 , 0 . 2 , 0 . 4 , 0 . 8 and 1 . 6 μg/50 μL ) was used to build the standard curve . Bands in silver-stained SDS-PAGE gels were digested as previously described [42] , with modifications . They were first incubated twice with 100% ( v/v ) acetonitrile for 15 minutes and then dried under vacuum for 15 minutes . The bands were reduced by incubating the samples with 65 mM 1 , 4-dithiothreitol at 56°C for 30 minutes . The reduction buffer was removed , and the bands were washed twice in 100 mM ammonium bicarbonate for 10 minutes , followed by a 5 minute wash in 100% acetonitrile . The bands were dried under vacuum for 15 minutes; 20 ng/μL trypsin was added , and they were then incubated for 45 minutes on ice . Excess trypsin was removed , and 40 mM ammonium bicarbonate was added . Hydrolysis proceeded overnight at 37°C and then for an additional 45 minutes at 56°C . Finally , the digested products were desalted using tip columns that were packed with Poros R2 resin ( Applied Biosystems ) and equilibrated with 0 . 1% ( v/v ) formic acid in water . After washing away nonbound material ( 10 x 20 μL ) by using equilibrium buffer , the peptides were eluted using 0 . 1% ( v/v ) formic acid in 50% ( v/v ) acetonitrile and dried under vacuum . Desalted tryptic peptides were redissolved in 1% ( v/v ) formic acid , and 4 μL of each sample was loaded onto a home-made capillary guard column ( 2 cm x 100 μm i . d . ) that was packed with 5 μm , 200 Å Magic C18 AQ matrix ( Michrom Bioresources , Auburn , CA , USA ) . Peptide fractionation was performed on an analytical column ( 10 cm x 75 μm i . d . ) with a laser pulled tip ( ~5 μm ) that was packed with the same matrix and coupled to an Ultimate 3000 RSLCnano chromatography system ( Thermo Fisher Scientific , Waltham , MA , USA ) . The analysis was conducted using an LTQ Orbitrap XL mass spectrometer ( Thermo Fisher Scientific , Waltham , MA , USA ) with a capillary temperature of 200°C , tube lens voltage of 100 V and a nanoESI source with a spray voltage set to 1 . 9 kV and no sheath gas . Samples were added with water containing 0 . 1% ( v/v ) formic acid into the trap column at 2 μL/min , while the chromatographic separation occurred at 0 . 2 μL/min . The peptides were eluted with a 2–40% ( v/v ) acetonitrile gradient in 0 . 1% ( v/v ) formic acid over 32 minutes , which then ramped to 80% acetonitrile in 4 minutes and was followed by a final washing step at 80% acetonitrile for an additional 2 minutes . The mass spectrometer operated in data-dependent mode , using the following settings: for MS1 , a 300–1700 m/z scan range , 1 x 106 automatic gain control ( AGC ) , 500 ms maximum injection time ( IT ) , centroid mode acquisition and resolution of 60 , 000 ( FWHM at m/z 400 ) . Up to 7 of the most intense precursor ions in each survey scan were selected for CID fragmentation in the LTQ using 35% normalized collision energy ( NCE ) . MS2 analysis was performed with the following parameters: 1 x 104 AGC , 100 ms IT , and centroid mode acquisition . The isolation window was 2 m/z , and only precursor ions with a charge state ≥ 2 were selected for fragmentation , setting the dynamic exclusion to 45 s . The spectrometer was calibrated using a calibration mixture composed of caffeine , peptide MRFA , and Ultramark 1621 , as recommended by the instrument manufacturer . The Pichia pastoris sequence database was obtained from UniProt ( Proteome ID UP000000314 , containing 5 , 073 protein sequences ) . The sequences of DM64 ( UniProt Q8MIS3 , excluding the signal peptide ) , DM43 ( UniProt P82957 ) and common contaminants ( ftp://ftp . thegpm . org/fasta/cRAP ) were also included in the protein database . Peaks Studio software ( version 7 . 5 ) was used for de novo sequencing assisted database search [43] using the following parameters: monoisotopic masses , carbamidomethylation of cysteine ( fixed modification ) , oxidation of methionine ( variable modification ) , semi-tryptic digestion , 10 ppm peptide mass error tolerance , 0 . 6 Da fragment mass tolerance , up to 2 variable modifications per peptide and a maximum of 2 missed cleavages . The data were filtered using the PEAKS decoy fusion approach , and false discovery rates ( FDR ) were set to a maximum of 1% at the peptide level . Only proteins that were identified with at least 2 unique peptides were accepted ( FDR values at the protein level ≤1% ) . Expression products in the supernatant and purified rDM64 were separated by 12% SDS-PAGE gel [39] , and the proteins were transferred to a 0 . 45 μm nitrocellulose membrane ( GE Healthcare Life Science , USA ) using ice-cold transfer buffer ( 25 mM Tris-HCl , pH 8 , 192 mM glycine , and 20% methanol ) . The membrane was blocked in TBS-Tween ( 25 mM Tris-HCl , pH 8 . 0 , 140 mM NaCl , 2 mM KCl , and 0 . 05% Tween 20 ) containing 5% ( w/v ) non-fat dry milk overnight at 4°C . The membrane was then washed three times in TBS-Tween and was incubated for 2 hours at room temperature with diluted [1:100 ( v/v ) ] crude rabbit serum that was raised against native DM64 [41] . The membrane was subsequently washed three times in TBS-Tween and was incubated with 1:5000 ( v/v ) peroxidase-conjugated secondary anti-rabbit antibody . The rDM64 bands were visualized using a DAB substrate kit . The following prestained low-range SDS-PAGE standards were used: phosphorylase B ( 103 kDa ) , bovine serum albumin ( 77 kDa ) , ovalbumin ( 50 kDa ) , carbonic anhydrase ( 34 kDa ) , and soybean trypsin inhibitor ( 28 . 8 kDa ) . The expression medium that contained the recombinant protein was concentrated using Amicon Ultra 15 mL centrifugal filters ( Merck Millipore , USA ) with a cutoff of 10 kDa . The expression medium was exchanged to buffer ( 20 mM Tris-HCl pH 7 . 5 ) , and the sample was injected into a column that had already been equilibrated with the same buffer . This column , a HiTrap NHS ( 7 x 25 mm , GE Healthcare Life Sciences , USA ) activated column that contained myotoxin II ( from B . asper ) that had been immobilized according the manufacturer’s instructions , was connected to an ÄKTA Purifier chromatography system ( GE Healthcare Life Sciences , USA ) . The rDM64 fraction was eluted at a flow rate of 1 mL/min with 0 . 1 M glycine , pH 2 . 7 , and collected over 1 M Tris to neutralize the solution . The fraction was concentrated using Amicon Ultra 4 mL centrifugal filters ( Merck Millipore , USA ) with a cutoff of 10 kDa . The recombinant protein fraction buffer was exchanged to the equilibration buffer of a Mono Q GL column ( 5 x 50 mm , GE Healthcare Life Sciences , USA ) ( 20 mM Tris-HCl , pH 7 . 5 ) . The recombinant protein was eluted during a 20 minute 0–1 M NaCl linear gradient at a flow rate of 0 . 5 mL/min . Protein concentrations in the collected fractions were determined using the bicinchoninic acid assay ( BCA method ) with BSA as standard [44] . Their electrophoretic profiles were analyzed by using SDS-PAGE under reducing conditions [39] . PNGase F was used to cleave between the innermost N-acetylglucosamine ( GlcNAc ) residues and the asparagine residues to which they are linked in this high-mannose-type recombinant glycoprotein . Initially , 3 μg of purified rDM64 was mixed with 0 . 5% SDS and 40 mM DTT , and the solution was incubated at 100°C for 10 minutes . Then , 50 mM sodium phosphate , pH 7 . 5 , 1% NP-40 , and 1 , 000 U of PNGase F were added , and the solution was incubated at 37°C for 1 hour . The reaction was stopped by adding 5 μL of denaturing buffer ( 5X , final concentrations: 0 . 3 M Tris-HCl , pH 6 . 8 , 2% SDS , and 0 . 1 M DTT ) , and the separation of the reaction products was visualized using 12% SDS-PAGE [39] gel with silver staining [40] . Native DM64 was used as a positive control in this experiment . The presence of sugars in rDM64 was determined with the periodic acid-Schiff method using the Pierce Glycoprotein Staining kit . The staining of 5 μg of rDM64 was done following the manufacturer’s instructions . Native DM64 was used as a positive control in this experiment . All samples were analyzed using 12% SDS-PAGE gels [39] . The interaction between myotoxin II ( PLA2-Lys49 ) from B . asper and rDM64 was monitored by using a 12% native PAGE gel [39] . The myotoxin-rDM64 complex ( 2:1 molar ratio ) was incubated in 20 mM Tris-HCl , pH 7 . 5 , and 150 mM NaCl at 37°C for 30 minutes . Native DM64 was used as a positive control in this experiment and the gel was stained with silver nitrate [40] . Murine myoblast cell line C2C12 ( ATCC CRL-1772 ) , which can fuse and differentiate into myotubes , was used . The cells were grown in Dulbecco’s Modified Eagle Medium ( DMEM ) that was supplemented with 44 mM sodium bicarbonate , 19 . 5 mM glucose , 2 μM L-glutamine , 100 U/mL penicillin , 0 . 1 mg/mL streptomycin , and 10% fetal bovine serum ( FBS ) in a humidified atmosphere with 5% CO2 at 37°C . Cells were harvested from near-confluent cell monolayers grown in 25 cm2 bottles . After detaching the cells by using 1500 U/mL trypsin containing 5 . 3 mM EDTA for 2 minutes at 37°C , the resuspended cells were seeded in 96 wells at an approximate initial density of 1 x 104 cells/well in the same medium . Upon reaching near confluence in 3 days , the growth medium was replaced with a differentiation medium ( DMEM supplemented with 1% FBS ) . When multinucleated myotube cells were observed ( after 6 days of culture ) , they were utilized in a cytotoxicity assay [25] . Myotoxin II and recombinant or native DM64 were preincubated ( 1:1 , 2:1 and 4:1 , mol:mol ) for 30 minutes at 37°C in 150 μL of DMEM supplemented with 1% FBS . After aspirating the old medium , the samples were added to the cell cultures that were growing in 96 wells to yield a total volume of 150 μL/well . After three hours of incubation at 37°C with 5% CO2 , 20 μL aliquots of supernatant were collected to determine lactate dehydrogenase activity . Controls for 0% and 100% toxicity consisted of assay medium and 0 . 1% Triton X-100 in assay medium , respectively . The results are presented as the mean ± standard deviation ( n = 3 ) . Statistical analyses were by one-way ANOVA followed by Student–Newman–Keuls post hoc test ( GraphPad Prism 5 . 0 software ) . P-values of 0 . 05 or less were considered significant .
Recombinant glycoprotein production was carried out in a 1 L culture flask using a two-step growth protocol that consisted of a glycerol batch phase and a methanol fed-batch phase . During the glycerol batch phase , P . pastoris was cultivated at 30°C for 24 hours , and the cellular concentration reached approximately 40 g/L . The methanol fed-batch phase was maintained at 30°C , and the induction lasted 264 hours , yielding 288 hours of total cell growth time . Pichia pastoris cells were grown in methanol in the absence ( Fig 1A ) or presence ( Fig 1B ) of 0 . 2 mM Pefabloc , a serine protease inhibitor . During the methanol fed-batch phase , cell growth without Pefabloc was maximum at 72 hours ( 96 hours of total cell growth time ) . From 72 to 264 hours of methanol induction , the cell concentration was constant , around 63 g/L ( Fig 1A ) . On the other hand , in the presence of Pefabloc , the cell growth continued after 72 hours , reaching a biomass yield of 90 g/L at 264 hours ( 288 hours of total cell growth time ) ( Fig 1B ) . The use of a serine protease inhibitor during fermentation was important to not only positively stimulate P . pastoris growth but also improve rDM64 expression yields . After 264 hours of methanol induction , the rDM64 concentration in the fermentation medium increased from 0 . 002 g/L without Pefabloc ( Fig 1A ) to approximately 0 . 02 g/L in the presence of Pefabloc ( Fig 1B ) . Without protease inhibitor , the recombinant protein concentration in the fermentation medium increased until 144 hours of induction ( 168 hours of cell growth time ) were reached and decreased thereafter until 264 hours ( 288 hours of cell growth time ) ( Fig 1A ) . The 65 kDa band in the SDS-PAGE gel that corresponds to rDM64 shows the same expression kinetics ( Fig 2A ) , as does immunoblotting ( Fig 2C ) . Several bands with lower molecular masses whose intensities progressively increase during 264 hours of methanol induction can also be observed by SDS-PAGE ( Fig 2A ) . On the other hand , the amount of rDM64 in the culture medium containing Pefabloc favored a progressive increase ( Fig 1B ) , which corresponds to the behavior of the main 65 kDa band observed by SDS-PAGE during the same time interval; there were also smaller amounts of protein bands with lower relative molecular masses ( Fig 2B ) . The bands indicated in red in Fig 2A were digested in gel , and the resulting peptides were analyzed by mass spectrometry ( S1 Table ) . Database searches using the masses of these peptides and their fragment ions led to the unequivocal identification of the 65 kDa bands ( bands # 3–5 ) as full-length DM64 . The identity of DM64 was further confirmed by Western blotting using polyclonal antibodies , as shown in Fig 2C . Additional fainter bands with higher ( bands # 1–2 ) and lower ( bands # 6–10 ) relative molecular masses were also identified as DM64 . In this last case , partial proteolytic degradation of rDM64 seems the most likely explanation , given the reduced abundance of these bands in the presence of Pefabloc . Higher molecular mass bands ( band #1: 136 kDa and band #2: 88 kDa , Fig 2A ) were also identified as DM64 , and two explanations may be envisaged: either there is a strongly aggregated form of rDM64 and/or the recombinant protein has been expressed with different levels of glycosylation . The culture medium containing the recombinant inhibitor was fractionated by liquid chromatography using an affinity column conjugated with myotoxin II , a Lys49-PLA2 isolated from B . asper venom [45] . DM64 specifically binds to the immobilized myotoxin , and for this reason , this first purification step efficiently removed most impurities ( culture medium proteins ) , allowing for the recovery of a protein fraction that was enriched in rDM64 ( Fig 3A ) . However , the SDS-PAGE profile of the sample under reducing conditions showed that the chromatographic peak corresponding to rDM64 was still heterogeneous , with both higher and lower molecular mass protein bands co-purified with full-length rDM64 ( Fig 3C ) . Native DM64 is an acidic protein with a pI of 4 . 5 [35]; thus , a second chromatography step using an anion-exchange column was necessary to improve protein purification , as shown in Fig 3 ( panels B and C ) . rDM64 was noted to be glycosylated following positive staining with periodic acid-Schiff reagent ( Fig 3D ) and was recognized by polyclonal antibodies raised against native DM64 ( Fig 3E ) . rDM64 was treated with the glycosidase PNGase F to remove N-linked oligosaccharides ( Fig 4 ) . Due to the limited amount of homogeneous protein , a partially purified fraction that was purified by affinity column was used instead of the protein preparation that was obtained by anion-exchange chromatography . Both native DM64 ( 71 kDa ) ( Fig 4 , lane 1 ) and rDM64 ( 65 kDa ) ( Fig 4 , lane 3 ) proteins were submitted to carbohydrate removal . Deglycosylated rDM64 showed a main molecular band at 56 kDa ( Fig 4 , lane 4 ) , which is closer to the molecular mass of native DM64 without the glycan moiety ( 55 kDa ) ( Fig 4 , lane 2 ) . The theoretical average molecular mass of DM64 based on its primary sequence is 53 . 3 kDa ( calculated using http://web . expasy . org/compute_pi ) . Therefore , given the accuracy ( ± 10% ) of molecular mass estimates by SDS-PAGE [46] , the molecular masses of deglycosylated DM64 ( native/recombinant ) proteins are in close agreement with the expected value . The ability of the recombinant inhibitor to bind to myotoxin II ( Lys49-PLA2 ) from B . asper venom was analyzed . Myotoxin II ( mt II ) and rDM64 were mixed at a 2:1 ( mol:mol ) ratio , and complex formation was monitored by electrophoresis under native conditions . Fig 5 ( lane 5 ) shows a band with an electrophoretic mobility that corresponds to the complex . Native DM64 was used as a control for complex formation ( Fig 5 , lane 2 ) . Due to the basic nature of myotoxin-II ( pI 9 . 1 ) [47] , it does not enter the gel under native conditions ( Laemmli buffer system without SDS ) . The band corresponding to the rDM64-mt II complex ( Fig 5 , lane 5 ) shows a different mobility than that of DM64-mt II ( Fig 5 , lane 2 ) due to the difference in the net charges of the proteins; this factor is a major variable that influences electrophoresis mobility on native/non-denaturing gels . rDM64 has a different charge than that of native DM64 due to the absence of sialic acid , which is negatively charged . The inhibition of the cytotoxicity induced by myotoxin II from B . asper on C2C12 myogenic cells by rDM64 was also evaluated . The toxin-inhibitor complex was incubated with myotubes for 3 hours and the rDM64 purified by affinity chromatography showed inhibitory properties , inducing a 92% reduction of the cytotoxic effect of myotoxin II when tested at a 1:1 molar ratio ( myotoxin:rDM64 ) ( Fig 6 ) . When a 2:1 molar ratio was used , the cytotoxicity was inhibited by 65% , whereas a 15% inhibition was observed when a 4-fold molar excess of the toxin was tested . When the inhibitory effects of equivalent concentrations of native and recombinant DM64 were compared , a slight but significant difference was observed at a single molar ratio ( i . e . , 2 myotoxin:1 DM64 ) . These results indicate that both DM64 and rDM64 show similar anticytotoxic effects .
Molecular biology techniques enable the production of recombinant proteins in large amounts; these proteins can then be used in therapeutics and scientific research . Currently , E . coli is the most widely used expression system , although many eukaryotic proteins are not efficiently expressed in this organism . Protein misfolding and the absence of post-translational modifications are important limitations in the use of E . coli . Pichia pastoris could be an eukaryotic organism alternative for the expression of recombinant proteins from mammals . This methylotrophic yeast presents advantages such as the presence of a strong promoter region regulated by methanol ( AOX1 ) , the possibility of allowing disulfide bond formation , the ability to secrete recombinant proteins , and the capacity to perform most post-translational modifications [48–50] . The vector pPICZαA was used for rDM64 expression ( Figs 1 and 2 ) . It is regarded as a good vector because it contains the Sh ble gene from Streptoalloteichus hindustanus , which confers resistance to Zeocin , allowing the best P . pastoris clone to be selected and screened for the inhibitor expression . The recombinant inhibitor was secreted due to the presence of the N-terminal α-factor signal peptide of S . cerevisiae . The vector also encodes a C-terminal c-myc epitope and a polyhistidine ( 6xHis ) tag , enabling the easy detection and purification of recombinant proteins , respectively . However , this extra 21-amino-acid sequence may modify the structure of the recombinant protein , compromising its biological activity . Santos-Filho and co-workers [51] have previously reported using the pPICZαA vector to produce recombinant BaltMIP , a myotoxin inhibitor from B . alternatus serum , adding the c-myc epitope and the His-tag at the C-terminus of rBaltMIP . However , the inhibitory effect of the rBaltMIP was lower than that of the native BaltMIP . For this reason , we decided to maintain the natural structure of the rDM64 , and such a C-terminal peptide was not added . Despite lacking these features , our results show that rDM64 was effectively purified using affinity ( column conjugated with myotoxin II ) and ion exchange chromatographies ( Fig 3 ) . For the unequivocal identification of the recombinant protein , we have used mass spectrometric data instead of just epitope recognition ( S1 Table ) . Fermenting Pichia pastoris generally starts in a shake-flask system before the culture is transferred to a larger volume fermenter . The shake-flask system provides suboptimal conditions due to the lack of data recording and regulatory control systems . Nonetheless , the production of rDM64 in shake flasks was advantageous because it is a low-cost and less complex method . However , proteolytic degradation is a recurrent problem when working in shake-flask cultures . In these experiments , methanol is the carbon source and induces the AOX1 promoter; therefore , recombinant protein induction also creates conditions that trigger an excess of protease production [48 , 52 , 53] . Methanol can also induce cell lysis by oxidative stress and heat-shock responses , eliciting a proteolytic response when the cells are growing exponentially , which results in high cell-density fermentation . In this regard , oxidative stress may also be responsible for recombinant protein degradation because of the increased amounts of reactive oxygen species that are produced during methanol induction . Our mass spectrometry analysis of protein expression in the medium identified proteases and intracellular proteins of yeast in addition to rDM64 . This analysis ( S1 Table ) suggests that yeast lysis occurred during expression in shake flasks . The growth of P . pastoris in shake flasks created stress conditions , such as starvation , heat , pH changes , and/or toxic chemicals . Although expression in shake flasks may have also generated proteolytic products and/or glycoforms in minor quantities , it apparently did not affect rDM64 activity . The biological activity of the partially purified rDM64 fraction was similar to that of native DM64 ( Fig 6 ) . During the expression of rDM64 , we added Pefabloc , a potent and irreversible serine proteinase inhibitor , to the expression medium ( Fig 1B ) . It has lower toxicity , improved solubility in water and higher stability in aqueous solutions than other inhibitors ( e . g . , PMSF and DFP ) . This serine protease inhibitor was used to protect against the proteolytic degradation induced by yeast serine proteases . This molecule may additionally inhibit the production of reactive oxygen species that are generated by NADPH oxidase [54–56] . Therefore , Pefabloc was also used to decrease ROS generation in yeast cultures in shake flasks , thus reducing rDM64 degradation . This report is the first to use a synthetic serine protease inhibitor during recombinant protein expression in a Pichia pastoris culture . However , as shown in Fig 2B , the use of Pefabloc during fermentation did not completely inhibit the proteolytic degradation of rDM64 . Although its recommended working concentration ranges from 0 . 1 to 1 mM , a maximum concentration of 0 . 25 mM should be used when working with cell cultures . The concentration used in the present study ( 0 . 2 mM ) may therefore represent a suboptimal inhibitor concentration considering a ) the increased production of P . pastoris extracellular/cell-bound proteases that may be elicited by methanol induction and b ) the release of intracellular proteolytic enzymes following cell lysis induced by oxidative stress . Nevertheless , Pefabloc favored both the growth of the cell culture and heterologous protein production , indicating that this strategy could be used for low-scale recombinant expression of other proteins . Comparing the molecular masses of native and recombinant DM64 showed that rDM64 had a slightly smaller mass ( Fig 4 ) . Previously , our group reported that the N-glycan moiety of native DM64 is of the complex type , being composed of N-acetylglucosamine , mannose , galactose , and sialic acid [57] . However , the glycosylation of foreign proteins by P . pastoris includes only mannose residues . The N-deglycosylation assay showed that the molecular masses of the recombinant and the native protein were similar ( Fig 4 ) ; hence the smaller molecular mass of the glycosylated recombinant protein is likely to be due to differences in the glycan moieties . It is important to observe that glycoproteins expressed in Pichia pastoris are less frequently hyperglycosylated than those produced in S . cerevisiae , although excessive glycosylation in P . pastoris has also been reported [48] . Glycosylation in the lumen of the endoplasmic reticulum after protein translation is similar in both yeast and mammalian cells , but in yeast Golgi , mannose residues are added and oligomannose units can be α-1 , 6 linked to the α-1 , 3 mannose in the Manα-1 , 3-Manβ-1 , 4-GlcNAc2 inner core . In this way , P . pastoris glycosylation results in diverse structural heterogeneity of the rDM64 , which is attributed to the Golgi glycosylation enzymes [48 , 58 , 59] . In the heterologous expression in this study , the higher molecular mass bands ( 136 and 88 kDa ) were also identified by mass spectrometry as rDM64 ( Fig 2A ) . Periodic acid-Schiff ( Fig 3D ) results suggest that these bands may represent glycoforms that are produced by P . pastoris , although the deglycosylation assay does not seem to be conclusive ( Fig 4 ) . Glycosylation is critical in several biological properties , including structural stability , biophysical characteristics , and resistance to proteolytic attack [60 , 61] . N-glycosylated proteins are abundant in eukaryotic cells , and N-linked glycans all contain a common trimannosyl-chitobiose core with one or more antennae attached to each of the two outer mannose residues [61 , 62] . Leon and co-workers [57] have previously shown that after sialic acid and galactose removal , DM64 was still able to interact with myotoxin II from B . asper . The same behavior was observed for DM43 , an antihemorrhagic protein homologous to DM64 ( 71% sequence identity ) that targets snake venom metalloproteases and does not inhibit myotoxins [63 , 64] . Interestingly , PNGase F-treated DM43 was half as effective as native DM43 in inhibiting the hydrolysis of azocasein by jararhagin [57] . On the other hand , our present results showed that the complex-type N-glycans of native DM64 can be fully replaced by a high-mannose N-glycan structure without impairing the inhibitory activity of rDM64 toward myotoxin II ( Fig 6 ) . Although other reports in the literature have shown that partial deglycosylation does not impair the structural stability of native proteins [65–67] , the presence of carbohydrate moieties improves the solubility of proteins and may also be important during in vivo folding of nascent glycoproteins [61 , 68] . In summary , local myonecrosis caused by Bothrops species is an important public health problem , as it may cause permanent disability of the victims in addition to generating high medical costs due to increased hospitalization times [4 , 26 , 69] . This effect is mainly induced by myotoxic phospholipases A2 and , indirectly , by hemorrhagic metalloproteases . Therefore , the local administration of effective myotoxin inhibitors that are based on the structure of rDM64 may represent a valid alternative to reduce tissue damage at the bite site . The present study demonstrated the successful expression of rDM64 in P . pastoris cultures in small volumes . The maintenance of the native-like structure of the inhibitor was fundamental to preserving its anticytolytic effect , suggesting that rDM64 may also inhibit the in vivo myotoxic effect of myotoxin II . The production of a biologically active myotoxin inhibitor by yeast cells can contribute to the development of a therapeutic alternative for the treatment of envenomation by bothropic snakes . The production of rDM64 can also be exploited by studies aiming to map the structure-function relationship of toxin inhibitors and their molecular targets . This result is very important since , while many works have reported the primary structure of inhibitors isolated from snake and mammalian sera , none of these works have shown the three-dimensional structures of their inhibitors [38 , 63 , 64 , 70] . Our group has made multiple unsuccessful attempts to crystallize inhibitors from animal serum . Although the inhibitor DM43 has been crystallized , only low-quality diffraction patterns have been obtained . The heterogeneity of complex-type glycans tends to impair crystallization , and whether the high-mannose glycan structures inserted in recombinant proteins result in more homogeneous structures must yet be tested . The expression of DM64 in P . pastoris could also assist the development of alternative low-resolution structural analysis techniques , such as cross-linking-mass spectrometry ( XL-MS ) , hydrogen-deuterium exchange-mass spectrometry ( HDX-MS ) , and small-angle X-ray scattering ( SAXS ) , using both full-length heterologous inhibitors and/or selected structural domains . Glycosylation is an important feature that maintains the structure of the inhibitor and its soluble state . Previous attempts to express toxin inhibitors using an Escherichia coli system were not successful , probably due to the absence of post-translational modifications . For future development , new research on rDM64 will be undertaken to increase the production scale for structural and in vivo assays .
|
Snakebite envenomation causes medical emergencies that , depending on the species responsible for the bite , involve different organs and tissues . Envenomation by snakebite is a worldwide problem , and Brazil presents a high incidence of Bothrops bites . Bothrops venoms cause pathological alterations with prominent local effects , such as edema , blistering , hemorrhage , dermonecrosis and myonecrosis , usually followed by poor tissue regeneration and permanent sequelae . Bleeding , coagulopathy , cardiovascular shock and renal failure are typical systemic effects of these venoms . The clinical treatment for snakebite envenoming is intravenous administration of the specific antivenom . However , serotherapy does not efficiently protect against local tissue damage . Additional challenges faced by classical antivenom therapy include the wide antigenic variation of venoms across species and even within the same snake species and the frequent occurrence of adverse reactions that are associated with the administration of immunobiologicals . The development of new effective toxin inhibitors based on the structure of natural antiophidic proteins is an attractive therapeutic alternative . DM64 is a myotoxin inhibitor that was isolated from opossum serum , and its expression as a recombinant protein is paramount to the characterization of its structure-function relationship , an essential step toward the development of alternative strategies to better manage bothropic snakebite envenomations .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"molecular",
"mass",
"pichia",
"pastoris",
"tropical",
"diseases",
"vertebrates",
"silver",
"staining",
"animals",
"toxicology",
"toxic",
"agents",
"fungi",
"reptiles",
"neglected",
"tropical",
"diseases",
"glycoproteins",
"gel",
"electrophoresis",
"snakebite",
"chemical",
"properties",
"physical",
"chemistry",
"venoms",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"electrophoretic",
"techniques",
"electrophoretic",
"staining",
"proteins",
"chemistry",
"recombinant",
"proteins",
"yeast",
"snakes",
"biochemistry",
"squamates",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"amniotes",
"glycobiology",
"organisms"
] |
2017
|
Heterologous expression of the antimyotoxic protein DM64 in Pichia pastoris
|
Two Kato-Katz thick smears ( Kato-Katzs ) from a single stool are currently recommended for diagnosing Schistosoma mansoni infections to map areas for intervention . This ‘gold standard’ has low sensitivity at low infection intensities . The urine point-of-care circulating cathodic antigen test ( POC-CCA ) is potentially more sensitive but how accurately they detect S . mansoni after repeated praziquantel treatments , their suitability for measuring drug efficacy and their correlation with egg counts remain to be fully understood . We compared the accuracies of one to six Kato-Katzs and one POC-CCA for the diagnosis of S . mansoni in primary-school children who have received zero to ten praziquantel treatments . We determined the impact each diagnostic approach may have on monitoring and evaluation ( M&E ) and drug-efficacy findings . In a high S . mansoni endemic area of Uganda , three days of consecutive stool samples were collected from primary school-aged children ( six - 12 years ) at five time-points in year one: baseline , one-week-post- , four-weeks-post- , six-months-post- , and six-months-one-week-post-praziquantel and three time-points in years two and three: pre- , one-week-post- and four-weeks-post-praziquantel-treatment/retreatment ( n = 1065 ) . Two Kato-Katzs were performed on each stool . In parallel , one urine sample was collected and a single POC-CCA evaluated per child at each time-point in year one ( n = 367 ) . At baseline , diagnosis by two Kato-Katzs ( sensitivity = 98 . 6% ) or one POC-CCA ( sensitivity = 91 . 7% , specificity = 75 . 0% ) accurately predicted S . mansoni infections . However , one year later , a minimum of three Kato-Katzs , and two years later , five Kato-Katzs were required for accurate diagnosis ( sensitivity >90% ) and drug-efficacy evaluation . The POC-CCA was as sensitive as six Kato-Katzs four-weeks-post and six-months-post-treatment , if trace readings were classified as positive . Six Kato-Katzs ( two/stool from three stools ) and/or one POC-CCA are required for M&E or drug-efficacy studies . Although unable to measure egg reduction rates , one POC-CCA appears to be more sensitive than six Kato-Katzs at four-weeks-post-praziquantel ( drug efficacy ) and six-months-post-praziquantel ( M&E ) .
Schistosomiasis remains a major public health concern despite praziquantel reaching over 30 million people in endemic areas in 2013 [1] . Goals to eliminate schistosomiasis by 2020 have been articulated by the World Health Organization's ( WHO ) ‘Roadmap for Neglected Tropical Disease ( NTD ) Implementation’ [2] , and the London Declaration of the NTD Coalition [3] . Accurate diagnostic techniques , recently highlighted by Gomes and colleagues [4] , are essential for monitoring and evaluation ( M&E ) of mass drug administration ( MDA ) programs at all stages [5]–[9] , and particularly when considering elimination [2] , [10] , [11] and/or drug-resistance pharmacovigilance [12] , [13] . The WHO recommends two Kato-Katz thick smears ( Kato-Katzs ) from a single stool [14] for Schistosoma mansoni diagnosis to determine prevalence to map areas for control interventions [15] . Kato-Katzs have assumed 100% specificity , but large inter- and intra-specimen variation [16]–[18] and low sensitivity for the detection of low intensity infections have been reported [19]–[21] . In Brazil , where M&E programs use only one Kato-Katz , S . mansoni prevalence has been significantly underestimated in low intensity regions [22] , [23] . This may be associated with overestimated cure rates ( CRs ) and praziquantel efficacy [24] , potentially missing drug resistance . Conversely , using one Kato-Katz can overestimate infection intensities [22] . These differences in sensitivity for prevalence versus infection intensity highlight complex interactions with egg reduction rates ( ERRs ) as true intensities decrease . Inter-day variation of excreted egg numbers post-treatment remains poorly understood . Two Kato-Katzs are commonly used for annual M&E of program impact and/or drug-efficacy studies , without being rigorously tested in these scenarios . Detailed analyses of the sensitivity and specificity of single and multiple Kato-Katzs following praziquantel treatment is urgently required . It is not possible to directly measure S . mansoni adult worm numbers , due to their location in the mesenteric system , with eggs counted from Kato-Katzs used as a proxy for infection intensity . Immunodiagnostics for adult worm circulating cathodic antigens ( CCA ) or circulating anodic antigens ( CAA ) detect current infections and are potentially more sensitive for diagnosis of cases in low transmission areas [25]–[27] . Recent developments using innovative immunomagnetic separation with several target CCAs [28] and novel monoclonal antibody diagnostics for serum CCA [27] show high sensitivity even in low endemic areas . Serum CAA tests are more accurate than urine CAA tests [29] , with advances made on a , not yet commercially available , point-of-care-CAA ( POC-CAA ) [30] . In contrast , urine CCA tests were more accurate than serum CCA tests [29] . Being easier to collect and more socially acceptable than stool or blood , urine POC-CCAs were developed for rapid , non-invasive diagnostics and proposed as alternatives to Kato-Katzs for S . mansoni prevalence mapping [31]–[37] . One urine POC-CCA is as sensitive as two Kato-Katzs for S . mansoni diagnosis for prevalence mapping [35] or two years post-treatment [36] and has now been used to map S . mansoni prevalence in low intensity areas in Uganda [37] . Major POC-CCA limitations are their ability to only semi-quantitatively measure infection intensity [32] , [34] , [38] , inaccuracy for S . haematobium infection detection [39] , [40] , and not measuring soil-transmitted helminth ( STH ) infections . Intensity of infection measures are vital for control program M&E and drug-efficacy evaluation . Therefore knowledge on the ability of POC-CCAs to detect S . mansoni intensity reductions is essential [34]–[36] , [41] . We compared the accuracy of the currently available POC-CCA and one to six Kato-Katzs ( two smears per day from three consecutive stool samples ) for S . mansoni diagnosis , in primary-school children in Mayuge District , Uganda , at baseline and after up to ten praziquantel treatments per child over three years ( STARD Checklist S1 and Figure 1 ) . We evaluated the epidemiological implications of diagnostic methods for control program M&E and praziquantel-efficacy studies . We inform on the number of Kato-Katzs required to accurately detect S . mansoni infections pre- and post-praziquantel treatment and whether POC-CCAs are suitable alternatives . Our detailed longitudinal design enabled novel investigations into individuals' recent versus total praziquantel treatments , aiding biological understanding of differences between Kato-Katz and POC-CCA results post-praziquantel treatment . We predicted that the accuracy of M&E and drug-efficacy findings are limited by Kato-Katz sensitivity at low infection intensities post-treatment . We also predicted that a single POC-CCA would have a higher sensitivity than multiple Kato-Katzs , and be more informative for prevalence monitoring as control programs progress .
Approvals were granted by the Uganda National Council of Science and Technology ( Memorandum of Understanding: sections 1 . 4 , 1 . 5 , 1 . 6 ) and the Imperial College Research Ethics Committee ( EC NO: 03 . 36 . R&D No: 03/SB/033E ) . Verbal assent was given by every child before inclusion into this study and at school committee meetings comprising of parents , teachers , and community leaders before the onset of the study . Written consent for the children to participate in the study was attained from each head teacher . Participation was voluntary and children could withdraw or be withdrawn from the study at any time without obligation . Children were treated with 40 mg/kg praziquantel and 400 mg albendazole ( active against STH infections ) as detailed below . Samples were collected , between 2004 to 2006 , from primary-school children , in a high S . mansoni-endemic area , in Mayuge district , Uganda from three schools on the shores of Lake Victoria: Bugoto Lake View , Bwondha , and Musubi Church of God . Children at Musubi were , to the authors' knowledge , praziquantel-naïve . Children at Bugoto and Bwondha had received 40 mg/kg praziquantel one year previously in 2003 [42] . Inclusion criteria were to have lived in the area since birth and to attend the schools sampled . In 2004 , samples were collected at five time-points: baseline , one-week-post- , four-weeks-post- , six-months-post- and six-months-one-week-post-praziquantel treatment ( Figure 1 ) . In 2005 and 2006 , samples were collected pre- , one-week-post- , and four-weeks-post-praziquantel re/treatment . On the third day of sampling , at baseline , six-months , one-year , and two-years all children were treated with 40 mg/kg praziquantel and 400 mg albendazole ( active against STH infections ) . At one-week post-treatment , children with infections of >100 S . mansoni eggs per gram of stool ( EPG ) were retreated with 40 mg/kg praziquantel . At all other time-points all children with positive diagnoses for S . mansoni or STHs were retreated with 40 mg/kg praziquantel and 400 mg albendazole respectively . Cohort and sample collection are described elsewhere [43] . In brief , 110 children from Bugoto , 110 from Bwondha and 68 from Musubi were recruited in 2004 with an equal sex ratio , aged six to 12 years , without prior knowledge of infection status and/or symptoms of S . mansoni infection . In addition , at one- and two-years , 30 praziquantel-naïve six year old children were recruited at each school and followed up with the original cohorts at the time points described above . This enabled monitoring of the impact of MDA on untreated children entering the school system , assessing diagnostic accuracies for Kato-Katzs and POC-CCA , in praziquantel-naïve and praziquantel-exposed children , as control programs progress . Diagnostic accuracy increases with the number of Kato-Katzs , however , in Brazilian low intensity regions , the additional benefit of more than six Kato-Katzs from repeated stools was negligible [21] , supporting our six Kato-Katzs ‘gold standard’ . Stool samples , marked with unique child IDs , were collected on three consecutive days , between 10:00 and 12:00 hours . Two 41 . 7 mg Kato-Katzs were prepared per stool and read onsite using a compound microscope with natural light source , by highly trained personnel from the Ugandan Vector Control Division , Ministry of Health . S . mansoni , hookworm , Ascaris lumbricoides , and Trichuris trichiura egg counts were recorded . Five percent of slides were reread after the study for S . mansoni , A . lumbricoides , and T . trichiura egg counts for quality control , but no significant differences were observed . One urine sample per child was collected between 10:00 and 12:00 hours on the first day . In the first year , at all five time-points , POC-CCAs ( European Veterinary Laboratory , The Netherlands ) were performed , according to the producer's protocols , by the first author , blind of other test results . Microhematuria was tested for using Hemastix ( Bayer , United Kingdom ) . SPSS version 19 ( SPSS , Inc . , Chicago , IL , United States of America ) was used for all statistical analyses . The double entered data were not normally distributed and could not be normalized by transformation , therefore non-parametric tests were used . Individuals without the full six Kato-Katzs were excluded from the study ( Figure 1 ) . Arithmetic mean infection intensities were categorized as by the WHO ( S . mansoni: light = 1–99 EPG , moderate = 100–399 EPG and high ≥400 EPG; A . lumbricoides: light = 1–4999 EPG; Hookworm: light = 1–1 , 999 EPG; T . trichiura: light = 1–999 EPG ) [15] . Exact confidence intervals ( CIs ) were calculated for prevalence measures and standard errors for EPGs .
Observed cure ( measured at four-weeks-post-praziquantel treatment ) and reinfection ( measured at six-months ) rates depended on sampling method and effort ( Figures 4 and 5 ) . Two Kato-Katzs underestimated S . mansoni reinfection whilst overestimating CRs ( Figure 4A ) ( two Kato-Katzs CR = 81 . 5%; six Kato-Katzs CR = 70 . 4% ) . Cure rates determined with POC-CCA-t− were 47 . 8% and 26 . 1% for POC-CCA-t+ . One-week-post-recent-praziquantel ( from data at both one-week and six-months-one-week ) , prevalence was significantly lower when measured by POC-CCAs than by six Kato-Katzs ( Figure 4B ) ( OR 0 . 33 ( 95% CI: 0 . 19 , 0 . 59 ) . Pre-re/treatment and at four-weeks-post-praziquantel-re/treatment in years zero , one , and two , the number of days of Kato-Katzs did not significantly affect the infection intensities ( Figure 5 ) ( all p>0 . 05 ) . However , at one-week-post-re/treatment ( one-week , six-months-one-week , one-year-one-week and two-years-one-week ) , two Kato-Katzs and to a lesser extent four Kato-Katzs systematically , and significantly , overestimated the mean infection intensity ( Friedman χ2 = 33 . 08 , d . f . = 2 , p<0 . 001 ) . Infection intensity measured by the strength of the POC-CCA bands ( graded from negative , + ( inc . trace ) , ++ and +++ ) did not accurately predict the six Kato-Katzs infection intensity categories ( negative , light , moderate , and heavy ) ( Table S1 ) . However , strong positive correlations were seen between the ordinal POC-CCA band strengths and Kato-Katzs infection intensity categories ( baseline: r = 0 . 402 , p = 0 . 003; one-week: r = 0 . 647 , p<0 . 001; four-weeks: r = 0 . 389 , p = 0 . 001; six-months: r = 0 . 413 , p<0 . 001; six-months-one-week: r = 0 . 424 , p<0 . 001 ) as well as between POC-CCA band strengths and individual arithmetic mean EPG ( Figures 6A to E ) . Pre-treatment , one POC-CCA-t+ , and two Kato-Katzs had sensitivities above 90% , however the POC-CCA-t− and t+ NPVs were extremely low , similar to one Kato-Katz ( Table 3 ) . At one-week-post-praziquantel , one POC-CCA was less sensitive and had lower NPVs than one Kato-Katz , whilst two Kato-Katzs had an NPV of only 55 . 6% . At four-weeks-post-praziquantel POC-CCA-t+ had high sensitivity and NPV ( >80% ) , whilst POC-CCA-t− and one to six Kato-Katzs all had sensitivities and NPVs of <60% . Indeed , two Kato-Katzs only detected a quarter of the infections . At six-months-post-praziquantel POC-CCA-t+ was more sensitive than six Kato-Katzs , however all diagnostics showed NPV values below 15% . Some Bugoto and Bwondha children had been treated once before the study and new praziquantel-naïve cohorts were recruited each year . We therefore also analyzed our data along timelines specific for each individual child's praziquantel exposure ( Tables S2 and S3 ) . Key results were not affected by re-analyzing the data in this manner . Pre-treatment , six-months , one-year and one-year-six-months POC-CCA-t+s showed high sensitivities but low NPVs throughout ( Table S2 ) . One-week-post-recent-praziquantel , three Kato-Katzs were required for >90% sensitivity in general , and four-weeks-post-recent-praziquantel four or five Kato-Katzs were required for >90% sensitivities ( Table S3 ) . An increased Kato-Katz sampling effort was required year on year to achieve sensitivities of >90% ( Table S3 ) , which was not clearly seen in the original M&E timeline ( Figure 2 , Table 2 ) . Praziquantel-naïve children and children one-year-post-praziquantel required three Kato-Katzs for accurate S . mansoni diagnosis , whilst four Kato-Katzs were required at two-years , and five Kato-Katzs at three-years .
At least four Kato-Katzs ( two smears per stool from two stools ) are required for M&E , in the early years of a MDA program in a highly endemic area , increasing to six Kato-Katzs ( two smears per stool from three stools ) by year three . One POC-CCA is a suitable alternative to current prevalence M&E protocols , but they provide no information on STHs and limited intensity data post treatment , therefore we recommend their use for S . mansoni M&E with Kato-Katzs performed in a subset of schools . For drug-efficacy studies , at least six Kato-Katzs ( two smears per stool from three stools ) are required for accurate prevalence assessment four-weeks-post-praziquantel treatment . POC-CCAs may be a promising alternative with low specificity findings potentially due to low Kato-Katzs sensitivity , however further work is required to elucidate POC-CCA's full potential for drug-efficacy studies . Further work on improved ‘gold standards’ is required to elucidate discordant POC-CCA and Kato-Katzs results . Data on multiple Kato-Katzs from a single stool post-treatment would ascertain if accuracies of multiple days of Kato-Katzs or POC-CCAs could be matched , minimizing logistical costs without overestimating M&E success and drug efficacy , whilst retaining vitally important intensity data .
|
Schistosomiasis is a parasitic disease infecting over 200 million people . It remains a major public health concern despite treatment of over 120 million people in sub-Saharan Africa alone . Accurate diagnostic methods are essential for monitoring drug efficacy and long-term control program success . The World Health Organization recommends two Kato-Katz thick smears ( Kato-Katzs ) from a single stool for Schistosoma mansoni diagnosis to map prevalence and areas for control interventions . Although highly specific , Kato-Katzs are thought to be insensitive at low egg counts . The recently refined urine point-of-care circulating cathodic antigen test ( POC-CCA ) has been proposed as a diagnostic alternative for mapping areas for interventions , and potentially for assessing drug efficacy . Over three years we assessed the accuracy of six Kato-Katzs and a single POC-CCA in detecting infections in Ugandan primary-school children at 11 time points with repeated praziquantel treatments . Our results demonstrate that two Kato-Katzs accurately detect S . mansoni infection pre-treatment , but at least three days of two Kato-Katzs per stool or one POC-CCA are required for annual monitoring and treatment evaluation and/or drug-efficacy studies . One POC-CCA may be more sensitive in measuring S . mansoni prevalence than six Kato-Katzs , but its accuracies for rigorous intensity measures are still to be proven .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"helminth",
"infections",
"schistosomiasis",
"plant",
"science",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"medicine",
"epidemiology",
"disease",
"surveillance",
"plant",
"pathology",
"biology",
"and",
"life",
"sciences",
"tropical",
"diseases",
"neglected",
"tropical",
"diseases",
"soil-transmitted",
"helminthiases",
"parasitic",
"diseases",
"epidemiological",
"methods",
"and",
"statistics"
] |
2014
|
Sensitivity and Specificity of Multiple Kato-Katz Thick Smears and a Circulating Cathodic Antigen Test for Schistosoma mansoni Diagnosis Pre- and Post-repeated-Praziquantel Treatment
|
Genome-scale metabolic models provide a powerful means to harness information from genomes to deepen biological insights . With exponentially increasing sequencing capacity , there is an enormous need for automated reconstruction techniques that can provide more accurate models in a short time frame . Current methods for automated metabolic network reconstruction rely on gene and reaction annotations to build draft metabolic networks and algorithms to fill gaps in these networks . However , automated reconstruction is hampered by database inconsistencies , incorrect annotations , and gap filling largely without considering genomic information . Here we develop an approach for applying genomic information to predict alternative functions for genes and estimate their likelihoods from sequence homology . We show that computed likelihood values were significantly higher for annotations found in manually curated metabolic networks than those that were not . We then apply these alternative functional predictions to estimate reaction likelihoods , which are used in a new gap filling approach called likelihood-based gap filling to predict more genomically consistent solutions . To validate the likelihood-based gap filling approach , we applied it to models where essential pathways were removed , finding that likelihood-based gap filling identified more biologically relevant solutions than parsimony-based gap filling approaches . We also demonstrate that models gap filled using likelihood-based gap filling provide greater coverage and genomic consistency with metabolic gene functions compared to parsimony-based approaches . Interestingly , despite these findings , we found that likelihoods did not significantly affect consistency of gap filled models with Biolog and knockout lethality data . This indicates that the phenotype data alone cannot necessarily be used to discriminate between alternative solutions for gap filling and therefore , that the use of other information is necessary to obtain a more accurate network . All described workflows are implemented as part of the DOE Systems Biology Knowledgebase ( KBase ) and are publicly available via API or command-line web interface .
Genome-scale metabolic models ( GEMs ) integrate available information about metabolism to provide a basis for holistic modeling and prediction of metabolic phenotypes [1] . GEMs have been utilized broadly [2] , [3] , [4] , [5] across all three domains of life [6] to accelerate research in such areas as network evolution [7] , [8] , [9] , synthetic biology [10] , [11] , and the discovery of novel drug targets [12] . However , achieving a sufficiently accurate metabolic model to enable high utility currently requires a very time-intensive manual reconstruction process , often taking many months or even years to complete [13] . As the throughput of sequencing technologies continues to increase and as research on microbial populations produces more and more genomes [14] , there is a growing need for methods that automate high-quality metabolic model reconstruction . Since the advent of genome-scale metabolic modeling , protocols [13] , databases [15] , [16] , [17] , algorithms [18] , [19] , [20] and toolboxes [20] , [21] , [22] have been developed to help systematize the lengthy and iterative process of collecting , curating , and integrating large volumes of biochemical knowledge . There have also been previous efforts to fully automate this process , including , notably , the Department of Energy's ModelSEED [20] . Despite these important advances , significant barriers to high-quality automated metabolic reconstructions still persist . Even with human curation , ambiguous or incorrect annotations are still pervasive [23] . Incomplete annotations leave gaps in the metabolic networks that need to be filled in order to make simulation possible [13] , [24] . Inaccurate annotations also give rise to the need to identify and assess the merits of alternative annotations for genes , a process that typically done manually as part of the model curation process [13] . An automated approach to model building that accounts for alternative annotations would help expedite manual curation and ensure that the draft models maximally account for alternatives that can be identified based on available data . Existing algorithms for filling gaps , or dead-end reactions , in metabolic networks broadly fall into approaches based on network topology [19] , [25] , pre-defined pathways [26] , or phenotype data [27] , [28] , [29] . Parsimony-based algorithms such as GapFill identify dead-end reactions in a metabolic network and identify the minimum number of modifications to the network that can be made to activate those reactions [19] . Variations of GapFill have been developed that assign specific penalties based on thermodynamics or database incompleteness [25] . Pathway-based algorithms , such as that implemented in the PathwayTools [26] , automatically complete pre-defined pathways that have sufficient representation in the draft model . Finally , several algorithms use phenotype data to help choose gap filling pathways , including OMNI , which maximizes model consistency with reaction rate data [27] , GrowMatch , which maximizes consistency with experimental growth/no growth results [28] , and MIRAGE , which maximizes the co-occurrence and co-expression of connected reactions [29] . Uniquely among these methods , MIRAGE also automatically identifies gene candidates for optimal gap fill solutions . While existing methods capably activate the necessary reactions to allow growth simulations , they often do so by fitting to phenotype data . Such fitting may result in the inclusion of spurious pathways which cause failures when testing the models against independent datasets [30] . This is epitomized by a recent article that showed that in some cases , pruning these spurious pathways can lead to significant improvements in simulation accuracy [30] . Although genomic evidence may be incorporated after gap filling through human curation of potential solutions [31] , these solutions are unlikely to fully reflect all the available knowledge of the genome . Methods that addresses both the resolution of dead-end metabolites and the identification of gene-reaction pairings for the reactions added to the model during the resolution of the gaps in the reaction network help researchers identify poorly-supported solutions when building models , thus helping to reduce over-fitting . The goal of our work is to improve the quality of automatically generated metabolic reconstructions and models by explicitly incorporating alternative potential gene annotations and their estimated likelihoods into the gap filling process . We have developed a likelihood-based gap filling workflow that ( 1 ) assigns likelihood scores based on sequence homology to multiple annotations per gene and , from these , likelihoods for reactions in a network and ( 2 ) identifies maximum-likelihood pathways for gap filling using a mixed-integer linear programming ( MILP ) formulation . We have also developed a workflow to iteratively identify pathways that activates gene-associated orphaned reactions in a network and assesses the likelihood of these pathways . Critically , the likelihood-based approach makes the gap filling solution genome-specific and provides users with putative gene-protein-reaction relationships and confidence metrics for each result . We show that our likelihood-based approach improves the quantity and quality of new gene annotations compared to the existing gap filling algorithm . The resulting models have comparable accuracy when simulating high-throughput growth phenotype data , when compared with previous parsimony-based gap filling algorithms . The workflow tools are fully integrated within the Department of Energy's System Biology Knowledgebase ( KBase ) , and are publicly available via both a web-based command line interface ( available at http://kbase . us ) and a web service API .
Confidence scores are useful for building models and assessing the quality of the annotations , reactions and pathways therein [13] . We have developed a quantitative likelihood measurement for the evidence that a gene carries a specific annotated function and a technique by which these likelihood estimates can be converted into the likelihood of existence of a reaction in a cell's metabolic network ( see Methods ) . Importantly , we simultaneously compute the likelihoods of multiple annotations for a single gene , which both broadens the space of testable annotation hypotheses in gap filling solutions and helps mitigate possible errors in the most likely annotation . We have also developed a method by which these annotation likelihoods are converted into likelihoods of metabolic reactions . These reaction likelihoods are useful to evaluate confidence in the inclusion of individual parts of a metabolic network . In order to assess the efficacy of the likelihood-based gap filling approach , we implemented four gap filling workflows ( Figure 1 and Text S2 ) . These four workflows use two gap filling algorithms ( parsimony-based versus likelihood-based gap filling ) as applied to two separate gap filling strategies ( targeted versus iterative gap filling ) . The goal of both algorithms is to alter the reaction network by adding new reactions or altering existing reactions . Here , we use the terms parsimony-based and likelihood-based to describe the two gap filling schemes according to their core mode of reaction addition . Both schemes can prioritize changes in reversibility or add special consideration to edge cases such as the addition of transporter reactions . However , in parsimony-based gap filling , the overall goal of parsimony-based gap fill is to make the least number of modifications in order to fill a gap . In general , this means that the shortest reaction path is incorporated into a network . In contrast , likelihood-based gap filling weights genomic evidence and takes into account how likely a reaction is to be included in the network . Likelihood-based gap filling favors reaction paths supported by evidence over paths without any supporting evidence from the genome . In targeted gap filling , which is the most commonly used in the field , gap filling is used to activate one particular reaction in a model such as the biomass reaction [19] . The target reaction is activated either by adding new reactions to the model from a universal reaction database or by changing the reversibility of existing reactions in a model . A successful application of targeted gap filling enables simulations to be performed on the resulting model using an increasingly large suite of constraint-based analysis algorithms [32] , [33] . In the second strategy , which we call iterative gap filling , targeted gap filling is applied iteratively to all inactive reactions in a network , hence maximizing the number of activated reactions in a network . High-priority reactions such as those in central metabolism are activated first ( see Methods ) . One could imagine such an approach would sacrifice specificity for sensitivity ( while targeted gap filling to achieve only biomass production would do the opposite ) . In the iterative gap filling workflows , a post-processing step is also used to reduce the redundancy resulting from attempting to activate every gene-associated reaction , to assess the value of each gap filled pathway in terms of how much of the original annotated network is corrected by the pathway , and optionally , to apply a cutoff to the cost of pathways added to the model ( see Methods and Test S2 and S3 ) . As part of the model curation process , it is necessary to evaluate the quality of each annotation and fix those which are found to be problematic [13] . We have implemented a simple method to estimate annotation likelihoods accounting for two sources of ambiguity: ( 1 ) sequence divergence between query genes and the genes in the reference database , and ( 2 ) inconsistencies in annotation within the reference database ( see Methods ) . We characterized the utility of our reaction likelihoods by comparing the gene-reaction links created using our automated likelihood-based approach to those present in manually curated metabolic networks of Escherichia coli K12 [2] and Bacillus subtilis str . 168 [25] ( note that iJR904 , an older E . coli model than the most recent , was used because the ModelSEED database does not yet link annotations to periplasmic reactions and the gap filling implementation does not yet properly support compartmentalized models ) . We found that highly likely gene-reaction links were significantly enriched in the models compared to less-likely gene-reaction links ( Figure 2 ) indicating that a higher likelihood score reflects higher confidence in the predicted function . We also identified large numbers of high-likelihood gene annotations that are not in the comparison models , which reflect promising candidates for further investigation and possible inclusion in the models . Unlike the parsimony-based gap filling approach , the likelihood-based approach is able to produce different solutions for different organisms , even if the starting network is identical , based on the organisms' genetic content . To demonstrate the utility of this approach in improving model quality , we identified a set of 32 reactions from the iBsu1103 genome-scale metabolic model of B . subtilis [25] that were predicted to be essential for growth and whose existence in the model was supported by literature evidence [34] . This set of reactions represented a gold standard set of reactions that should be incorporated into gap filling solutions if they were missing . We then removed all 32 gold standard reactions from the iBsu1103 model and applied the targeted parsimony-based gap filling and likelihood-based gap filling algorithms to restore biomass production in the knockout model . In order to evaluate the effects of parameterizing each algorithm , we performed a sensitivity analysis on the effects of modifying penalties for adding transporters and for making thermodynamically unfavorable reversibility changes ( see Supplemental Material ) . Larger penalties for these types of changes make it less favorable for the algorithm to pick them over other changes to the network when fixing network gaps . Since the removed set of reactions did not include transporters or reversibility changes , higher penalties for transporters or reversibility changes led to greater accuracy in the returned reactions . The penalty for transporters that maximized the accuracy of the returned pathways was higher for parsimony-based gap filling ( 55 or greater - equivalent to adding about 7 intracellular reactions on average ) than for likelihood-based gap filling ( 25 or greater ) . The penalties for thermodynamically unfavorable reversibility changes that maximized accuracy were also higher for parsimony-based than likelihood-based gap fill ( 40 and 12 , respectively ) . Therefore , likelihood-based gap filling reduced the need to have very large penalties for these categories of network changes in order to obtain accurate solutions . Although both methods had the same number of tuning parameters available , likelihood-based gap filling successfully outperformed the parsimony-based method by replacing a maximum of 31 of the 32 gold-standard reactions . Parsimony-based gap filling only replaced only a maximum of 24 reactions , regardless of the chosen penalties ( Dataset S2 ) . The failures in parsimony-based gap filling were a result of picking shorter pathways to fill certain gaps for which longer pathways are the correct choice . For example , the synthesis of isopentyl diphosphate ( IPDP ) , a primary precursor for lipid synthesis , can occur by one of two routes , the mevalonate pathway and the non-mevalonate pathway [35] . B . subtilis uses the non-mevalonate pathway for IPDP synthesis [36] , [37] . The mevalonate pathway contains fewer reactions than the non-mevalonate pathway , and thus the parsimony-based gap filling approach incorrectly used the mevalonate pathway to restore IPDP production ( Figure 3 ) . However , all of the knocked out reactions in the non-mevalonate pathway had high estimated likelihoods . Hence , likelihood-based gap filling correctly chose this pathway to restore production of IPDP . Given the extremely high magnitude of optimal penalties for parsimony-based gap filling and the more moderate magnitude of optimal penalties for likelihood-based gap filling , the optimal penalties for likelihood-based gap filling were selected as the defaults for the algorithm . These values were used for the remainder of the results in this manuscript . They are also the default parameters in the provided workflow script , though users are able to modify them at will . One important step in curating gap filling solutions is identifying genes in the genome that could be responsible for catalyzing the gap filled reactions and assessing the quality of the genomic evidence behind these assignments [31] . Mapping between genes and reactions allows for a useful connection to genetic manipulations , drug targets , and experimental validation . We have compared the ability to identify genes with likelihood-based and parsimony-based gap filling by using the estimated Gene-Protein-Reaction relationships ( GPR ) from our likelihood computations to assign genes to reactions that are gap filled using each approach . We found that likelihood-based gap filling produced significantly more links between genes and reactions and more gene-associated reactions than post-processing of parsimony-based gap filling results ( i . e . seeking for gene homology after reactions were gap filled ) . This was true for both the targeted gap filling and iterative gap filling approaches , though the effect was greater for iterative gap filling ( p<10−6 , Wilcoxon signed-rank test; Figure 4 ) . By virtue of the optimization formulation , the average quality of annotation hypotheses generated from gap filling was also significantly greater for likelihood-based gap filling , as measured by computing the average likelihood of gene associations added using the likelihood-based approach vs . post-processing parsimony-based gap filling solutions ( p<0 . 01 , Wilcoxon signed-rank test; Figure 5 ) . Therefore , the likelihood-based approach yields a larger quantity of better-supported candidate annotations for genes than simply searching for genes associated with parsimony-based gap filling results post-hoc . One commonly-used method to verify the integrity of genome-scale metabolic models is to compare their predictions with high-throughput phenotyping data , such as knockout lethality screens [13] . To test the impact of each of our workflows on the accuracy of model phenotype predictions , we applied our workflows to construct and fill gaps in genome-scale models for 22 organisms for which either Biolog or gene knockout lethality data was available . We then compared the predictions of these models to the phenotype data , without fitting to the data ( Table 1 ) . There were many differences in the pathways identified using likelihood-based gap filling compared to parsimony-based gap filling: between 5% and 30% of the reactions in a likelihood-based gap filling solution were not found in the parsimony-based solution , despite using the same parameters for each ( see Text S1 ) . However , the use of likelihoods did not significantly affect the phenotype predictions . For Biolog data , iterative gap filling increased the aggregate sensitivity by 11% compared to targeted gap fill , but decreased aggregate specificity ( more false positives ) by 10%–13% . The aggregate accuracy decreased by about 1% for iterative compared to targeted gap filling . Since gap filling only adds a small number of genes to the model compared to the number in the draft model ( about a 6% increase for iterative gap filling ) , the sensitivity and specificity of knockout lethality predictions were very similar for all four workflows . The aggregate sensitivity was 84–86% for all four workflows while aggregate specificity was 64–68% . We also examined the lethality predictions specifically for genes added in gap filling ( Figure 6 ) . The negative predictive value was essentially identical for all four workflows at 40% . However , there was a notable improvement in the positive predictive value in the iterative gap filling workflows ( 80% ) compared to targeted gap filling workflows ( 35% for parsimony-based and 55% for likelihood-based gap filling ) . Taken together , these results indicate that iterative gap filling mostly adds genes predicted to be nonlethal knockouts , and that most of these predictions are correct .
An important feature of the likelihood-based gap filling algorithm is that it can differentiate between genomes by assigning organism-specific likelihoods for each reaction in a network . As a direct result , the gap filling solutions resulting from this algorithm are also organism-specific . This direct link back to evidence in the genome directly enables the identification of pathways that are not parsimonious , but that are most consistent with genomic data . We have shown that the likelihood-based approach increases both the quality and the quantity of hypothesized gene associations from gap filling , especially when using the iterative approach to maximize the number of activated reactions subject to evidence constraints . Of course , when building a high-quality network model , gap filled pathways should be evaluated by experts to evaluate the evidence cited in the algorithm , to search for existing experimental evidence in favor of or refuting the suggested solutions , or to design new experiments to test the existence of the hypothesized functions in the modeled organism [13] , [21] . The reported confidence metrics for annotations and for reactions will help curators target these curation efforts . The likelihood-based gap filling methods described in this manuscript use genomic evidence-based metrics for the confidence that can be placed in annotations and reactions . The pathways that result from maximizing confidence have a greater genomic coverage and stronger evidence for inclusion of genes compared with the common procedure of post-processing parsimony-based solutions . Therefore , including likelihoods into the gap filling procedure directly improves on the state of the art in evaluating and selecting gap filling solutions . Despite the significantly increased level of evidence for gap filling solutions resulting from likelihood-based gap filling , we did not observe a significant difference from existing approaches in the accuracy of knockout lethality or growth ( Biolog ) predictions . This result suggests that using phenotype data to filter gap-filling solutions may not result in a more accurate metabolic network ( that is , one that better reflects biological evidence for the specific components included ) . Indeed , validation metrics such as consistency with knockout lethality predictions have a large number of ways in which they could be fit to become consistent with phenotype data , which can lead to decreases in observed accuracy when the model is tested on new data not available during its construction [30] . This tendency to overfit models makes the use of reaction confidence metrics essential when evaluating discrepancies between models and phenotype data . The proposed integration of likelihoods into gap filling can serve as a tool for hypothesis generation in biology . An initial pool of potential annotations with associated likelihoods can be generated using many different methods such as protein co-localization or co-occurrence [38] , [39] or from high-throughput ‘-omics’ datasets such as metabolomics . This initial pool can be quite broad ( many alternative functions for each gene ) . The likelihood-based gap filling approach we have outlined is sufficiently general to incorporate likelihoods based on any type of evidence . The gap filling algorithm then selects from this broad pool of hypotheses for the new annotations that best explain a complex combination of biological observations , providing insights into enzyme promiscuity , adaptation , and evolution . Finally , the KBase framework in which all of our workflows are implemented helps address the need for a unified framework for systems biology . In addition to gap filling , KBase includes implementations of many other modeling and reconstruction tools such as tools for the automatic generation of compartmentalized community models [40] and phenotype reconciliation tools such as the gap generation algorithm implemented as part of the ModelSEED framework [25] . We anticipate that the use of likelihoods to guide solutions of these algorithms would also lead to better-supported networks and improvements in the ability to assess solution quality independently of the test data itself , thus reducing the impact of overfitting . As systems biology expands to incorporate a greater number of high-throughput biological measures , the utility of computational frameworks for leveraging this vast knowledge in toto becomes increasingly important .
The first step of each workflow ( Figure 1 ) is importing an annotated genome into a workspace in the KBase system . KBase workspaces provide a way for users to store , share , and manage data objects that they have uploaded or generated by running KBase analyses . The genome data is imported into the workspace as a Genome typed object , which is a standardized format compatible with all KBase tools that expect genomes as input . All results from subsequent analyses are also stored in a KBase workspace as typed objects ( of different types ) , and the methods used to generate them are tracked to ensure data provenance . The annotations for genes in the genome are used to generate a draft model using the ModelSEED algorithm as previously described [20] . A description is available in Text S1 . The computation of reaction likelihoods for likelihood-based gap filling is achieved by first estimating the likelihood of multiple annotations for each gene in the query organism based on sequence similarity , and then by using mappings from annotations to reactions found in the ModelSEED reaction database [20] to convert these likelihoods into reaction likelihoods . Annotation likelihoods are computed in reference to a database of genes with high-confidence annotations . For this purpose , we compiled a list of the protein sequences for all proteins whose function was either literature-supported or called as part of at least one SEED subsystem [41] . Functional annotations in the SEED subsystems are manually curated using multiple sources of information such as sequence similarity , phylogeny and gene context , and therefore represent a high-confidence reference set . To minimize the amount of redundancy in the list of target proteins , they were binned into organism taxonomic units ( OTUs ) with roughly 97% 16S rRNA similarity [42] . The final target database included at most one protein from each OTU for each functional role . When possible , the representative protein was chosen from the representative organism of the OTU , which tends to be a better-understood organism with higher-quality annotations such as Escherichia coli K-12 . If the representative organism for an OTU did not have a protein with that role in a subsystem or with a literature backing , a representative protein was chosen at random from another member of the OTU . The computation of annotation likelihood scores was designed based on the principle that genes with more similar sequences are more likely to share the same function , but recognizing that these relationships are far from perfectly predictive [43] . The computation thus attempts to quantify the uncertainty in relation to the available database of high-confidence annotations by accounting for both the similarity of the query gene to genes in the reference database and the distribution of annotations of the reference genes ( Figure S3 ) . Annotation likelihoods were calculated by first running BLASTP [44] , [45] with an E-value cutoff of 10−5 against all of the genes in the high-confidence gene annotation data set . A log-score for each ( query , target ) pair was computed as:Eij is the E-value for the BLASTP hit between the protein products of genes i and j and Sij is the log-score between them . The parameter k = 10−200 was used to prevent the log E-value from being undefined due to a reported E-value of zero . After calculating log-scores for all ( query , target ) pairs , a likelihood score that each gene i ∈ GO , where GO is the set of genes in the organism , is also a member of the set Aa of genes with annotation a was computed as follows:Aa represents the set of genes with annotation a , the maximum score of BLASTP hits from a gene in the query organism to a gene in the high-quality database , and PC = 40 is a pseudocount used to dilute the likelihoods of annotations for annotations with weak homology to the query . The sum in the numerator is over all BLASTP hits from gene i to genes with a particular annotation a and the sum in the denominator is over all BLASTP hits from gene i . Squaring the scores prevents large numbers of weak hits from dominating the computed likelihood . The pseudocount was chosen to set the likelihood of a gene having moderately high homology ( E = 1E-40 ) to a single protein in the database to 50% and is a typical parameter used in tools such as BLAST [45] to ensure that we offset for potentially incomplete or biased database representations of gene families . According to this formulation , p ( i ∈ Aa ) will be high only if the protein product of gene i possesses strongly significant sequence similarity to reference proteins with annotation a and does not possess similarly strong similarity to proteins with other annotations . Therefore , the metric takes into account two different sources of annotation ambiguity: divergence of sequence and disparity of annotations for similar proteins in the target database . Annotation likelihoods are useful for evaluating individual annotations , but must be converted into likelihoods for metabolic reactions in order to use them in the context of evaluating a metabolic network . Reaction likelihoods represent the confidence in the inclusion of a particular reaction in a metabolic network . The conversion from annotation likelihoods into reaction likelihoods takes into account the facts that annotations can imply multiple functional roles , a protein with a particular functional role could be part of multiple protein complexes , and multiple complexes could catalyze the same reaction . We used the ModelSEED reaction database [20] as the source of all of these links . The first step in computation of reaction likelihoods is the conversion of gene annotations into functional roles , to account for the possibility that an annotation implies multiple protein functions . The likelihood that each gene i ∈ GO also belongs to the set of genes Rr with functional role r was computed as the sum of the likelihood that gene i had each annotation that maps to role r . Here the mapping a→r from annotations to roles is defined by the ModelSEED database . This definition ensures that if a protein could be multi or single-functional , the final reaction likelihoods reflect each of those possibilities . In the second step , the likelihood that at least one gene in GO had role r was computed as the maximum likelihood of the role across all genes in GO . The genes most likely to fulfill role r ( within 80% of the maximum ) were retained and linked with an OR relationship to form a Boolean Gene-Function relationship . In the third step , the ModelSEED reaction database was used to compute the likelihood of existence of protein complexes from the likelihood of existence of functional roles . A protein complex represents a set of protein functions that must all be present in order to build a multi-subunit enzyme ( for example , an ATP binding subunit and a translocating subunit must both be present to build certain ABC transporters ) . Since all of the subunits must be present to perform the function , the likelihood of the existence of a complex c in the cell was computed as the minimum likelihood of the roles associated with it . The mapping r→c from roles to complexes is provided by the ModelSEED reaction database . The sets of genes possessing each function in a complex were linked with an AND relationship to form a Boolean Gene-Protein relationship . In the fourth step , reaction likelihoods were computed from protein complex likelihoods using complex-reaction links in the ModelSEED . Since multiple complexes can independently catalyze a reaction , the likelihood of the existence of a reaction x in the cell was computed as the maximum likelihood of the possible complexes that could catalyze it . The complexes that could catalyze the same reaction were linked with an OR relationship to form a Boolean Gene-Protein-Reaction relationship ( GPR ) [46] . Only complexes with a likelihood within 80% of the maximum complex likelihood associated with a reaction were retained in the GPR for that reaction . The computed GPR was used in simulations of gene knockouts for gap filled reactions , and the reaction likelihoods were used as weights in the objective function for likelihood-based gap filling ( see below ) In all workflows , targeted gap filling is performed first in order to activate the biomass equation and achieve growth on ‘complete’ media . Complete media consists of all compounds for which the organism has transport reactions in the draft reconstruction , and hence the solution to this gap filling problem represents reactions that would be necessary for growth on any more limited media for which the organism possesses transporters . Targeted gap filling is performed using either the parsimony-based or the likelihood-based approach . The parsimony-based gap filling approach , used in the ModelSEED for auto-completing models [20] , has been described previously [19] , [25] . The parsimony-based approach minimizes the number of additions to a model . Penalties are added for use of less-confident biochemical databases , adding reactions with ambiguous compounds , adding of transport reactions , or making thermodynamically unfavorable reversibility changes . Higher penalties make it less favorable for the algorithm to make these types of modifications . Detailed descriptions of the formulation and penalties are available in Text S1 . The likelihood-based gap filling approach uses the same MILP formulation as parsimony-based gap filling . However , likelihood-based gap filling uses reaction likelihoods to re-weigh the objective coefficients . To do this , the likelihoods of reactions p ( x ) are first converted into costs C ( x ) by inverting them:Then , modified gap filling objective coefficients λgapfill , x are computed as follows:where λgapfill , x is the objective coefficient in the gap filling formulation for reaction x and the P-values represent the same penalties as used in the existing parsimony-based approach ( see Text S1 ) . In our modified formulation , higher-likelihood reactions are given lower costs ( though the thermodynamic penalties for adding a reaction in the wrong direction are not changed ) and are therefore favored in the optimization provided their benefit outweighs costs of other reactions in a pathway . The numeric parameters ( 12 and 10 ) in the equation make changing the reversibility of a reaction with low estimated Gibbs energy equivalent to adding ( on average ) one to two intracellular reactions in a favorable direction , while changing a reaction with reaction with an estimated Gibbs energy of 10 kCal/mol is equivalent to adding three average intracellular reactions in a favorable direction . Due to the difficult nature of mixed-integer programs , obtaining a solution can take a long time for certain problems . Therefore , we have implemented a system in which a time limit is initially imposed on solution time and automatically increased if solving fails . An error is ultimately thrown if the solver fails to find a solution within one day or if no solution exists . In iterative gap filling work flows , all reactions in the model that are associated with one or more genes are targeted to enable flux . Iterative gap filling is similar to the previously published gap find and gap fill algorithms [19] , but operates on inactive reactions rather than dead-end or orphaned metabolites . This is accomplished by performing targeted gap filling on one reaction at a time until as many reactions as possible are functional . These targeted gap fillings can be performed using either the parsimony-based approach ( parsimony-based iterative gap filling ) or the likelihood-based approach ( likelihood-based iterative gap filling ) . The results of iterative gap filling depend on the order in which the targets are processed . In our studies , the order was selected based on the region of metabolism in which the reaction occurs . Central carbon reactions were gap filled first to ensure that core metabolism was functional . These were followed by reactions involved in biosynthesis of essential metabolites ( amino acids , nucleotides , and cofactors ) , finally culminating in reactions involved in peripheral utilization and degradation pathways . After ordering reactions according to this priority , flux variability analysis [47] was used to determine if each reaction had a non-zero maximum flux . If the maximum flux was zero , gap filling ( likelihood or non-likelihood-based ) was run to attempt to activate the reactions with pathways from the ModelSEED reaction database . If a gap filling solution was found , it was integrated into the model before moving onto the next reaction in the model . The final result was a set of pathways that activated a maximum number of reactions in the model . Since iterative gap filling attempts to fill the maximum number of gaps in a model , solutions that fill different gaps in the model are often redundant or very poorly supported . To solve this problem , we have implemented a reaction sensitivity analysis that identifies for each gap filled reaction ( including reversibility changes for existing reactions in a network ) : a ) whether each gap filled reaction causes other reactions in the model to become inactive when it is removed and b ) whether the gap filled reaction is predicted to be essential for growth . After performing reaction sensitivity analysis , any non-contributing reactions , which are non-essential gap filled reactions that do not activate any other reactions in the network , were removed from the model . For parsimony-based iterative gap filling , reaction sensitivity was performed on all reactions in the reverse order in which they were added , so that lower-priority gap filling solutions would be tested for removal first . For likelihood-based iterative gap filling , reaction sensitivity analysis was done in order from lowest to highest likelihood so that gap filled reactions that were unsupported by genetic evidence would be tested for removal first . In order to support simulations of Biolog data and achieve greater completeness of the metabolic network , the models gap filled on complete media are gap filled again to achieve non-zero biomass production on a minimal media . In this study Carbon-D-Glucose was used as the minimal media . This gap filling step can be performed using either the targeted or likelihood-based approach . To perform phenotype simulations , high-throughput data is imported into the KBase workspaces and saved as PhenotypeSet objects in the workspace . These objects contain links from phenotype sets to specific media and genes in a genome , and represent the results of Biolog or knockout lethality experiments in a consistent format . In this study , the ModelSEED algorithm [20] was used to build a draft model for each of 22 organisms for which either gene knockout lethality data ( 8 organisms ) , Biolog data ( 9 organisms ) , or both ( 5 organisms ) was available [48]–[60] . All four gap filling workflows ( targeted parsimony-based , targeted likelihood-based , iterative parsimony-based , and iterative likelihood-based ) were independently applied to the draft model to build working models of each of these organisms . The gap filled models were verified to predict positive biomass production on Carbon-D-Glucose media using flux balance analysis [61] before performing further simulations . To simulate gene knockout lethality phenotypes , the models growing on Carbon-D-Glucose media were first further gap filled ( if necessary ) to achieve nonzero biomass production on the media in which knockout experiments had been performed ( this was only necessary for Mycobacterium tuberculosis ) . Subsequently , the knockouts were simulated by evaluating the Boolean GPR rules for each reaction and setting the maximum rate of each reaction whose GPR evaluated to FALSE to 0 . Flux balance analysis was then used to maximize the biomass equation . The knockout was considered lethal if the predicted biomass production rate was less than 10−9 hr−1 . For parsimony-based gap filling knockout simulations were performed both before and after integrating predicted gene protein reaction associations for gap filled reactions into the model . To simulate Biolog data , the models growing on Carbon-D-Glucose media were modified to possess transporters for every compound in every media in the Biolog array . After this modification , growth on each media was tested by setting exchange reactions for each compound not in the media to zero and using flux balance analysis to predict the biomass production rate . The model was considered non-growing if the predicted biomass production rate was less than 10−9 hr−1 . All gap filling for this manuscript was performed outside of KBase using CPLEX under an academic license ( IBM Corporation , version 12 . 5 ) [62] . Due to licensing restrictions , gap filling performed on KBase servers is done using SCIP 3 . 0 . 2 [63] . Phenotype simulations and sensitivity analysis were performed using GLPK version 4 . 43 . The gap filling and likelihood computations are implemented in the KBase framework with web service APIs and a web interface ( http://iris . kbase . us ) . Detailed descriptions of all steps in the workflow are available in Text S2 and S3 . Wilcoxon signed-rank tests were performed using the signrank function in MATLAB statistics toolbox version 7 . 1 , using the ‘exact’ method and 2-tailed p-values .
|
Genome-scale metabolic modeling is a powerful approach that allows one to computationally simulate a variety of metabolic phenotypes . However , manually constructing accurate metabolic networks is extremely time intensive and it is thus desirable to have automated computational methods for providing high-quality metabolic networks . Incomplete knowledge of biological chemistries leads to missing , ambiguous , or inaccurate gene annotations , and thus gives rise to incomplete metabolic networks . Computational algorithms for filling these gaps in a metabolic model rely on network topology based approaches that can result in solutions that are inconsistent with existing genomic data . We developed an algorithm that directly incorporates genomic evidence into the decision-making process for gap filling reactions . This algorithm both maximizes the consistency of gap filled reactions with available genomic data and identifies candidate genes for gap filled reactions . The algorithm has been integrated into KBase's metabolic modeling service , an automated metabolic network reconstruction framework that includes the ModelSEED automated metabolic reconstruction tools .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"systems",
"biology",
"computer",
"and",
"information",
"sciences",
"network",
"analysis",
"biology",
"and",
"life",
"sciences",
"metabolic",
"networks",
"computerized",
"simulations"
] |
2014
|
Likelihood-Based Gene Annotations for Gap Filling and Quality Assessment in Genome-Scale Metabolic Models
|
Ebola virus ( EBOV ) infections are characterized by deficient T-lymphocyte responses , T-lymphocyte apoptosis and lymphopenia . We previously showed that disabling of interferon-inhibiting domains ( IIDs ) in the VP24 and VP35 proteins effectively unblocks maturation of dendritic cells ( DCs ) and increases the secretion of cytokines and chemokines . Here , we investigated the role of IIDs in adaptive and innate cell-mediated responses using recombinant viruses carrying point mutations , which disabled IIDs in VP24 ( EBOV/VP24m ) , VP35 ( EBOV/VP35m ) or both ( EBOV/VP35m/VP24m ) . Peripheral blood mononuclear cells ( PBMCs ) from cytomegalovirus ( CMV ) -seropositive donors were inoculated with the panel of viruses and stimulated with CMV pp65 peptides . Disabling of the VP35 IID resulted in increased proliferation and higher percentages of CD4+ T cells secreting IFNγ and/or TNFα . To address the role of aberrant DC maturation in the IID-mediated suppression of T cell responses , CMV-stimulated DCs were infected with the panel of viruses and co-cultured with autologous T-lymphocytes . Infection with EBOV/VP35m infection resulted in a significant increase , as compared to wt EBOV , in proliferating CD4+ cells secreting IFNγ , TNFα and IL-2 . Experiments with expanded CMV-specific T cells demonstrated their increased activation following co-cultivation with CMV-pulsed DCs pre-infected with EBOV/VP24m , EBOV/VP35m and EBOV/VP35m/VP24m , as compared to wt EBOV . Both IIDs were found to block phosphorylation of TCR complex-associated adaptors and downstream signaling molecules . Next , we examined the effects of IIDs on the function of B cells in infected PBMC . Infection with EBOV/VP35m and EBOV/VP35m/VP24m resulted in significant increases in the percentages of phenotypically distinct B-cell subsets and plasma cells , as compared to wt EBOV , suggesting inhibition of B cell function and differentiation by VP35 IID . Finally , infection with EBOV/VP35m increased activation of NK cells , as compared to wt EBOV . These results demonstrate a global suppression of cell-mediated responses by EBOV IIDs and identify the role of DCs in suppression of T-cell responses .
The 2013–2016 outbreak of Ebola virus ( EBOV ) in West Africa claimed the lives of 11 , 300 people [1] . EBOV infections are characterized by ‘immune paralysis’ , the profound immune deficiency resulting in uncontrolled viral replication [2] . A characteristic feature of EBOV infections is lymphopenia , which is observed in both humans and experimentally infected nonhuman primates ( NHPs ) [3–10] and is particularly pronounced during fatal human cases [9–11] . Fatal human cases and studies with EBOV-infected NHPs also demonstrated apoptosis of T cells accompanied by upregulation of tumor necrosis factor related apoptosis inducing ligand ( TRAIL ) and Fas/FasL [11 , 12] . Moreover , EBOV infection of macaques resulted in depletion of T-cells , NK-cells but not CD20+ B cells , and no detectable activation of T-cell [4] . The lack of T cell activation in infected macaques contrasts a recent study of EBOV survivors , which received EBOV-specific antibody treatment and demonstrated a substantial immune activation of T and B cells [13] . Thus , the available information on the effect of EBOV on cell-mediated response is incomplete and controversial . Type I interferons ( IFN-I ) are well-characterized inflammatory mediators whose interaction with IFNα/β receptors ( IFNAR ) is critical for controlling viral infections [reviewed in reference[14] . IFNAR induces the Janus activated kinase-signal transducer that results in activation of transcription JAK-STAT pathway in the majority of cells , along with other pathways , some of which are cell type-specific , which jointly transcriptionally control expression of hundreds of IFN-stimulated genes ( ISG ) [15] . IFN-I directly regulates activation of numerous immune cell types including dendritic cells ( DCs ) , T-lymphocytes , B-lymphocytes and NK cells [16–22] . IFN-I has been shown to affect monocyte and macrophage functions and differentiation [14 , 23 , 24] . Furthermore , IFN-I stimulates antibody-dependent cytotoxicity of macrophages while exerting both positive and negative regulation of secreted inflammatory mediators [14 , 25] . IFN-I triggers macrophages to upregulate nitric oxide synthase 2 , resulting in enhanced IFNγ-induced oxidative response and eventually enhanced phagocytosis [17 , 26 , 27] . With regards to DCs , IFN-I has multiple effects , including differentiation from monocytes , maturation and migration [23 , 28–32] and enhancing their antigen presentation capacity [14 , 16 , 23 , 24 , 28–30 , 33] . The presence of IFN-I during antigen-dependent maturation of DCs has been shown to strongly enhance their capacity to induce human antibody responses and CTL expansion [18 , 19 , 34 , 35] . IFNα/β stimulation of immature DCs leads to a rapid upregulation of cell surface markers associated with the initiation of an adaptive immune response including MHC class I , MHC class II , CD40 , CD80 , CD86 and CD83 [16 , 24 , 33] . INFα has been shown to promote expression of chemokine receptors such CC chemokine receptor type 7 ( CCR7 ) while both IFNα and IFNβ are required for the migration of plasmacytoid DCs to the marginal zones occupied by T-lymphocytes in vivo [36] . Signaling through IFNAR in T-lymphocytes is critical to the development of their effector functions . As noted above , IFN-I increases presentation of MHC-associated antigenic peptides on the surface of antigen presenting cells ( APCs ) that in turn results in antigen-specific activation of T-lymphocytes . IFN-I can exert it effects on immune cells either directly , through IFNAR signaling , or indirectly by the induction of chemokines , which promote recruitment of immune cells to the site of infection and result in the release of a second wave of immune modulatory cytokines [36] . Studies in mice have shown that IFNα promotes efficient cross-priming of antigen-specific CD8+ T-cells and secretion of IFNγ [18 , 37] . IFNα has been shown to be a critical regulator of genes involved in the CTL responses [38 , 39] . IFN-I stimulation of naïve CD4+ T-cells results in their differentiation into IFNγ-producing Th1 cells [19] . IFN-I directly enhances the functional role of CD4+ T-cells in the development of antibody responses [20] . Furthermore , IFN-I have been shown to directly inhibit regulatory T-cell function thereby promoting optimal antiviral T-cell responses during acute infection [21] . Conversely , IFN-I , along with IFN-III , may directly suppress proliferation of CD4+ T cells in context of viral infection [40] . Hence , IFN-I exhibits complex , pleotropic effects that are T lymphocyte subset specific . IFN-I also has a profound effect on immune cells other than DCs and T cells . In addition to the ability to enhance antibody responses through the effects on DC and CD4+ T cells mentioned above , IFN-I also directly stimulates the ability of B cells to secrete antibodies [20] . IFN-I is critically associated with the production of all subtypes of immunoglobulin G ( IgG ) and the development of long-lived plasma cells and immunological memory [20 , 41–43] . IFN-I was shown to enhance secretion of IFNγ and numerous other cytokines by NK cells through an autocrine IFNγ-dependent activity and enhance cytolytic activity [22 , 44] . IFN-I also promotes both expansion and survival of proliferating NK cells via IFN-I/STAT1-dependent production of IL-15 [14 , 45 , 46] . Due to these pleiotropic effects , viruses have evolved targeted subversion strategies aimed at blocking IFN-I signal transduction by targeting the JAK-STAT pathway . One of the characteristic features of EBOV infection is the strong antagonism of IFN-I responses by IFN-inhibiting domains ( IID ) located in the viral proteins VP24 and VP35 , including the suppression of cytosolic sensing of double stranded RNA by VP35 IID and the subversion of IFN-induced signaling by both VP35 and VP24 IIDs [reviewed in reference[47] . Another important feature of EBOV infections is the lack of maturation of DCs despite their susceptibility to the virus [48 , 49] . We recently demonstrated that a point mutation disabling the VP35 IID effectively unblocks maturation of DCs exposed to the virus , while a mutation disabling VP24 IID promotes partial maturation [50] . These mutations result in a global modulation of infected DCs transcriptome profiles with only partial overlapping being observed; hence , VP24 and VP35 IIDs are associated with distinct antagonistic mechanisms [51] . In this study we attempted to determine the effects of EBOV VP24 and VP35 IIDs on global adaptive and innate cell-mediated responses . We used recombinant viruses carrying point mutations disabling IID in VP24 ( EBOV/VP24m ) , VP35 ( EBOV/VP35m ) or both ( EBOV/VP35m/VP24m ) previously generated in our lab [50 , 51] each expressing green fluorescent protein ( GFP ) to visualize the infection of susceptible cells . For comparisons , GFP-expressing virus with no mutations in IIDs [52] which otherwise is identical to the mutated viruses , referred here as wild type ( wt ) EBOV , was used . Many important findings on EBOV pathogenesis utilize a mouse model; however , IFN-I has different effects on human and mouse lymphocytes [53 , 54] . We therefore used only primary human immune cells in these studies . We found that IIDs result in a global attenuation and dysregulation of cell-mediated responses , including T , B and NK cells .
To determine the extent of EBOV-mediated suppression of cell-mediated immune response in our experimental systems , we assessed the ability of mock and EBOV-infected DCs to stimulate CMV-specific T-lymphocytes isolated from healthy individual donors seropositive for cytomegalovirus ( CMV ) . CD14+ monocytes were isolated from peripheral blood mononuclear cells ( PBMCs ) using magnetic bead-based methods and differentiated into DCs for 7 days . Differentiated immature DCs were infected with wt EBOV expressing enhanced green fluorescent protein ( GFP ) from an added gene referred here as wild type ( wt ) EBOV [52] and simultaneously pulsed with pooled peptides overlapping CMV pp65 immunodominant protein followed by incubation for an additional 18–24 hours . In parallel , autologous PBMCs were stimulated with CMV peptides for 48 hours , at which point CMV-specific CD137+ T-lymphocytes were isolated and expanded for 8 days . Expanded responder cells were added at day 8 at a 1:1 ratio with mock or EBOV-infected DCs . Following overnight stimulation , cells were stained as described in the Materials and Methods and analyzed for markers of activation . Non-peptide stimulated DCs were used as a control to determine specific induction of CMV responders over background . A significant increase in percentages of T-lymphocytes positive for Ki67 , a nuclear marker of proliferation [55] , was detected as compared to no peptide control ( Fig 1A ) . When T-lymphocyte responders were cultured in the presence of EBOV-infected DCs , percentages of Ki67+ cells were significantly reduced as compared to mock . Intracellular cytokine staining paralleled these findings as infection of DCs with EBOV significantly reduced percentages of T cells positive for activation markers IFNγ , IL2 or TNFα , as well as for the marker of degranulation CD107α+ [56] ( Fig 1A and 1B ) . Despite a reduction in the proportion of activated T cells in response to EBOV infection of DCs , we also observed a moderate increase of the proportion of activated T cells over no peptide controls . This is likely due to the presence of DCs with the lack of high level EBOV replication , which are expected to display MHC-CMV peptides , as infection of DCs with EBOV results in different levels of viral replication in individual cells present in the population [50 , 51] . Overall , these findings are highly correlative with previous in vivo data on the deficient T cell response to EBOV infection . To determine the effects of IID on T cell response , we utilized a panel of recombinant EBOV mutants , each expressing GFP from an added gene to visualize the infection , which included EBOV/VP24m with the VP24 IID disabled by the mutation K124A [50] , EBOV/VP35m with the VP35 IID disabled by the mutation R312A , and EBOV/VP35m/VP24m with both mutations [51] ( Fig 2A ) . PBMCs from CMV-seropositive donors were infected with the panel of recombinant EBOVs and simultaneously stimulated with CMV pp65 peptides for 7 days , and then re-stimulated with the peptides for 6 hours ( Fig 2B ) . Multi-color flow cytometry analysis of CD4+ T cells secreting IFNγ , IL-2 or TNFα as markers of activation demonstrated that disabling of the VP35 IID , but not the VP24 IID , resulted in significantly increased percentages of IFNγ+ and TNFα+ cells ( Figs 2C and S1 , S1 Table ) . The observed effects were more prominent in the dividing 5 , 6-carboxyfluorescein diacetate succinimidyl ester ( CFSE ) -negative cell population . Since the functionality of T cells generally correlates with the number of cytokines and other markers of activation simultaneously expressed by individual cells [57 , 58] , we next quantified CD4+ T cells positive or negative for all 32 possible combinations of IFNγ , IL-2 , IL-4 , IL-17a or TNFα by Boolean gating . We found two abundant populations: IFNγ+ and IFNγ+TNFα+ ( S2 Fig ) . Comparison of these populations and IFNγ+TNFα+IL-2+ , expected to be the most highly activated , demonstrated that disabling of VP35 IID results in the increase of IFNγ+TNFα+ cell populations ( Fig 2D , S2 Table ) . We also detected the increase in the percentages of single-positive IFNγ+ cells in the majority of donors , but the effect did not reach statistical significance . Disabling of any of the two IIDs did not significantly affect the percentage of GFP+ cells , although some reduction was observed when both IIDs were disabled ( S3A Fig ) . These data suggest that the effects of mutations on activation of T cells are not related to changes in viral replication . We next determined the levels of IL-4 , IL-5 and IL-13 , which are markers of Th2 response , in supernatants . Wt EBOV strongly induced the expression of IL-5 and IL-13 ( but not IL-4 ) , while disabling of the VP35 IID resulted in their significant reduction ( Fig 2E ) . These findings suggest that VP35 IID strongly reduces Th1 response . Since T cells are resistant to EBOV infection [3] , mechanisms associated with the VP35 IID-induced suppression of the Th1 response may be related to infection of multiple non-lymphoid immune cells susceptible to EBOV . We hypothesized that the effect results from the deficient maturation of DCs associated with VP35 IID [50] . To test the hypothesis , we established a co-cultivation system , which included monocyte-derived DCs from CMV-positive donors , which were infected with the panel of viruses in the presence of CMV pp65 peptides and subsequently cultured with autologous purified CD4+ T cells ( Fig 3A ) . Following 7 days of co-cultivation , cells were re-stimulated with CMV peptides for 6 hours , and CD4+ T cells were analyzed by multi-color flow cytometry . We found that disabling the VP35 IID strongly increased the proportion of total CD4+ T cells populations secreting IFNγ , TNFα or IL-2 ( Figs 3B and S4 , S3 Table ) . The effect was more pronounced in the actively dividing CFSE- population . Once again , disabling of any of the two IIDs did not significantly affect the percentage of GFP+ cells , and a small reduction was observed with both IIDs disabled ( S3B Fig ) , suggesting the effects of the mutations are not due to changes in the viral replication . To check if the time of peptide stimulation affects the observed effects , we compared simultaneous infection and stimulation used in our experiments , with stimulation 24 hours after infection . We found that the time of stimulation did not affect the percentage of GFP+ DCs ( S5A Fig ) or the percentage of activated IFNγ+ CD4+ T cells ( S5B Fig ) . We next quantified CD4+ T cells positive or negative for IFNγ , IL-2 , IL-4 , IL-17a and TNFα by Boolean gating , resulting in 32 possible combinations . The vast majority of cells exposed to the panel of viruses or SEB belonged to the IFNγ+ , TNFα+ and IL-4+ single-positive populations , and IFNγ+TNFα+ double-positive populations ( Figs 3C , S6 and S7 , S4 Table ) . Disabling of VP35 IID resulted in a significantly increased percentages of IFNγ+ , IFNγ+TNFα+ and IFNγ+TNFα+IL-2+ cells . The increases were observed both in the dividing ( CFSE- ) and the total cell populations . To determine the effects of IID under conditions when the majority of CD4+ T cells respond to stimulation , CMV-specific CD4+ T cells from three donors were expanded and used in co-culture assays with autologous DCs ( Fig 3D ) . Under these conditions , not only the disabling of VP35 IID , but also VP24 IID resulted in significant increases of cytokine secreting T cells compared to wt EBOV ( Fig 3E ) . We next quantitated cytokines and chemokines in the medium of expanded CD4+ T cell responders cultured for 24 hours with CMV-pulsed DCs pre-infected with the panel of viruses ( Fig 4 , S5 Table ) . Infection of DCs with wt EBOV resulted in a significant reduction of the majority of cytokines and chemokines analyzed in comparison to mock-infected DCs , with a few exceptions , including IL-4 , IL-5 and IL-13 ( Fig 4A ) . These data further confirm suppression of T cell responses in general , and Th1 responses , by EBOV demonstrated in Figs 1 and 2 . Disabling of either VP24 IID or VP35 IID resulted in an increase , as compared to mock-treated cells , of the majority of the cytokines analyzed in donors 2 and 3 , and a lesser number of cytokines in donor 1 . The effects of the two IIDs were not identical; disabling of the both of them increased almost all cytokines analyzed in donors 2 and 3 , and approximately half in donor 1 . Remarkably , the levels of IL-4 , IL-5 and IL-13 did not increase compared to wt EBOV-infected DC . The effects of the mutations on chemokine expression was even stronger as the majority of those analyzed were upregulated in response to either or both mutations ( Fig 4B ) . We next determined if the induction of Th1 response associated with disabling of VP35 IID is related to soluble factors , by assessing the ability of conditioned media to activate naïve CD4+ T-cells . We therefore infected DCs with the panel of viruses or mock-infected and pulsed with CMV-peptides , incubated for 5 days , collected cell-free supernatants and transferred them to naïve CD4+ T cells . Following overnight stimulation , brefeldin A and monenesin were added to media for an additional 6 hours that was followed by intracellular staining of IFNγ . Culture of CD4+ T-cells in conditioned media from DCs infected with EBOV/VP35m or EBOV/VP35m/VP24m resulted in significantly increased percentages of IFNγ+ CD4+ T cells ( Fig 4C and 4D ) . To further demonstrate the direct role of VP35 IID in the restriction of a Th1-response , lentiviral vectors expressing wild type or mutant VP35 were used to transduce DCs . Co-culture of CD4+ T-cells in the presence of DCs transduced with the mutant VP35 resulted in a significant increase in the percentages of IFNγ+ CD4+ T-cells compared to wild type VP35 ( S8 Fig ) . Similar results were obtained both in the presence or absence of CMV peptides . These data suggest that the restriction of Th1 response by VP35 IID at least in part is mediated by soluble factors , and the observed effects are not related to changes in biological properties of the virus due to the introduced mutation . Taken together , the results demonstrate that the VP35 IID , and in a lesser degree VP24 IID , suppress activation of CD4+ T cells as a result of the IID-associated deficient maturation of DC . Previous reports have indicated that DCs require IFN-I response to mature [59]; however , induction of the IFN-I signaling pathway , but not release of IFN was reported to be a requirement for maturation of DCs following induction by negative-strand RNA viruses [60] . We therefore determined whether the observed suppression of DCs maturation by VP35 IID is a consequence of the suppression of IFN-I signaling and whether released IFN has any effect on this phenotype . First , we blocked IFN receptor 2 ( IFNAR2 ) , since the anti-viral activities of IFN-I correlate well with its binding to IFNAR2 , rather than IFNAR-I [61] . For this purpose , IFNAR2 blocking antibodies were added to DCs at a concentration 30 μg/ml; we previously demonstrated that a 1 , 000-fold lesser dose of the antibody suppresses the expression of the IFN-inducible genes Mx1 and ISG56 in T cells [62] . Cells were incubated with IFNAR2 blocking antibodies at 37°C for 1 hour , followed by inoculation with the viruses as indicated previously . At 40 hours post infection , we examined the expression levels of markers associated with DC maturation including CD86 , CD80 and CD54 ( Figs 5A and S9A ) . Even though wt EBOV only induced low expression of CD86 , IFNAR2 blockade further reduced it to the level of mock-infected DCs . No significant effects of the blockade were detected for CD80 , and the effect on CD54 was similar to that of CD86 , but somewhat less pronounced . The reduction of the level of CD86 in DCs infected with wt EBOV by the IFNAR2 blockade is likely a result of the synergistic effect of the blockade with the effects of the VP35 and VP24 IID present in the virus . In contrast , in DCs infected with EBOV/VP35m , the blockade did not result in significant changes in the expression of any of the three markers of maturation in both GFP+ and GFP- cells . Since EBOV/VP35m has the intact VP24 IID , which inhibits IFN-I signaling , and VP35 IID suppresses not only double stranded RNA cytosolic sensing but also inhibits other host defense pathways [reviewed in reference[47] , these findings suggest that DC maturation associated with disabling of VP35 IID is only partially dependent on IFN-I . This hypothesis is supported by the release of numerous cytokines including TNFα following infection with EBOV/VP35m ( Fig 4A ) , which could stimulate DC maturation by induction of pathways other than the IFN-I pathway . Furthermore , these findings are consistent with the increased Th1 response by conditioned media from DCs infected with EBOV/VP35m ( Fig 4C and 4D ) and with the lack of difference between maturation of infected ( GFP+ ) and uninfected ( GFP- ) DCs exposed to EBOV/VP35m we reported earlier [50] . These results differ from the previous observation with an HIV gag vaccine in mice , when blocking of IFN-I receptors prevented effective DCs maturation [59] . In separate experiments , the effects of exogenously added IFNα2 or IFNβ were determined . Due to the anti-viral effects of IFNs , DCs were infected and incubated for 24 hours before IFNα2 or IFNβ was added at the concentrations 1 , 000 and 800 IU , respectively . This was followed by an additional 20 hour-long incubation , and analysis of the expression of the maturation markers CD86 ( Fig 5B ) , CD80 , and CD54 ( S9B Fig ) . No significant effect of exogenous IFN on the expression of CD86 and CD80 was found . Contrary to the expectations that exogenous IFNs would facilitate DCs maturation , the expression of CD54 was reduced , rather than increased in both the infected and mock-infected cells . In addition , no significant differences in the expression of the three maturation markers were found between GFP+ and GFP- cells . These results support the previously published observation that secreted IFN-I is irrelevant for the induction of DC maturation by viruses [60] . Taken together , our results suggest that the suppression of IFN-I signaling and the prevention of IFN-I release by EBOV per se play only a limited role in the suppression of DCs maturation by the IID that is consistent with their diverse and highly redundant mechanisms of the suppression of the innate immune response by EBOV IIDs [47] . Stimulation of T cells by DCs is a key step , which affects the magnitude of the immune response to a viral infection [63] . To determine if VP35 IID interferes with the ability to form immunological synapses between DCs and T-lymphocytes , we co-cultured purified CD4+ T-cells with autologous CMV-pulsed DCs infected with wt EBOV or EBOV/VP35m . Twenty four hours following infection , autologous CD4+ T-cells were added at a 1:1 ratio , cultured for an additional 4 hours , fixed , and stained as described in the Materials and Methods . Immunological synapses between DCs and T cells were visualized by confocal microscopy , which demonstrated co-localization of HLA-DR and CD3ε in mock-infected cultures . Infection of DCs with wt EBOV resulted in a significantly reduced number of immunological synapses as compared to mock-infected DCs , while disabling of VP35 IID significantly increased the number in comparison to wt EBOV ( Fig 6A and 6B ) . In addition , the intensity of the staining of the immunological synapses and their sizes were greater in DCs infected by EBOV/VP35m than in cells infected with wt EBOV . Furthermore , we noted that infection with wt EBOV reduced the HLA-DR expression compared to mock-infected cells , while no reduction was observed with EBOV/VP35m , the finding further confirmed by flow cytometry analysis ( Fig 6C ) . Taken together these data suggest that EBOV VP35 IID interferes with IS formation by reducing DC maturation , thereby limiting the capacity of DCs to effectively generate an adaptive immune response . We next examined the effects of VP24 and VP35 IID on proliferation of T cells . Total PBMC from four healthy CMV-seropositive donors were labeled with CFSE , inoculated with the panel of viruses with or without simultaneous pulsing with CMV peptides , and cultured for 7 days ( Figs 7A and S10 , S6 Table ) . Infection with wt EBOV did not affect the percentages of the dividing CFSElow cells without peptide stimulation , but reduced it , compared to uninfected cells , in presence of peptides . Infection with EBOV/VP35m resulted in an increase , as compared to wt EBOV , in proliferation of both CD4+ and CD8+ T cells in PBMCs from each donor , with or without peptides , although the effect was weak for CD4+ T-cells with peptides , and the overall effect did not reach statistical significance due to the high donor-to-donor variability . In contrast , no consistent effect was observed in PBMCs infected with EBOV/VP24m or EBOV/VP35m/VP24m . Based on these data , we hypothesized that IID also suppresses expression of CD69 , which is involved in lymphocyte proliferation as a signal transmitting receptor [64] , and Ki67 , a nuclear marker of proliferation . We utilized the enriched CMV responder assay described in Fig 3D , in which DCs and autologous isolated CMV-specific CD137+ CD4+ T-lymphocytes were cultured for 8 days , infected with the panel of the viruses , and pulsed with CMV peptides . Next , expanded CMV-specific T-lymphocyte responders were combined at a 1:1 ratio with the infected DCs and after 24 hours long co-cultivation , analyzed by flow cytometry . In T-cells cultured with wt EBOV infected DCs , the percentages of both CD69+ and Ki67+ CD4+ T cells were reduced as compared to T-cells cultured with uninfected DCs ( Fig 7B and 7C ) that suggests suppression of T cell activation/proliferation by EBOV and is consistent with the data shown in Figs 1 and 7A . Again , disabling of VP35 IID completely reversed the suppressive effect of wt EBOV . Unexpectedly disabling of VP24 IID , in addition to VP35 IID , or both , also completely reversed the suppressing effect of IIDs in this system . Taken together , these data demonstrate the suppressive effect of both VP35 and VP24 IID on T cell proliferation . Previous studies have demonstrated that DCs and other APCs are vital to the survival of CD4+ T-cells due to MHC-TCR complex-dependent signal transduction that require a direct cellular contact [65–67] . We therefore sought to examine the effects of IID on signal transduction using co-cultures of unstimulated DCs and autologous CD4+ T-cells . Following a four-day co-culture with mock or EBOV-infected DCs , the phosphorylation cascades associated with TCR signaling were analyzed by Western blotting ( Fig 8A ) . Specifically , we examined phosphorylation of TCR complex-associated adapters and downstream signal molecules , which have previously been shown to remain phosphorylated in the absence of antigen-dependent activation , and whose phosphorylation depends on DC-T-cell contact [65 , 68–70] . In the absence of DCs , a limited phosphorylation of signal transduction mediators was detected in CD4+ T-cells . In contrast , CD4+ T-cells cultured in the presence of mock-infected DCs exhibited phosphorylation of molecules activated at both early and late stages of TCR-mediated signaling . Presumably , the observed phosphorylation status represents basal phosphorylation events that while supporting survival of CD4+ T-cells , remain below the threshold required for activation . Infection with wt EBOV resulted in the increase in phosphorylation of Lck; however , phosphorylation of additional adapters including ZAP70 , PLCγ1 and SLP76 appeared to be blocked . Consistent with the data demonstrating reduced formation of immunological synapse ( Fig 6 ) , the absence of downstream phosphorylation events may be the result of impaired synaptic formation and/or aberrant signal transduction . Phosphorylation of ZAP70 is dependent on the formation of TCR microclusters that form scaffolding for downstream signaling events at the intracellular sites of immunological synapse formations [71] . Thus the absence of phosphorylated ZAP70 in co-culture of T-cells with wt EBOV-infected DCs strongly indicates at suboptimal engagement of TCR . In agreement with that , infection of DCs with wt EBOV significantly reduced expression of HLA-DR ( Fig 6C ) . It is not immediately clear why a relatively strong induction of Lck phosporylation was observed following wt EBOV infection; however , it is plausible that secondary signal transduction may be suboptimal resulting in impaired downstream signaling . Disabling of VP35 IID unblocked phosphorylation of ZAP70 and PLCγ1 , and to a lesser degree , SLP76 , while disabling of VP24 IID unblocked only ZAP70 , but to a much greater degree than disabling of VP35 IID . Unexpectedly , infection with EBOV/VP24m resulted in an increase in the relative amount of phosphorylated ZAP70 despite the lack of phosphorylation of the downstream signaling molecules PLCγ1 and SLP76; however , this is consistent with the lack of effective T cell activation by this mutant . This discrepancy may result from differential kinetics associated with altered signaling events , the absence of co-activating signals or upregulation of negative regulators of PLCγ1 and SLP76 or other inhibitory factors downstream ZAP70 . Surprisingly , disabling of both IIDs resulted in only weak phosphorylation of PLCγ1 and SLP76 , but not ZAP70 . We note however , the absence of CD3ζ and Lck phosphorylation in EBOV/VP35m/VP24m co-cultures , suggesting that the phosphorylation of PLCγ1 and SLP76 may be associated with alternate signaling pathways . Although not examined in these studies , it is possible that both phosphorylation and dephosphorylation kinetics may be variable following co-culture . To further characterize the capacity of infected DCs to transmit survival signals , we determined the relative levels of the anti-apoptotic Bcl-2 family members . These proteins are involved in the survival of T cells and are upregulated in response to survival signals [72–74]; on the other hand , survival signal transduction was previously associated with reduced Bcl-2 levels [75–77] . We also determined phosphorylation of Src , which has previously been shown to be essential for naïve T-cell survival and also TCR-dependent [78] ( Fig 8B ) . Bcl-XL was undetectable in control CD4+ T-cells alone , but was readily detectable at relatively similar levels in CD4+ T-cells cultured with mock , wt or mutant EBOV infected DCs . Bcl-2 was readily detectable in CD4+ T-cells cultured with mock or wt EBOV-infected DCs , but greatly reduced when VP35 and/or VP24 IID were mutated . Furthermore , the phosphorylated Src was detectable in T cells cultured with mock-infected DCs and DCs infected with the mutated but not wt EBOV , showing correlation with activation of infected DCs ( Figs 2–7 ) . Taken together , these data demonstrate that cultivation of T cells with EBOV-infected DCs blocks phosphorylation of ZAP70 , PLCγ1 and SLP76 involved in TCR signaling and the pro-survival molecule P-Src , and that the levels of P-Src correlate with T cell activation , while that of Bcl-2 correlate with T cell survival in the autologous system . Furthermore , these data identify the role of VP35 IID in blocking phosphorylation of these molecules , and role of VP24 IID in blocking phosphorylation of ZAP70 . Paradoxically , the two IIDs demonstrated an opposite ( i . e . stimulating ) effect on phosphorylation of Lck , as well as on expression of the prosurvival molecule Bcl-2 . A schematic which illustrates the positive and negative signals associated with IIDs as they relate to the observed phosphorylation status of adapter molecules in Fig 8A and 8B is presented in Fig 8C . To date , only limited data concerning the effects of EBOV infection on B and NK cell function have been published . To determine the effects of EBOV infection and IID on B-cells , PBMCs were infected with wt or mutated EBOVs at MOI 1 . 0 PFU/cell , incubated for 7 days , and analyzed for markers of activation , maturation and those associated with specific subsets . The percentages of naïve B-cells ( CD19+CD27-IgD+ ) were reduced after infection with wt EBOV or the mutants except the double mutant ( S11A Fig ) . The percentages of memory B-cells ( CD19+CD27+IgD- ) did not change after infection with wt EBOV , while disabling of VP35 ( but not VP24 ) IID increased this population ( Fig 9A ) . Of note , these results inversely correlated with observed changes in the percentage of naïve B-cell populations ( S11A Fig ) . We next examined the effects of IID on the percentages of class-switched memory B-cells ( CD19+CD27-IgD-IgM-CD20+CD38++ ) ( Figs 9B , S11B and S11C ) . Infection with wt EBOV slightly reduced the percentages of this population , while disabling of VP35 IID significantly increased it . Examination of post-class switched memory B-cells ( CD19+CD27+/DimIgD-IgM-CD20-CD38- ) revealed a similar effect ( Fig 9C ) ; again , the decrease in SEB-treated cells is consistent with an increase in other subsets in samples treated with SEB ( Fig 9B and 9D ) . These results suggest that VP35 IID may , by a yet to be determined mechanism , block the development of post-class switched memory cells . Analysis of plasma cells ( CD19+CD38+CD138+ ) demonstrated the lack of effect of wt EBOV , but as much as 4-fold increase in their percentage after infection with EBOV/VP35m ( Fig 9D ) . Interestingly , infection with the double mutant also increased the percentages of this cell population , but only ~1 . 5-fold , suggesting that VP24 IID reduces the effect of VP35 IID . These data demonstrate that VP35 IID suppresses induction of memory B cells , their class switching , and their differentiation into plasma cells . The primary role of NK cells is the continuous surveillance of “stressed” cells , which is regulated by detection of both activating and inhibitory signals , as well as by cytokines . One of the major molecules , whose expression affects activation of NK cells , is MHCI; reduced MHCI expression levels in virus-infected cells are sensed by NK receptors , which typically result in activation of NK cells [79 , 80] . In addition , IFN-I is required for the optimal NK cell response and promotes the activation and effector functions of these cells [81] . As noted above , our previous study demonstrated that disabling of IID effectively unblocks maturation of EBOV-infected DCs , as evidenced by increased expression of multiples maturation markers , although expression of MHCI was not tested , and increased expression of IFN-I [50] . We therefore examined the effects of IID on NK function in PBMCs by analyzing expression of both activation and inhibitory markers on CD56+CD3- NK cells . We tested expression of the two activation markers: CD38 ( Fig 10A ) , which triggers cytolytic response [82] and NKp46 ( Fig 10B ) , which is an immunoglobulin-like natural cytotoxicity receptor [83] , and three inhibitory receptors: CD27 ( Fig 10C ) , which is expressed by mature cytotoxic effector NK cells [84] , CD158 killer immunoglobulin-like receptor ( Fig 10D ) , which inhibits NK cytotoxicity [85] and the lectin-like receptor KLRG1 ( Fig 10E ) [86] . Paradoxically , infection with wt EBOV increased the percentages of NK cells positive for the activating marker NKp46 and all inhibitory markers tested , as well as the percentage of dead NK cells ( Fig 10F ) . Disabling of VP24 IID resulted in a strong increase in the percentages of cells expressing NKp46 , and a limited non-statistically significant increase of CD38 , and reduced the percentages of dead cells . On the other hand , disruption of VP35 IID resulted in strongly ( 10 . 1-fold versus wt EBOV ) increased expression of CD38 but not NKp46 , which showed an opposite trend perhaps due to altered signaling events . Furthermore , it reduced expression of all three inhibitory receptors . These data suggest , based on multiple activating and inhibitory receptors , that both IIDs suppress activation of NK cells , with the exception of the effect of VP35 IID on the activating receptor NKp46 .
Our findings provide new insights into functional role of IID in EBOV pathogenesis and identify their role as suppressors or modulators of both the innate and adaptive cell-mediated immune responses . We show that EBOV infection of PBMCs resulted in only a limited activation of T cells , while disabling of VP35 IID significantly increased their activation ( Fig 2 ) . As lymphocytes , along with NK cells , are refractory to EBOV infection , we hypothesized that the IID-associated effects are indirect and result from interaction of these cells with other cells susceptible to the virus , such as DCs . To identify the role of DCs , two co-culture systems were used , which included DCs infected with the panel of viruses and CMV-stimulated T cells ( Fig 3A–3C ) , and its modified version with expanded CMV responder T-lymphocytes ( Fig 3D and 3E ) . Analysis of supernatants of CMV-peptide stimulated DCs infected with wt EBOV co-cultured with expanded responder T-lymphocytes demonstrated that most of the cytokines and chemokines analyzed were below that observed in mock-infected control ( Figs 2E and 4A ) . In addition , analysis of supernatants of wt EBOV-infected cultures where indicative of Th2 skewed response , as IL-5 and IL-13 were elevated . This is consistent with the induction of Th2 response in EBOV patients [87] and in macaques infected with Marburg virus [88] , which similarly to EBOV belongs to the family Filoviridae and causes human infections with high case fatality rates . Supernatants of EBOV/VP24m-infected cells also demonstrated elevated levels of IL-13 . However , flow cytometry analysis of T cell phenotypes of CD4+ T-cells from CMV-responder assays indicated that disabling of VP24 IID promotes expression of Th1-associated cytokines ( Fig 3E ) in addition to Th2 , suggesting induction of a mixed Th1/Th2 response . Strikingly , flow cytometry analysis also demonstrated that disabling of VP35 IID results in the significant expansion of activated Th1 population suggesting that VP35 IID limits the capacity of EBOV-infected DCs to initiate the Th1 response . Consistent with that , analysis of EBOV/VP35 IID supernatants demonstrated several fold reduced levels of IL-5 and IL-13 indicating induction of a uniformly Th1-like response . Overall , the cytokine analysis provides a clear indication regarding the inhibitory effects of VP35 and VP24 IIDs , as their disruption by point mutations dramatically increased cytokine/chemokine secretion . A clear cumulative effect was observed in EBOV/VP35m/VP24m cultures as the levels of virtually all cytokines and chemokines were elevated in comparison to either single mutant . This finding suggests that activation of immune cells and their migration to sites of infection may be severely impaired in vivo due to the presence of IIDs . We note that in T cell activation analyses , that the effect of VP24 IID in the double mutant often countered the effects of the VP35 IID in a single mutant ( Figs 2 and 3 ) , which is consistent with our recent transcriptome analysis of DCs infected with the panel of viruses [51] . In general , the EBOV/VP24m mutant exhibited a phenotype consistent with more moderate effects in comparison with EBOV/VP35m . Interestingly , while the effects of VP35 IID on activation and proliferation of T cells in the DC-T cell co-cultivation system ( Figs 2C , 2D , 3B , 3C and 7A ) were strong , the effects of VP24 IID were minimal . We therefore expanded CMV-specific CD4+ T cells that resulted in not only better identification of the effects of VP35 IID , but also demonstrated the effects of VP24 IID ( Figs 3E and 7B ) . The limited formation of immunological synapses in co-culture experiments with wt infected DCs ( Fig 6 ) demonstrated the mechanism by which VP35 effectively blocks the development of an adaptive immune response . As noted above , previous studies demonstrated that infection of DCs with wt EBOV results in their aberrant maturation . Hence , DC-associated ligands required for the formation of synapses are likely inadequately expressed resulting in fewer immunological synapse formations between DCs and T-cells . This reduction results in an inability to reach a signal transduction threshold necessary for cellular activation . This presumption is consistent with both the limited T cell activation observed in co-cultures experiments and the altered phosphorylation cascade profiles when T-cells were cultured with wt EBOV-infected DCs ( Fig 8 ) . The block in immunological synapses formation and the limited T cell activation were highly correlative with the presence of functional VP35 IID in wt EBOV , as disabling the VP35 IID reversed the suppressive effects . The EBOV VP35 IID-mediated suppression of cytosolic sensing and induction of IFN-I response , which otherwise would induce an antiviral state are well established [reviewed in reference [47] . Thus both the lack of proper DC maturation and impairment in the development of an antiviral state blunt the initiation of a T cell response to EBOV infection . Interestingly , the suppression of IFN-I signaling and the prevention of IFN-I release per se during EBOV infection appeared to play only a limited role in the suppression of DCs maturation by the IID ( Fig 5 ) . On the other hand , transfer of conditioned media from DCs infected with EBOV/VP35m , but not wt EBOV , to DC-T cell co-cultures effectively stimulated secretion of IFNγ by T cells ( Fig 4C and 4D ) , despite the block of IFN signaling by intact VP24 IID present in the virus . These data suggest that that some cytokines , such as TNFα whose expression was unblocked by disabling VP35 IID ( Fig 4A ) could contribute the observed DC maturation . More studies are required to mechanistically connect the IFN-inhibiting effects of EBOV IIDs with their effects on maturation of DCs and ultimately phosphorylation of TCR-associated adaptors and downstream signaling molecules . These studies also demonstrate aberrant B cell and NK cell activation by EBOV , thus suggesting the global impairment of the adaptive and innate cell-mediated immune responses by IIDs . As indicated above , B-cell function and maturation are highly affected by IFN-I stimulation in the presence of cognitive antigen ( 35 , 40–42 ) . Previous findings have indicated that survivors of EBOV infection develop and maintain antigen specific adaptive immune responses , which included both CTL and humoral responses but did not include direct examination of antigen-specific B-cell function [89 , 90] . We demonstrated an overall increase in the percentages of class switched and post class-switched memory B-cells in response to disruption of VP35 IID ( Fig 9 ) . These data suggest a suppressive effect of VP35 IID on class switching and identify the role of IFN-I in B-cell differentiation . Disruption of VP35 IID also led to an increase in the overall percentage of B-cells positive for markers associated with plasma cells . Furthermore , consistently with in vivo data demonstrating loss of NK cells in EBOV-infected macaques [4] , infection of PBMCs with wt EBOV appeared to increase the rate of NK cell death . Similarly to our findings regarding increased functional activity , proliferation and differentiation of lymphocytes associated with disruption of VP35 IID , the mutation altered the expression of several NK cell markers in a direction generally corresponding to a greater cytotoxicity , and also reduced NK cell death ( Fig 10 ) . Intriguingly , disruption of VP24 IID dramatically increased the percentage of NK cells expressing the natural cytotoxicity receptor NKp46 . We note that this study was entirely performed with primary human immune cells from donors , which allowed us to detect the magnitude of the effects on cells with different genetic background , as can be seen by a relatively high donor-to-donor variability , and avoid producing skewed results related to a specific genetic background of inbred mice , such as the Th2-skewed response in BALB/c mice [91] . Despite that , the inhibition of activation of T cells co-cultured with EBOV-infected DCs contrasts the T cell activation in EBOV patients [13] . The most obvious difference is that most of CD8+ T cells in patients were positive for Ki-67 , while only 12% of CD4+ T cells in our study were Ki67+ ( Fig 7C ) . This discrepancy can be explained by a much greater increase in the numbers of Ki67+ CD8+ T cells as compared to Ki-67+ CD4+ T cells , as demonstrated with human immunodeficiency virus infection [92] . In addition , activation , including non-antigen specific activation , of T cells in infected patients by other types of cells not included in our studies , or by stimulation related to high doses of EBOV-specific antibodies administered to patients may contribute to the observed in vivo activation through an unknown mechanism . These studies provide evidence of the dual role of the VP35 and VP24 IIDs in the pathogenesis of EBOV . While IIDs are intimately linked to the ability of EBOV to block IFN-I production and signaling , they also counter the ability of DCs to initiate the adaptive immune response . The lack of DC maturation following EBOV-infection presents a significant obstacle in development of the adaptive immune response but also may render immune cell populations susceptible to premature cell death due to aberrant signal transduction and/or the absence of survival signals . Importantly , the suppressive effect of IIDs is not limited to T and B cells , which are the central components of the adaptive response , but also extends to NK cells , the key players in cell-mediated innate immune response . Taken together , these findings suggest global suppressive effects of EBOV IIDs on cell mediated response , and also indicate the potential benefits of blocking the immunosuppressive effects of IIDs as a potential therapy for EBOV-infection .
All work with EBOV was performed in BSL-4 facilities of the Galveston National Laboratory . Flow cytometry was performed either in BSL-4 using the Canto-II instrument ( BD Biosciences ) , or cells were treated with 4% paraformaldehyde in PBS for 48 hours according the UTMB standard operating procedure and removed from BSL-4 for analysis with LSRII Fortessa flow cytometer ( BD Biosciences ) available at the UTMB Flow Cytometry Core Facility . Cells for confocal microscopy were placed on slides , stained , fixed in 4% paraformaldehyde for 24 hours , which was replaced with a fresh solution , incubated for additional 48 hours , and taken out of BSL-4 . To remove supernatants of EBOV-infected cells from BSL-4 , they were gamma-irradiated with the 5 Mrad dose according the UTMB standard operating procedure protocol . Staining and mounting procedures are described below . The staff had the U . S . government permissions and appropriate training for work with EBOV . Generation of the recombinant EBOVs carrying the mutation R312A in the VP35 IID or K142A in the VP24 IID , or both , each expressing GFP from an added gene , and the control GFP-expressing EBOV with no mutations was described previously [50 , 51] . The viruses were propagated on Vero-E6 monolayers and quantitated by plaque titration as previously described [93] . Peripheral blood nuclear cells ( PBMC ) were obtained from buffy coats from anonymous healthy adult blood donors from the UTMB blood bank according to a clinical protocol approved by the UTMB Institutional Review Board . Study population included both CMV-positive and CMV-negative individuals as tested using the Beckman Coulter PK CMV-PA System for qualitative detection of IgG and IgM antibodies to CMV . A pool of 138 CMV peptides ( 15-mers overlapping by 11 amino acid residues ) was obtained from the NIH AIDS Research and Reference Reagent Program . Peptides were reconstituted at 1 mg/ml in DMSO and stored at -70°C in 200 μl aliquots and used to stimulate PBMC or dendritic cells at final concentration of 2 μg/ml . Total PBMC from CMV+ donors were resuspended at 1x106 per ml using 50 ml conical tubes and inoculated at an MOI of 2 PFU/cell with the recombinant strains of EBOV depicted in Fig 2A and simultaneously stimulated with 2 μg/ml of CMV pp65 peptides in media containing 10% human serum ( Corning , Celgro ) at 37°C , 5% CO2 . Staphylococcal Enterotoxin B ( SEB ) ( Sigma Aldrich ) was used as a positive control at a final concentration of 2 μg/ml . Additional controls included mock-infected cells and cells stimulated with 15-mer CMV pp65 peptides only . After 4 hours of incubation , cells were washed twice by centrifugation at 200 x g for 5 min with 2% human serum media and cultured at 1x106 per ml in Advanced RPMI 1640 medium ( Gibco , Life technologies ) supplemented with 10% human serum ( Gemini Bio-Products ) , 2 mM L-glutamine , 200 IU/ml penicillin , and 200 μg/ml streptomycin sulfate ( Invitrogen ) in 6-well plates . PBMC were isolated by density gradient centrifugation ( histopaque; Sigma life science ) . CD14+ monocytes were purified by positive selection using anti-CD14 monoclonal antibody-coated magnetic microbeads according to the manufacturer’s instructions ( Quadro Macs; Miltenyi Biotech ) . CD14+ monocytes were cultured in T-225 flasks ( Corning Incorporated , Costar ) at 6 x 105 cells per ml in Advanced RPMI 1640 medium ( Gibco , Life Technologies ) supplemented with 10% heat-inactivated bovine serum ( Quality Biologicals ) , 2 mM l-glutamine ( Invitrogen ) , 0 . 05 mM β-mercaptoethanol , 50 ng/ml granulocyte-macrophage colony-stimulating factor ( R&D Systems ) , 16 ng/ml interleukin-4 ( R&D Systems ) , 200 IU/ml penicillin , and 200 μg/ml streptomycin sulfate ( Invitrogen ) . The cells were incubated for 7 days at 37°C , 5% CO2 as described previously [50] . Immature DCs generated as described above were harvested , infected with the panel of EBOVs and stimulated with CMV pp65 peptides either simultaneously or 24 hours following infection . After 4 hours of peptide stimulation , cells were washed twice and co-cultured with CFSE ( Molecular Probes , Life Technologies ) -labeled autologous PBMC or purified CD4+ T cells at 37°C , 5% CO2 for 7 days . The DC: lymphocyte ratio was 1:10 ( 2 x 105 DC: 2 x 106 lymphocytes ) in 2 ml of Advanced RPMI 1640 medium . For assays involving purified CD4 or CD8 T cells , these cell populations were purified by negative selection using a primary cocktail of antibodies conjugated to biotin and secondary anti-biotin antibody conjugated to magnetic microbeabs in order to deplete non-CD4+ or non-CD8+ T cells including γ/δ T cells , B cells , NK cells , DC , monocytes , granulocytes and erythroid cells . Lentiviral vectors encoding wt and R312A mutant VP35 were prepared as previously described [94] . Briefly , plasmids encoding the lentivirus , HIV-Gag-pol , VSV-G and SIV-Vpx were transfected into 293T using PEI ( Sigma Aldrich ) transfection reagent . Cell monolayers were incubated overnight , fresh medium was added , and monolayers were incubated for additional 48 hours . Thereafter cell-free supernatants were collected and lentiviral vectors were titrated in 293T cell monolayers . DCs were transduced twice with ~8 hour culture period between the addition of lentiviral stocks . Following three days of culture , mock or medium containing CMV peptides were added . Autologous CD4+ T-cells were added to transduced T-cells at a 1:1 ratio , and the percentages of IFNγ+ CD4+ T-cells were determined 24 hours after initiation of co-culture . After 7 days of culture , cells were harvested , washed and 2 x 106 cells were stimulated for 6 hours in culture medium with 10 μg/ml Brefeldin A ( Sigma-Aldrich ) , 0 . 7 μg/ml monensin ( GolgiStop , BD Biosciences ) , 1 μg/ml anti-CD28 ( BD Biosciences ) , 1 μg/ml anti-CD49d ( BD Biosciences ) , 20 μg/ml DNase ( Calbiochem ) and 2 μg/ml of 15-mer CMV pp65 peptides . PMA ( Sigma Aldrich ) at 20 μg/ml and ionomycin ( EMD Chemicals ) at 1 μg/ml were also used as an additional positive control . Following stimulation , cells were washed 2 x with wash buffer ( PBS , 1% FBS , 0 . 02% sodium azide ) followed by PBS . CD4+ T cells were stained extracellularly with anti-CD3 antibodies labeled with anti-CD3 BD Horizon Brilliant Ultraviolet 395 ( clone UCHT1 , BD Biosciences ) and anti-CD4 PE-CF594 ( clone RPA-T4 , BD Biosciences ) ; for analysis of activation , cells were also stained with CD69-PE/Dazzle ( clone FN50 , BioLegend ) or anti-Ki67-brilliant violet 421 ( Clone B56 , BD Biosciences ) . CD8+ T cells were stained extracellularly with anti-CD3 BD Horizon Brilliant Ultraviolet 395 ( clone UCHT1 , BD Biosciences ) and anti-CD8 BD Horizon PE-CF594 ( clone RPA-T , BD Biosciences ) . Both CD4+ and CD8+ T cell subsets were also stained extracellularly with Live/Dead Fixable Aqua or Near-Infra Red ( Invitrogen ) to discriminate dead cells by flow cytometry for 30 minutes at 4°C . Following extracellular staining , cells were washed , fixed and permeabilized with CytofixCytoperm ( BD Biosciences ) according to manufacturer’s instructions . CD4+ and CD8+ T cells were stained intracellularly with the following antibodies: anti-IFNγ-PE ( clone B27 ) , anti-IL2-allophycocyanin ( APC ) ( clone MQ1-17H12 ) , anti-IL-4-peridinin chlorophyll protein PerCP ) /Cy5 . 5 ( clone 8D4-8 ) , anti-IL17a-BD Horizon V450 ( clone N49-653 ) and -TNFα-Alexa Fluor 700 ( clone Mab11 ) ; all from BD Biosciences . Flow cytometry analysis of DCs for HLA-DR was performed with anti-HLA-DR-PE/Dazzle 594 ( clone L243 , BioLegend ) . B-cell subset staining was performed as follows: anti-CD19-PerCP/Cy5 . 5 ( clone HIB19 , BioLegend ) , anti-IgD-PE/CF594 ( clone IA6-2 , BD Biosciences ) , anti-IgM-brilliant violet 786 ( clone R6-60 . 2 , BD Biosciences ) , anti-CD27-brilliant violet 510 ( clone M-T271 , BioLegend ) , anti-CD24-PE ( clone ML5 , BioLegend ) , anti-CD20-Alexa Fluor 700 ( clone 2H7 , BioLegend ) , anti-CD38-APC ( clone HIT2 , BioLegend ) , anti-IgG-Brilliant Violet 421 ( clone G18-145 , BD Bioscience ) , anti-CD138-fluorescein isothiocyanate ( FITC ) ( clone DL-101 , BioLegend ) . To analyze CD56+CD3- NK cell function/activation , the following antibodies were used: anti-CD38-Alexa Fluor 488 ( clone HIT2 , BioLegend ) , anti-NKp46-Brilliant Violet 786 ( clone 9E2 , BD Bioscience ) , anti-CD27-Brilliant Violet 510 ( clone M-T271 , BioLegend ) , anti-CD158B-APC ( clone DX27 , BioLegend ) , anti-KLRG1-PE-CF594 ( clone 14C2A07 , BioLegend ) , anti-CD56-Brilliant Violet 421 ( clone HCD56 , BioLegend ) and anti-CD3-Brilliant Ultraviolet 395 ( clone UCHT1 , BD Biosciences ) . Cells were subsequently washed twice with Perm/Wash and one time with PBS , fixed with 4% paraformaldehyde ( Polysciences ) and taken out of BSL-4 according to an approved protocol . Cells were washed , resuspended in PBS and 200 , 000 to 500 , 000 events were acquired on the BD LSR II flow cytometer ( BD Biosciences ) . For data analysis , FlowJo version 10 ( Tree Star ) and SPICE ( National Institute of Allergy and Infectious Diseases ) software was used to create Boolean gate arrays that allowed us to determine the frequency of 32 possible response patterns based on the five cytokines tested . PBMCs from CMV-positive donors were resuspended in RPMI media supplemented with 2% human AB serum ( Corning , Celgro ) , and labeled with 5 μM CFSE ( Molecular Probes , Life Technologies ) . CFSE-labeled PBMCs were inoculated with the recombinant strains of EBOV at an MOI of 2 PFU/cell , with or without simultaneous stimulation with 2 μg/ml of CMV pp65 peptides , SEB treated , or mock treated for 4 hours . PBMCs were washed twice to remove virus inoculum and cultured at a concentration of 2 x 106 cells/ml for 7 days in Advanced RPMI medium supplemented with 10% human AB serum , 2 mM L-glutamine , 200 IU/ml penicillin , and 200 μg/ml streptomycin sulfate ( Invitrogen ) in 6 well plates . PBMCs were harvested , stained extracellularly with the following antibodies: anti-CD3 PE-Cy7 ( clone SK7 , BD Biosciences ) , anti-CD4-PerCP/Cy5 . 5 ( clone SK3 , BD Biosciences ) , anti-CD8 APC-Cy7 ( clone SK1 , BD Biosciences ) , and Live/Dead-Far Red Dead cell stain ( Molecular Probes , Life Technologies ) . Cells were subsequently washed , fixed with 4% paraformaldehyde and removed from the BSL-4 according to an approved protocol . Cells were washed with PBS and analyzed by flow cytometry on BD LSR II . PMBCs were pulsed with CMV peptides as previously described . Following 24–48 hours , CD137+ T-lymphocytes were isolated using magnetic bead separation in accordance with the manufacturer’s protocol ( Miltenyi ) . Cells were expanded for 8 days in complete RPMI ( Gibco ) /10% Human serum ( Cellgro ) media supplemented with recombinant IL-2 ( 10 ng/ml ) and IL-7 ( 10 ng/ml ) ( R&D Systems ) and Dynabeads Human T-Activator CD3/CD28/CD137 beads ( ThermoFisher Scientific ) to promote the expansion of CMV-specific responders . DCs were infected at an MOI of 3 and pulsed overnight with CMV peptides . Expanded responders were cultured with 2x105 autologous DCs at a 1:1 ratio for an additional 24 hours . Intracellular staining was performed as described previously . Frozen CD4+ T-cells were recovered at 37°C for 2 hours prior to co-culture with autologous DCs infected overnight with wt or mutant EBOVs ( MOI of 3 ) at a 1:1 ratio . CD4+ T-cells were harvested after 4 days of co-culture , and analyzed by Western Blotting . Briefly , cell pellets were lysed in Laemmli lysis buffer ( Invitrogen ) , separated on 4–12% SDS-PAGE gradient gels ( ThermoFisher Scientific ) and transferred to nitrocellulose membranes using the I-blot system ( ThermoFisher Scientific ) . Thereafter blots were incubated with primary antibodies provided in the T-cell Signaling Antibody Sampler and the Pro-Survival Bcl-2 Family Antibody Sampler Kits ( Cell Signaling Technologies ) . GAPDH was used as an internal loading control ( Cell Signaling Technologies ) . HRP-conjugated secondary antibodies ( Santa Cruz Biotech ) were used for chemiluminescent detection on Hyperfilm ( Amersham ) . For analysis of cytokines and chemokines in CMV-responder—DC co-culture assays , cell-free supernatants were irradiated at 5 Mrad , stored at -80°C and shipped to Eve Technologies on dry ice . Undiluted samples were analyzed using Human Cytokine/Chemokine Array 65-Plex Panel . Heatmaps of values normalized to mock-infected samples were generated using GENE-E software ( Broad Institute ) . In some experiments , conditioned media were collected from DC-T-cell co-culture plates , clarified of cellular debris by low-speed centrifugation and stored at -80°C . Following overnight recovery , CD4+ T-cells were plated in 96-well plates in 100 μl of RPMI1640 medium supplemented by 10% FBS , followed by the addition of 150 μl of conditioned media . Cells were incubated overnight followed by the addition of brefeldin A and monensin for an additional 5 hours . Cells were then stained as previously described . In experiments requiring IFNAR2 blockade , DC were pre-incubated with 30 μg/ml of blocking antibody specific to IFNAR subunit 2 ( CD118; PBL Assay Science ) or mouse immunoglobulin G2a ( IgG2a; R&D Systems ) as an isotype control ( the endotoxin levels in the two antibody preparations were <1 and 0 . 1 endotoxin units per 1 μg of antibody , respectively ) , for 1 hour before virus inoculation . In the experiments involving exogenous IFNs , recombinant human IFNα2a or IFNβ1a ( PBL Assay Science ) was added to DCs at 24 hours after infection at a final concentration of 1 , 000 IU/ml or 800 IU/ml , respectively , and after additional 20 hours , cells were analyzed by flow cytometry . DC were analyzed for cell surface expression of markers of maturation by flow cytometry at 40 hours post-infection . Most DCs were collected by pipetting , and the remaining cells , which were attached to the bottom of the plates , were collected by applying staining buffer ( PBS containing 2% fetal bovine serum and 2 μM EDTA ) . Cells were pelleted by centrifugation at 200×g at 4°C for 5 minutes , buffer was removed , and the pellet was re-suspended in staining buffer . DCs were incubated for 20 min on ice in the dark with the monoclonal antibodies anti-CD86-PerCP-Cy5 . 5 ( clone FUN-1 ) , anti-CD80-APC-H7 ( clone L307 . 4 ) , and anti-CD54-PE ( clone HA58 ) . In addition , IgG2a-PerCp-Cy5 . 5 , IgG1-APC-H7 , and IgG1-PE were used as the respective isotype control antibodies ( all from BD Biosciences ) . Following incubation , cells were washed three times with the staining buffer and re-suspended in 200 μl of the same buffer . The Far Red fluorescent dye ( Invitrogen ) was used to evaluate cell viability by flow cytometry . Data were acquired using a flow cytometer FACSCanto II in BSL-4 facility or fixed with formalin as described above , taken out of BSL-4 and analyzed with BD LSR flow cytometer . Data were analyzed using FlowJo 7 . 6 . 1 software ( Tree Star ) . Suspensions of DCs and lymphocytes were fixed with 4% paraformaldehyde for 15 minutes and washed 3 times with PBS . Cells were then pelleted by centrifugation at 600xg for 6 minutes and resuspended in 50 μL of PBS . Cells were loaded on charged slides and dried overnight at 4°C . For staining , cells were rehydrated with PBS and permeabilized with 0 . 5% Triton X100 solution ( ThermoFisher Scientific ) in PBS for 15 minutes , washed with PBS , incubated with 0 . 5 M glycine in PBS for 10 minutes at room temperature , and washed 3 times with PBS . Antigen blocking was performed using 5% donkey serum ( Sigma Aldrich ) diluted in stain buffer ( 1% BSA and 0 . 1% Triton X100 in PBS ) for 30 minutes . Mouse monoclonal antibody targeting HLA-DR ( ThermoFisher Scientific , clone 7 . 3 . 19 . 1 , 1:10 dilution ) and goat polyclonal targeting CD3ε ( Santa Cruz , 1:50 dilution ) were used as primary antibodies and diluted in stain buffer as indicated . After 1 hour incubation at 37°C , slides were washed 3 times in washing buffer ( 0 . 1% Triton X100 in PBS ) , incubated with a mixture of two secondary antibodies: donkey anti-mouse conjugated with Alexa Fluor 647 ( ThermoFisher Scientific ) and donkey anti-goat conjugated with Alexa Fluor 594 ( ThermoFisher Scientific ) both diluted at 1:200 in stain buffer for 1 hour . Next , washed cells were incubated with 6-diamin-2-phenylindole dihydrochloride ( DAPI ) ( ThermoFisher Scientific ) at 1 μg/ml for 2 minutes , and washed 3 times in PBS . Slides were then fixed in 4% paraformaldehyde and removed from BSL-4 . Slides were washed in 0 . 5 M glycine , washed in PBS , and mounted with coverslips using PermaFluor mounting medium ( ThermoFisher Scientific ) . Infected cells were identified by expression of GFP encoded by the recombinant viruses . Slides were analyzed by laser scanning confocal microscopy using Olympus FV1000 confocal microscope . Lasers with 405 nm wavelengths were used for DAPI excitation , 488 nm for GFP , 543 for Alexa Fluor 594 , and 635 nm for Alexa Fluor 647 . This was followed by counting of immunological synapses , defined as colocalizations of CD3 and MHC-II ( HLA-DR ) . For each experimental group , the data presented are based on count of at least 10 confocal imaging acquisitions . Results of the count were expressed as numbers of synapses per 100 cells . Statistical analyses and generations of graphs were performed using GraphPad Prism version 6 . 05 ( GraphPad Software ) . Statistical significances were calculated using a paired T-test . Statistical significance was set at p < 0 . 05 .
|
The extensive investigation of interferon antagonism mediated by Ebola virus ( EBOV ) over the last 16 years resulted in identification of two interferon inhibiting domains ( IIDs ) located in the VP24 and VP35 proteins of the virus and of multiple mechanisms by which the domains disable the innate immune system and promote replication of the virus . However , the effects of these domains on cell-mediated immune response had not been investigated . To determine the effects of IIDs on cell-mediated responses , we used a panel of recombinant strains of EBOVs with point mutations disabling the VP24 and/or VP35 IIDs . The viruses were used for infection of peripheral blood mononuclear cells ( PBMCs ) or dendritic cells ( DCs ) , which were subsequently co-cultured with T cells . We found that IIDs block activation and proliferation of T cells as a result of their functional role in suppressing maturation of DCs and limiting the formation of immunological synapses . Similarly , IIDs were demonstrated to suppress activation and differentiation of B cells , and skew activation of NK cells present in infected PBMCs . These data provide evidence of previously unknown effects of IIDs on the adaptive and innate cell-mediated immune responses and identify a novel mechanism of “immune paralysis” during EBOV infections .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"phosphorylation",
"flow",
"cytometry",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"immunology",
"cloning",
"developmental",
"biology",
"molecular",
"development",
"molecular",
"biology",
"techniques",
"cytotoxic",
"t",
"cells",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"t",
"cells",
"molecular",
"biology",
"spectrophotometry",
"immune",
"system",
"biochemistry",
"antibody-producing",
"cells",
"cytophotometry",
"cell",
"biology",
"post-translational",
"modification",
"b",
"cells",
"nk",
"cells",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"spectrum",
"analysis",
"techniques"
] |
2016
|
The Ebola Interferon Inhibiting Domains Attenuate and Dysregulate Cell-Mediated Immune Responses
|
The mechanosensitive channel of large conductance ( MscL ) has become a model system in which to understand mechanosensation , a process involved in osmoregulation and many other physiological functions . While a high resolution closed state structure is available , details of the open structure and the gating mechanism remain unknown . In this study we combine coarse grained simulations with restraints from EPR and FRET experiments to study the structural changes involved in gating with much greater level of conformational sampling than has previously been possible . We generated a set of plausible open pore structures that agree well with existing open pore structures and gating models . Most interestingly , we found that membrane thinning induces a kink in the upper part of TM1 that causes an outward motion of the periplasmic loop away from the pore centre . This previously unobserved structural change might present a new mechanism of tension sensing and might be related to a functional role in osmoregulation .
Mechanosensitive ion channels are ubiquitous membrane proteins that enable a cell to respond to deformation forces in the surrounding lipid bilayers or cytoskeleton . This process , known as mechanosensation , is thought to have evolved to protect bacterial cells from sudden osmotic shock [1] , [2] . In eukaryotes , mechanosensation is involved in a physiological processes including hearing , touch sensation and gravitropism [2] . Shortly after the discovery of mechanosensitive ion channels in Escherichia coli bacteria [3] the gene of the mechanosensitive ion channel of large conductance ( Eco-MscL ) was identified and cloned [4] . The crystal structure of the closed pore MscL [5] from Mycobacterium tuberculosis ( Tb-MscL ) revealed a homo-pentameric channel where each subunit consists of two transmembrane ( TM ) helices , TM1 and TM2 , connected by an extracellular loop and cytoplasmic N- and C-terminal . In the closed state , the TM1 helices are tightly packed to form a narrow constriction referred to as the hydrophobic gate . Gating is induced by tension in the surrounding lipid bilayer that triggers a large conformational change to form an open channel of approximately Å diameter [6]–[9] . The structure and function of MscL , have been investigated extensively using a range of techniques including patch clamp studies ( see [10] , [11] and [2] for reviews ) , mutation studies [12]–[16] , FRET [9] , [17] , EPR spectroscopy [7] , [18] , [19] , structural modelling [20] , [21] and MD simulations [22]–[32] . Based on the original crystal structure [5] and the large conductance of the pore it was initially thought the open pore is lined by all 10 TM helices . A series of EPR experiments [7] , [18] , [19] resulted in a revised model of the open pore that is mostly lined by TM1 . This can only be achieved by a large tilting motion of the TM helices which allows them to cover the increased surface of the pore . The suggested gating mechanism involves an iris-like opening that creates an open pore with a significant decrease in membrane span . A large number of experimental and computational studies have provided insight into the structure and function of MscL but there remain questions about the details of the open pore structure and the gating mechanism . It is also unclear how the channel senses the tension in the membrane that subsequently triggers gating . Answering these questions is essential for our understanding of mechanosensation at the molecular level in both bacterial and eukaryotic cells , and to aid in the design of engineered channels with novel functionalities [33] . Molecular Dynamics ( MD ) simulations are well suited to examine the structure and dynamics of proteins but atomistic MD simulations of membrane proteins are computationally expensive and are commonly restricted to 10's or 100's of ns . This timescale is considerably shorter than most physiological processes such as channel gating , which often take place in the millisecond range . Furthermore , the large conformational changes associated with ion channel function are often separated by significant energy barriers and long simulations are necessary to ensure sufficient conformational sampling . Different approaches have been used to address the timescale and sampling issues of standard atomistic MD simulations . One way is to apply external forces to the protein to induce or accelerate the wanted conformational change . A number of studies used different forms of surface tension or negative pressure on the membrane and/or radial forces applied to the protein to model the gating of MscL [23]–[28] . From these simulation studies it is evident that long simulations are needed to obtain an open pore of the MscL protein without the use of radial forces or large tensions . An alternative way of addressing the sampling issue it to use restraints based on experimental data which directs the evolution of the system through the conformational space as demonstrated by several studies to model the structure of membrane proteins [7] , [9] , [34] , [35] . The timescale issue of MD simulations can be addressed by using coarse graining ( CG ) . The idea is to group several atoms into a single particle which reduces the system size and removes the fastest degrees of freedom such that a larger time step can be used , making simulations in the range feasible ( see [36] , [37] for a review of CG models for proteins ) . In addition , CG simulations also show some enhanced sampling efficiency as a direct result of the reduced number of effective interactions between particles . Yefimov et al . [29] used CG MD simulations to model the WT and mutants of Tb-MscL with and without tension . Several simulations of multiple were carried out and produced an expanded and stable water filled pore . The tension used was lower than in previous simulation studies of MscL but still exceeded the tension that is required to open the pore under physiological conditions . Here we combine CG MD simulations with experimental restraints by incorporating inter-subunit distances and solvent accessibility data from EPR and FRET experiments into a CG model of MscL . The aim of this study is to obtain an open pore structure of the MscL protein that is consistent with experimental data and thus gain insight into the structural re-arrangements during gating . Multiple simulations in the range allowed us to achieve much greater conformational sampling and observe structural changes that were not seen in single shorter simulations . By combining CG simulations with experimental restraints we were able to induce gating and model the open pore structure of MscL without using excessive tension . Furthermore , we observed previously unseen structural changes that may have an important functional role .
Figure 1 provides an overview of the simulation approach . The starting structure of the closed state MscL protein was an atomistic homology model of Eco-MscL [38] which is based on the crystal structure of Tb-MscL [5] and the structural models of Sukharev et al [20] , [21] . The simulation system was modelled using MARTINI , a biomolecular force field for CG simulations [39] , [40] . In the MARTINI model four heavy atoms are , on average , represented by a single CG particle . All non-bonded particles interact via a Lennard-Jones potential energy function and the strengths of this interaction is used to mimic the chemical nature of the different particle types . In addition , charged particles interact via a Coulombic energy function . The MARTINI model has been successfully used to model lipid membranes [41]–[43] and membrane proteins [29] , [30] , [32] , [44] , [45] . Note , secondary structure is maintained in the MARTINI model using standard bonded potentials . An additional elastic network was not used . An equilibrated CG structure of the closed state with water and lipid was used as the starting structure for all simulations . The inter-subunit distances and the solvent accessibility data from EPR and FRET experiments were converted into restraints and incorporated into the CG model . A series of simulations with different combinations of solvent and distance restraints and tension were carried out . Two different tensions were used: 12 dynes/cm , the tension required to induce gating in patch clamp experiments of MscL [8] , [46] , and 30 dynes/cm . The latter value was used as experience with previous CG MD simulations of MscL showed that a larger tension was needed to induce gating [29] , [30] . The different simulation protocols are summarised in Table 1 . Simulations without any restraints ( ) and with tension only ( ) were mainly used as references for comparison during data analysis . Different combinations of using the full set of EPR and FRET distance restraints with and without tension were carried out . In addition , simulations with a reduced set of distance restraints were carried out , in which 10% of randomly selected restraints were removed . For each protocol we aimed to collect a minimum of two trajectories to ensure reproducibility . In all restrained simulations , both distance and solvent restraints were introduced over a period of 1000 ns . After that , the simulation was continued for another 500 ns with the restraints in place . For simulations with distance restraints this was followed by a further 500 ns where the distance restraints were removed . Simulations were carried out using GROMACS 4 . 0 . 4 [47] with a 25 fs time step . The protein was embedded in a 1-Palmitoyl-2-oleoylphosphatidylcholine ( POPC ) lipid bilayer and solvated with water . The final simulation system consisted of 1 MscL protein ( 1435 particles ) , 1152 POPC lipids ( 14'276 particles ) and 23'868 water particles bringing the total particle count to 40'279 . The simulation system was contained in a rectangular box of dimension 20×20×12 nm and periodic boundary conditions were applied in all directions . The system was equilibrated with backbone particles restrained to their starting position for 20 ns followed by an unrestrained simulation of 10 ns . The system was simulated in an NPT ensemble at 310 K using a Berendsen temperature coupling scheme with a time constant of 0 . 25 ps . Berendsen pressure coupling with a reference pressure of 1 bar and a time constant of 0 . 5 ps was used . For simulations without tension , a semi-isotropic pressure coupling was used with a reference pressure of 1 bar in the x/y and z direction . Tension was simulated using a surface-tension pressure coupling scheme where the surface tension is coupled to the x/y dimension to simulate constant lateral tension in the membrane while normal pressure coupling is used in the z direction . For each simulation a final average protein structure was calculated by minimising the protein structure obtained from the average position of all backbone particles from the last 50 ns of simulation . The average structures were only used for visual inspection of kinks in TM1 but not for the quantitative analysis , the calculation of RMSD values , pore radius or radial position . Details on analysis protocols are given in the Text S1 . In addition to the simulations in a POPC lipid bilayer we carried out simulations of the MscL protein in short-chain lipids . The simulation system was identical to the one used for restrained simulations in POPC except that the POPC lipids were replaced by Dimyristoyl phosphatidyl choline ( DMPC ) lipids . POPC lipids contain a 16 carbon and a 18 carbon-chain tail while DMPC has two tails with 14 carbons each . The simulations in DMPC did not contain solvent restraints or distance restraints . Table 2 summarises the raw experimental data used for the restraints while Figure 2 depicts the position of all residues for both the solvent and distance restraints . EPR experiments provided data for lipid and water accessibility and inter-subunit distances for residues in the TM region [7] and accessibility data for the periplasmic loop [38] . Additional inter-subunit distances for selected residues across the protein were obtained from FRET spectroscopy [9] . Distance restraints representing distance between equivalent residues in each subunit from FRET were implemented using a flat bottom harmonic potential where the target distance is equal to the inter-subunit distance in the open state obtained from FRET experiments and the width is equal to the experimental uncertainty . The EPR distance restraints were implemented using harmonic half potentials as all inter-subunit distances in the open state are 15 Å . Based on the geometry of the pentameric structure of MscL each inter-subunit distance resulted in 10 distance restraints . A total of 90 distance restraints from FRET and 650 distance restraints from EPR were incorporated into the GROMACS topology of the CG model of MscL using tabulated potentials . Solvent restraints were based on the lipid and water accessibility measurements of residues in the TM helices and the periplasmic loop from EPR experiments . The raw experimental data was converted into restraints by changing the interactions of the side chain particles with water to represent the change in solvent exposure of selected residues upon opening of the channel . Although lipid and water accessibility are separate measurements in EPR experiments , they are conceptually related . To simplify the implementation , the data from lipid and water accessibility were combined into a single parameter which represents a change in hydrophobic character of a given residue during gating of the channel . Using the reported results from 54 residues in TM1 and TM2 [7] and 28 residues in the periplasmic loop [38] was determined for each of these residues and incorporated into the MARTINI force field [39] . We created a new set of particles that showed stronger or weaker interactions with water particles but unaltered interactions with all other particle types . These new particle types were used to change interactions between the side chain particles of selected residues and water particles based on their value ( see Tables S1 to S3 in Text S1 ) . The final simulation system contained 740 distance restraints and 265 solvent restraints that represent a change in inter-subunit distances and solvent exposure between the closed and the open channel . It was important to slowly introduce the restraints rather than impose a sudden change onto the closed pore . This was achieved using the coupling parameter from the free energy perturbation method in GROMACS . Further details of the implementation of the restraints are given in Text S1 .
Before examining the potential open structures resulting from the different simulations we investigate which combinations of restraints are most effective in producing a stable open pore and examine how the different restraints and tension affect the various functional domains of the protein . Monitoring the pore radius is the most direct structural measure to determine which restraints are most effective in producing an open pore . Figure 3 shows the pore radius as a function of time for some of the simulation protocols . All simulations involving distance restraints ( and ) show a steady increase in pore radius during the first half of the simulations in which the restraints are introduced . In contrast , simulations using solvent restraints and/or tension ( and ) show only a slightly larger pore radius than to simulations without restraints ( ) . Even simulations using a tension of 30 dynes/cm did not cause spontaneous opening . Previous CG simulations [30] suggest that this is likely a kinetic effect and higher tension is required to observe spontaneous channel opening in the range without restraints . Similar observations were made in previous CG simulations of MscL where tensions above 60 dynes/cm were required to observe a significant pore opening [29] . These results raise the question if the solvent restraints or tension have any effect on the structure when distance restraints are applied . The pore radius of the simulation starts to drop off shortly after the restraints are removed while the pore radius of and simulations remains more stable , especially for simulations with tension . This suggests that tension and solvent restraints both have a stabilising effect on the open pore structure , something confirmed by comparing the distance restraint energy of the different protocols . A more detailed analysis of the distance restraints can be found in Figure S1 in Text S1 . To investigate the effect of the restraints and tension on the different domains of the protein we prepared plots of root mean square deviation ( RMSD ) vs time for the various functional domains and simulation protocols ( Figure 4 and Figure S2 in Text S1 ) . The RMSD of the hydrophobic gate and TM1 show the same pattern as seen in Figure 3 since the pore radius is calculated from residues in the hydrophobic gate , which is part of TM1 . All the simulations involving tension have a larger RMSD of the periplasmic loop than those without , suggesting that tension induces structural rearrangements in this domain . Results for other domains also indicate that the solvent restraints and tension not only stabilise the open pore but have an effect on the structure of the protein independent of the distance restraints . We selected a first set of potential open pore structures from the simulation protocols that produced a structure with a pore radius larger than 12 Å . Comparison of structures from and simulations indicates that the pore size at the hydrophobic gate is a necessary but not sufficient structural change to produce an open pore , as can be seen in Figure 5 . The structures from simulations show a widened pore in the TM region but the periplasmic loop failed to expand , occluding the extracellular side of the pore . In comparison , the periplasmic loop from the simulation has expanded along with the TM domains . This is also visible from the RMSD vs time plots ( Figure 4 ) which show that the tension increases the extent of structural changes of the periplasmic loop . Analysis of structures from simulations showed that solvent restraints have some effect on the position of the periplasmic loop such that it is not occluding the pore entrance like in the simulation . A more detailed analysis showed that in simulations with tension the periplasmic loop lies more flat on the membrane surface and has a more pronounced tendency to become integrated into pore lining than in the absence of tension . Close inspection of the structures from ( not shown ) revealed that the integrity of the pore is compromised . No such effects were observed in other simulations suggesting that combining the full set of solvent and distance restraints with tension is exerting too much force on the protein . After this visual inspection structures from simulations were rejected and not considered for further analysis . Figure 6 shows plots of RMSD vs residue for three open pore structures from and simulations in comparison to a simulation without restraints ( ) . The graphs in Figure 6A were produced by aligning the entire protein of the open pore structure to the equilibrated closed pore structure . The overall pattern of the graphs is very similar to the ones reported in previous MD simulations of MscL [22] , [48] . Two structures from simulations with tension ( ) show an increase in RMSD in the periplasmic loop in comparison to the simulation confirming the effect seen in RMSD vs time plots ( Figure 4 ) . The increased RMSD suggests that this domain of the protein undergoes more pronounced structural changes in the presence of tension . The increased RMSD of TM1 and TM2 compared to simulations is consistent with an increased mobility of most residues in TM1 and selected residues in TM2 in the open state reported by EPR experiments [7] . To differentiate between rigid body movements of domains and changes in the internal structure of these domains , we prepared plots of RMSD vs residue where the RMSD was calculated by aligning the open and closed structures using consecutive segments of 5 residues ( Figure 6B ) . The data shows that the majority of residues in TM1 and TM2 show low RMSD compared to other functional domains suggesting that the helices maintain structural integrity as may be expected given the bias to maintain secondary structure in the MARTINI force field . However , residues corresponding to the periplasmic end of TM1 ( residue 38–43 ) and the cytoplasmic half of TM2 ( residues 90–107 ) show levels of RMSD as high as the periplasmic loop and the C-terminal . That this arises in all open pore models is indicative that some deformation of the TM helices occurs during channel gating which is discussed in more detail below . The most distinctive features of the open channel are the pore radius and the tilting of TM1 and TM2 , making these structural measurements an obvious choice for comparing our open pore structures to previously reported models of the open channel . Table 3 lists the pore radius and the helix tilt for TM1 and TM2 for the three open structures from this study along with data from previous studies . The pore radius of the 3 structures , calculated as an average over the last 50 ns of simulation , ranges from 13 . 7 Å to 15 . 8 Å and is in good agreement with data from EPR experiments ( Å ) [7] , FRET experiments ( radius change of 8 . 0–8 . 5 Å ) [17] , estimates based on conductance ( 15–18 . 5 Å ) [8] , [20] and recent atomistic MD simulations ( 12 . 0–14 . 5 Å ) [9] and CG simulations ( 11 . 6 Å , minimum radius ) [30] . Previous unrestrained CG simulations of Tb-MscL at higher tension ( 77 dynes/cm , 323 K ) produced a significantly smaller pore of 4 Å in radius and an only slightly larger pore of 6 Å [29] at increased temperature ( 338 K ) suggesting that combining tension with restraints is more effective that simply increasing the tension in the bilayer . The helix tilt values from simulations without tension ( ) are very close to the values from atomistic MD simulations [9] . Furthermore , both results from this study and from atomistic simulations of MscL [9] predict an increased helix tilt for TM1 and TM2 in the presence of tension . This is consistent with the notion that the helix tilting and the resulting flattening of the pore is a consequence of the protein trying to adapt to the tension-induced change in hydrophobic thickness of the membrane [49] and the suggestion that changes in membrane thickness accelerate the conformational changes involved in the gating process [48] . To estimate the relative motion of the helix pairs that form the pore we calculated the RMSD between the open and closed state by aligning TM1 and TM2 of the same subunit as well as TM1 of one subunit with TM2 of the neighbouring subunit ( Table 4 ) . RMSD from TM helices of neighbouring subunits is lower then for TM helices of the same subunit and much lower than from aligning the entire TM domain , suggesting adjacent TM1 and TM2 helices show little relative motion during the gating . Figure 7 shows the radial position of each residue for the structures shown in Figure 5 , to visualise the structural change involved in gating and to compare these structures to previous open state models . In addition , the plot shows the distance restraints from FRET and EPR . The radial position for the same structures as in Figure 5 are shown as these are the structures that show a pore radius above 12 Å . A comparison of the , and structures to the closed pore confirms that tension is essential to obtain a fully open pore as seen in the much smaller radial position in the TM2 and loop domain in structures without tension , despite an opening in the hydrophobic gate . This confirms the observations from the results in Figure 5 . Comparing the and structures to open pore models from previous studies shows a smaller radial position of the residues in the hydrophobic gate . At the same time the pore radius from these structures is in close agreement with these other models ( Table 3 ) . This suggests that the same sized pore can be reached with less movement of the backbone . The residues in TM2 from our fully open pore also shows less movement than earlier models . Our results support the notion that a stable and wide pore can be obtained with less structural changes in TM1 and TM2 than suggested by earlier models , as has been noted by Corry et al [9] . Furthermore , there is a marked difference between the C-terminal in our models and previous ones as the radial position remains unchanged in comparison to the closed pore . While the distance restraints are based on five-fold symmetry of the closed pore structure , the use of half and flat harmonic potentials for the implementation of the distance restraints means that the symmetry was not rigorously enforced during the simulation . As visible in Figure 8 ( as well as Figure S3 and S4 in Text S1 ) the open pore structures show some loss of symmetry . Based on the data from FRET and EPR experiments on fully labelled MscL proteins it is not possible to tell if and how much the pentameric symmetry is retained during gating . Furthermore , recent unrestrained CG simulations of MscL [30] suggested that the gating mechanism might be asymmetric . We therefore did not enforce or maintain the symmetry during the simulation and did not symmetrize the open pore structures . The structures from the and simulations are available as part of the supplementary material ( Dataset S1 ) . While there is a general consensus on the iris-like opening and the helix tilting mode of gating there has been little agreement on the potential rotation of the TM helices . The open pore model by Sukharev et al . [20] , [21] proposes a clockwise rotation of TM1 while the model reported by Perozo et al . [7] , [19] suggested a counterclockwise rotation which was supported by other experimental studies [50] , [51] . A similar interpretation was made of results obtained by modifying electrostatic interactions and cross-linking [52] . In contrast , the open pore model obtained from restrained atomistic MD simulations [9] and results from a disulphide cross-linking study [53] showed little rotation of TM1 and no rotation is seen in the crystal structure of the tetrameric Staphylococcus aureus MscL ( Sa-MscL ) in an expanded intermediate state [54] . Figure 8 depicts the equilibrated closed pore structure in comparison to the open pore structure obtained here from a simulation with distance restraints and tension . The open pore structure from other simulation protocols can be found in Figure S4 in Text S1 . Using the position of the residues V23 , G26 , I92 and I96 relative to the pore centre as an indication of TM1 rotation we note that the majority of the residues showed no change in orientation between the closed and open states . Thus , there is little evidence of rotation in the TM helices independent of the restraints used . The lack of rotation and the low RMSD of TM helices from neighbouring subunits indicates that the pair of helices move radially and tilt as a rigid body in agreement with the results of a recent study of the open MscL structure determined by FRET spectroscopy and all atom MD simulations [9] . Note that rotation refers to the change in relative position of the helices as a whole and does not involve changes in secondary structure and hence is not affected by the bias of the MARTINI force field towards maintaining secondary structure . In addition to the structure of the pore there has been considerable attention to the structure and possible function of the C- and the N-terminal domains [12] , [13] , [16] , [22] , [55]–[57] . An earlier model of MscL suggested that the N-terminal acts as a second gate by forming a helical bundle that occludes the pore on the cytoplasmic side [20] , [21] , [53] . However , a recently re-examined structure of the Tb-MscL in the closed state [58] and a systematic study of the N-terminal domain [57] shows that the N-terminal forms a helical structure running along the membrane water interface . In all our simulations , independent of pore opening or restraints applied , the N-terminal remained stable , and lies parallel to the cytoplasmic membrane ( Figure 8 and S3 in Text S1 ) . The same behaviour was observed in atomistic [9] and coarse grained [30] MD simulations of MscL . The combined results from experimental and simulation studies make it unlikely for the N-terminal to act as a second gate . In the original crystal structure of Tb-MscL [5] the C-terminal was solved to R118 ( corresponding to R126 in Eco-MscL ) and the remaining residues were modelled as a continued -helix . However , the conformation of C-terminal was unusual as a number of hydrophobic residues were pointing outwards to face the aqueous cytosol while some of the charged residues where pointing inside the helical bundle . It has been suggested [59] that this is an artefact of the low pH and the presence of detergent molecules during crystallisation . MD simulations [22] , [59] and a re-examined structure of Tb-MscL [58] suggest a more reasonable conformation with the charged residues pointing outwards . The starting structure in our simulations was a homology model of Eco-MscL with a C-terminal similar to the revised Tb-MscL with L121 and L128 facing outwards and L122 , I125 and L129 facing inwards . After a 10 ns equilibration without any restraints or tension the orientations of the already inward facing I125 , L122 and L129 did not change significantly while L121 and L128 had a tendency to move towards an inward facing conformation . Comparison of the equilibrated and the final structures from different simulation protocols showed that the orientation of the side chains did not change significantly during the 2 simulations and that there is no difference between closed and equilibrated open pore structures in this respect . Based on our simulations it is likely that these hydrophobic side chains are pointing inside the helical bundle . In addition to the conformation of the C-terminal there has also been a different hypotheses as to the functional role of the C-terminal . Some studies suggested that the C-terminal dissociates during opening of the pore [8] , [54] , [60] while others suggested it might act as a size-exclusion filter [59] . As depicted in Figure 8 and Figure S3 in Text S1 the helical bundle remained intact in all our simulations which is consistent with results from cross-linking [59] and EPR [38] , [61] experiments . Comparison of the structures from different simulations revealed an upward movement of the C-terminal that was present in all simulations independent of restraints or tension . Figure 8 depicts the position of the C-terminal in the equilibrated closed pore in comparison to the open pore structure from a simulation with distance restraints and tension . Structures from other simulations can be found in Figure S3 in Text S1 . The upward motion of the C-terminal results in contacts between the C-terminal and transmembrane domains . A number of results from the different analyses point towards the role of tension in structural changes during gating . Visual comparison of the open and closed pore structures revealed a change in position of the periplasmic loop as well as structural rearrangements in the upper part of TM1 . In the closed pore the periplasmic loop tends to point towards the extracellular space . This is in contrast to the open pore in which the loop shows an outward motion towards the membrane surface . Comparison of structures from different protocols revealed that this structural change is much more pronounced in simulations with tension . In the absence of tension the loop mostly remains in the extracellular space and often occludes the pore entrance while tension causes the loop to move towards the membrane surface and away from the pore centre . This conformational change of the loop is not correlated with pore opening at the hydrophobic gate . simulations resulted in a channel with an increased pore radius at the hydrophobic gate but the loop showed no outward motion , causing the loop to occlude the pore entrance ( in Figure 5 ) . In contrast , and simulations did not induce opening at the hydrophobic gate but the loop moved towards the membrane surface resulting in an accessible pore . These observations imply that tension is strongly related to the outward motion of the periplasmic loop during gating and that this is independent of the opening at the hydrophobic gate . Similar observation were reported in a recent study using MD simulations to investigate the energetic contributions to MscL gating [32] showing a tension-dependent increase of area on the extracellular side of the protein even when the hydrophobic gate remains closed . The results from this study further showed that a large part of the gating energy is actually provided by the adaptation of the channel at the lipid-water interface . Further inspection of the open pore structures showed that the re-orientation of the periplasmic loop is accompanied by the formation of a kink in the upper part of TM1 ( around residue 40 ) ( see also [62] ) . The formation of the kink is only possible and might be induced by the simultaneous thinning of the membrane as a straight helix is required to span the resting thickness of the membrane . In contrast , the kink allows the top of the helix to lie along the surface of the thinned membrane . To further investigate the formation of the kink in simulations with and without tension we first carried out a visual inspection of the average structures from the different simulation protocols to identify the presence of a kink and its position . Figure S5 in Text S1 summarises the results by showing the fraction of structures for which a visible kink is present as well as the position of the kink for simulations in POPC without tension , simulations in POPC with tension and simulations in DMPC without tension . In simulations without tension the kink occurs outside of TM1 while structures from simulations with tension show a shift in the position , with the majority of kinks occurring between residues 38 to 44 , corresponding to the periplasmic end of TM1 . This effect is also visible in Figure 6 that revealed a change in the internal structure of the upper part of TM1 around residues 37 to 45 . The combined results support the observation that the formation of the kink and the concurrent outward movement of the periplasmic loop during gating are related to presence of tension in the membrane . To further support the hypothesis that there is an increased occurrence of kinks in the upper part of TM1 in the presence of tension we carried out a more quantitative analysis where the kink was defined as the deviation from a straight helix as found in the closed pore structure . An RMSD vs residue was calculated for the TM1 helices in the 17 structures by comparing them to the close pore . Based on these datasets we calculated the % occurrence of kinks for simulations with and without tension . 39% of subunits from simulations without tension showed a deformation of the external end of TM1 compared to the closed structure and this number increased to 51% in simulations with tension . Based on these findings we hypothesise that the thinning of the membrane is associated with the formation of a kink in the upper part of TM1 and that this structural change is accompanied by an outward motion of the periplasmic loop towards the membrane surface . To support our hypothesis regarding the effect of tension on the periplasmic loop we conducted simulations of the MscL embedded in DMPC , a lipid that yields a thinner membrane in comparison to POPC without the application of tension . The pore radius of two final structures from DMPC simulations is 8 . 5 Å , which is almost equal to the pore radius of the equilibrated closed pore . That the hydrophobic gate is closed is further confirmed by a nearly constant RMSD for residues in the hydrophobic gate ( Figure S6 in Text S1 ) . There is a clear increase of RMSD of the periplasmic loop in DMPC simulations to the same level as seen in POPC and POPC simulations indicating that the periplasmic loop responds to the thinner membrane in the same way as it responds to tension . This increased activity of the periplasmic loop is also seen in plots of RMSD vs residue for DMPC and POPC simulations ( Figure S7 in Text S1 ) . The results confirm that the periplasmic loop exhibits similar behaviour in DMPC simulations as in POPC simulations with tension . Apart from a small increase in RMSD in TM1 there are no significant structural changes in the other functional domains of the protein . None of the characteristics of the open pore such as an increase in pore radius and helix tilt were observed in DMPC simulations further supporting the notion that the activity in the periplasmic loop is independent of channel gating . We repeated the analysis to identify the position of the kink for structures from DMPC simulations . Figure S5 in Text S1 demonstrates that there is an increased occurrence of kinks around residues 39 to 41 in DMPC simulations . Our quantitative analysis showed that 44% of subunits from the DMPC simulations showed a kink . It is worth emphasising that these simulations did not employ any restraints or tension . Hence the thinning of the membrane in DMPC was enough to induce the formation of kinks similar to simulations in POPC with tension , independent of other major structural changes and pore opening .
The aim of this study was to model the open pore structure of the MscL protein and investigate the structural changes involved in gating using restrained CG MD simulations . The simulations were carried out using a homology model of Eco-MscL [20] , [21] , [38] instead of the crystal structure of Tb-MscL as the restraints in the simulations are based on data from FRET and EPR experiments that were carried out on Eco-McsL . To model the open pore we integrated solvent accessibility and inter-subunit distances from EPR and FRET experiments into a CG model of the closed state of MscL and carried out a series of simulations using different combinations of restraints and membrane tension . The use of a CG model allowed us to run multiple simulations in the range and observe structural changes not seen in single shorter simulations . Furthermore , combining the CG model with restraints allowed for much greater conformational sampling than has previously been possible . While previous simulations used either large tensions or radial forces to induce opening of the pore within the available simulation time , here the opening is achieved through the use of restraints based on experimental data . While this also involves the application of external forces that move the protein towards a predetermined goal and which can distort the protein if the biasing forces are applied too quickly , we hope that these forces at least lead to a conformation that is consistent with the known structural data from experiments . In addition , the long simulation time allowed us to slowly introduce the restraints , apply tension that is of the same magnitude as the physiological tension of MscL as well as keep the force constants of all restrains much lower than in shorter , atomistic simulations . Furthermore , we carried out multiple simulations with different combinations of restraints , testing the sensitivity of the final structure to the restraints used . However , as a consequence of using experimental restraints to direct the evolution of the system through conformational space towards an open pore structure , the simulations should not be used to study intermediate structures or analyse the order of structural changes during gating . From the restrained simulations we produced a set of plausible open pore structures that showed some common features . The pore radius of the open pore structures ranged from 13 . 7 Å to 15 . 8 Å , in good agreement with results from experimental [7] , [8] , [17] , [20] and other simulation studies [9] , [30] . Our results suggest that a stable open pore can be achieved with less structural changes than previously reported . The open pore structure showed no significant rotation of the TM helices and adjacent TM1 and TM2 helices move together as fairly rigid unit . The open pores structures showed deviations from the five-fold symmetry of the open state , which was also observed in previous CG simulations [30] that suggested an asymmetric gating mechanism . Analysis of the final structure from all simulations showed that the N-terminal sits at the cytoplasmic lipid interface making it highly unlikely to act as a second gate . Our results also support that the C-terminal remains as an intact helical bundle and shows no outward motion away from the central axis running down the pore centre . In addition , the outward facing conformation of the hydrophobic residues in the C-terminal are likely to be an artefact of the crystallisation . The simulations did not contain any solvent restraints and only a single distance restraint in the C-terminal . Furthermore , we observed that the C-terminal bundle is free to move and , in comparison to the crystal structure , moves upwards to contact the TM section of the protein . While it could be an artefact of the simulations , this upward movement was observed in all simulations independent of tension and pore opening . It is possible that the low position of the C-terminal in the crystal structure is another artefact of crystal packing and the long simulations allowed this domain to move into a more natural position . On the other hand , this upward movement might be of functional relevance . The upward movement of the C-terminal reduces the size of the entrance to the pore and is possibly related with the hypothesis that the domain acts as a size exclusion filter [59] . It is also possible that the upward motion reveals interactions between the C-terminal and the TM helices that have not been observed previously . These interactions could stabilise either the open or closed state , depending if the pore is closed or open when the contact arises . In the case of the open state , the contact might prevent the closing of the pore by blocking the movement of the TM helices which might be related to the increased frequency of sub-conducting states in 110–136 mutants of MscL [13] , [59] . A blocking of the open gate by the C-terminal was also observed in recent coarse grained simulations of MscL [30] . On the contrary , recent results from patch-clamp experiments [54] showed that Sa-MscL C26 reconstituted in liposomes is more stable in the open state than WT channel . The authors suggested that this is because the C-terminal stabilises the closed state and hence the truncated mutant is more likely to sample the open state . The interactions between the C-terminal and the TM helices resulting for the upward movement of the C-terminal could be related to this effect . One of the most interesting result from our simulations is related to the structural changes of the periplasmic loop . Simulations with membrane thinning , induced by tension or the use of short-chain lipids , showed a significant structural change in the periplasmic loop and an outward motion of the entire loop away from the pore centre such that it lies close to the membrane surface . In contrast , in simulations without membrane thinning the loop remains around the pore entrance on the extracellular side where it tends to partially block this end of the pore . The reason for this structural re-arrangement of the periplasmic loop appears to be the formation of a kink in the upper part of TM1 which allows the upper end of TM1 and the start of the periplasmic loop to move outwards towards the surface of the thinned membrane . Both the outward motion and the formation of the kink are independent of pore opening . Based on these observations we propose that membrane thinning induces a kink at the top of TM1 which results an in outward motion of the periplasmic loop . These previously unobserved structural changes might have important implications as they reveal a new mechanism of sensing and transducing tension by or to the extracellular domains of the protein . This is an alternative to the conventional response to hydrophobic mismatch in which the protein senses membrane thinning by tilting of the TM helices in an attempt to avoid exposure of hydrophobic residues at both ends of the TM helices to aqueous environment . This new mechanism of tension sensing by helix kinking can act in concert with helix tilting and might contribute to the still unanswered question of how MS channels sense tension in the bilayer . Our observations are consistent with the results from a number of other studies . MD simulations of MscL in a curved bilayer showed spontaneous restructuring of the periplasmic loop leading to contacts between the loop and lipid head groups [62] . This contact has also been documented experimentally [38] , [63] . EPR experiments showed that , on average , the lipid accessibility of residues in the periplasmic loop increases in the open channel [38] . Cleaving of the loop [56] or mutations [13] in this domain altered the channel's sensitivity to tension . Finally , patch-clamping experiments [64] on WT-MscL and a series of Eco-MscL mutants reported that mutations of residues in the upper part of TM1 show reduced or complete loss of channel activity . We can propose two different but related functional consequences of the observed structural change of the periplasmic loop . First , the formation of the kink and the outward motion of the periplasmic loop might prevent unwanted or random opening in the absence of tension as it makes it harder for TM helices to move and open the gate unless the periplasmic loop has first extended into the outward position . This proposition is in agreement with experiments that show that the energy of activation is lower in MscL reconstituted in phosphatylcholine ( PC ) containing 16 carbons and higher in PC with 20 carbons compared to the commonly used POPC with 18 carbons [19] . The results also agree with a recent study showing that introducing lipids of shorter chain lengths to WT-MscL reconstituted in azolectin reduces the tension required to open while adding cholesterol to stiffen the membrane causes an increase in gating tension [64] . It has been suggested previously that the periplasmic loop acts as a “spring that resists the opening of the channel" [56] and our results relate this functional role of the periplasmic loop to specific structural rearrangements of that domain . If the loop resists opening it is likely that this domain is the first to undergo structural changes during gating , as previously suggested [62] . Secondly , in the case of an unwanted opening of the gate in the absence of tension or membrane thinning , the partial blocking of the extracellular end of the pore by the periplasmic loop might prevent the excessive loss of solutes . The open pore models produced in this study provided us with valuable insight into the structural changes involved in MscL gating and they form a good starting point for further refinement of the open pore model . By running and comparing two simulations for almost all protocols and running simulations with reduced data sets we ensured a basic level of reproducibility . Nevertheless , to further refine the open pore models it would be useful to run multiple simulations using the protocols that proved most effective in producing open pores and perform cluster analysis on a number of open pore structures . A set of representative open pore structures can then be used to carry out atomistic simulations based on the CG model to obtain an atomistic open pore model . The methodology of combining CG simulations with restraints based on experimental data has enabled us to conduct an extensive analysis of the structural changes of the protein during gating and to find some previously unreported structural changes . This approach is not limited to modelling mechanosensitive ion channels and may be useful to investigate the structure and function of other membrane proteins . The determination of high resolution structures for membrane proteins is challenging , as demonstrated by the low number of membrane protein structures in comparison to soluble proteins in the Protein Databank . Furthermore , for many membrane proteins the available high-resolution structure only represents one conformation but in most ion channels and membrane transport proteins the structure of multiple conformations is needed to elucidate the structure-function relationship . MD simulations that use low resolution , structural data can be very useful to refine the structure of membrane proteins in a given conformation or to model the remaining conformational states . While the integration of geometry restraints such as distances , angles and relative positions is implemented in most MD simulation packages the use of solvent accessibility is not as straight-forward as these restraints require changes to the non-bonded interactions in the force field . Nevertheless , restrained MD simulations are a promising tool to model the structure and conformational changes in membrane proteins .
|
Cells in biological organisms have to be able to respond to mechanical forces during processes such as touch , hearing , pain sensation and tissue growth . One way this is achieved is through mechanosensitive ion channels , membrane embedded proteins that initiate electrical signalling upon tension within the cell or cell membrane . The malfunction of such channels is also associated with a range of diseases including muscular dystrophy and cardiac arrhythmia . In this manuscript , we study in detail the mechanosensitive channel of large conductance ( MscL ) from bacteria , a model system in which to understand the principles of mechanosensation . Despite many years of investigative work the details of how the protein senses tension in the surrounding membrane remain unknown . By combining structural data from experiments with computer simulation we are able to model the open channel structure of the protein and report previously unobserved structural changes that might present a new mechanism of sensing tension . The methods developed in this paper are not limited to the study of mechanosensitive ion channels and may be useful in understanding the structure and function of other membrane proteins .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"signal",
"transduction",
"molecular",
"cell",
"biology",
"membrane",
"receptor",
"signaling",
"biology",
"biophysics",
"simulations",
"biophysics"
] |
2012
|
Structural Investigation of MscL Gating Using Experimental Data and Coarse Grained MD Simulations
|
Plesiosaurians are an extinct group of highly derived Mesozoic marine reptiles with a global distribution that spans 135 million years from the Early Jurassic to the Late Cretaceous . During their long evolutionary history they maintained a unique body plan with two pairs of large wing-like flippers , but their locomotion has been a topic of debate for almost 200 years . Key areas of controversy have concerned the most efficient biologically possible limb stroke , e . g . whether it consisted of rowing , underwater flight , or modified underwater flight , and how the four limbs moved in relation to each other: did they move in or out of phase ? Previous studies have investigated plesiosaur swimming using a variety of methods , including skeletal analysis , human swimmers , and robotics . We adopt a novel approach using a digital , three-dimensional , articulated , free-swimming plesiosaur in a simulated fluid . We generated a large number of simulations under various joint degrees of freedom to investigate how the locomotory repertoire changes under different parameters . Within the biologically possible range of limb motion , the simulated plesiosaur swims primarily with its forelimbs using an unmodified underwater flight stroke , essentially the same as turtles and penguins . In contrast , the hindlimbs provide relatively weak thrust in all simulations . We conclude that plesiosaurs were forelimb-dominated swimmers that used their hind limbs mainly for maneuverability and stability .
Plesiosaurians ( = plesiosaurs ) are an extinct group of highly derived predatory marine reptiles with a global distribution that spans 135 million years from the Early Jurassic to the Late Cretaceous . During their long evolutionary history [1] , plesiosaurs maintained a unique body plan with two pairs of large wing-like flippers—a unique adaptation in the animal Kingdom [2 , 3 , 4 , 5 , 6 , 7 , 8 , 9] . Although plesiosaurs were a key component of Mesozoic marine ecosystems , there are no extant ‘four-winged’ analogues to provide insights into their behavior or ecology , and their locomotion has remained a topic of debate since the first complete plesiosaur skeleton was described in 1824 [10] . Key areas of controversy have concerned the most efficient biologically possible limb stroke , and how the four limbs moved in relation to each other . Previous studies of plesiosaur locomotion have endorsed a variety of strokes and gaits . Stroke hypotheses include a rowing stroke akin to the oar of a boat [4] , a flight stroke [5 , 3] akin to penguins and turtles [11 , 12] , and a modified flight stroke [6] akin to sea lions [13] . Study of plesiosaurian musculature does not rule out either rowing or flight strokes [14] . Gait hypotheses include synchronous motion with all four limbs moving in phase [15] , asynchronous motion with the forelimbs and hindlimbs out of phase [12 , 7] , and semi-synchronous motion [3] . Some authors have proposed that the hindlimbs provided most of the propulsion [16] , whereas others suggest that the forelimbs provided the majority of thrust [8] . The question has been approached experimentally using robotics [17] and human swimmers with fabricated paddles [3] . These studies , although informative , are limited because they do not deal with accurate representations of the plesiosaur form . There is therefore still no consensus on how plesiosaurs swam , especially how they moved all four limbs relative to each other . Our approach uses computer simulation to address the question of how plesiosaurs swam using a three-dimensional plesiosaur model in a simulated fluid . The computational model explores a given range of joint motion to discover the swimming stroke and gait that propels the creature forward the greatest distance . There are two main advantages to using computer simulation to investigate swimming motions of plesiosaurs . First , we can construct a digital representation of the body that accurately matches the known body and limb shapes of a particular species . Second , we can run thousands of trials with different strokes and gaits to explore the space of possible swimming motions . We use an optimization method to search for the highest quality motion for a given range of joint angles . Computer simulation has been used to investigate the motions of several types of modern-day swimmers , including fish [18–21] , tadpoles [22] , and copepods [23] . To our knowledge , our work is the first use of computer simulation to study the swimming of plesiosaurs .
We constructed a life-sized plesiosaur model based on Meyerasaurus victor , a small ( 3 . 35 meters long ) taxon from the Lower Jurassic of Germany , because it is known from an almost complete articulated skeleton ( SMNS 12478 ) with all four limbs preserved in their entirety [24] ( Fig 1A ) . In addition , Meyerasaurus possesses a generalized morphotype among plesiosaurs , with a moderately long neck , so it can be considered representative of the clade Plesiosauria as a whole , which contains long- and short-necked morphotypes [2] . The shape and proportions of the model were based directly on the skeleton , with three-dimensional data and soft tissues ( e . g . muscles , cartilage , integument ) reconstructed based on evidence from other taxa . Two dimensional cross-sectional data for the body was estimated by tracing around the skeleton in the horizontal plane . Since the holotype of Meyerasaurus is dorsoventrally compressed , information from other plesiosaur specimens was used to estimate cross sections in the vertical plane [25] ( Fig 1B ) . Transverse cross-sections of the limbs were derived from the three-dimensionally preserved propodials of the closely related Rhomaleosaurus thorntoni [26] . The restored limbs are cambered hydrofoils in section with a narrower postaxial trailing edge . The postaxial edge of the limb was extended beyond the osteological anatomy to reflect fossil evidence for a soft tissue trailing edge in this region in Seeleysaurus guilelmiimperatoris [27] and Hydrorion brachypterygius [28] . The tail of the model was reconstructed with a short dorsally expanded mediolaterally compressed fin based on evidence from several plesiosaurian taxa , including the sister taxon to Meyerasaurus: Rhomaleosaurus [29] . The virtual model was constructed in Maya , a widely used CAD tool . First , the two-dimensional cross sections were aligned in the horizontal and vertical planes ( Fig 1B ) . Second , the model was constructed using these sections as a reference . Since our grid-based fluid simulator cannot detect features under 67 mm thick , the thinnest parts of the anatomy , such as the trailing edge of the limbs , were artificially dorsoventrally thickened . For simplicity , the density of the plesiosaur model is identical to that of the fluid ( i . e . it is neutrally buoyant ) , and therefore our simulations do not take into account possible variation in buoyancy along the body of the animal due to air-filled lungs , or gastroliths [30] . The life size final constructed plesiosaur model is 3 . 35 meters long from head to tail ( Fig 1C ) . An alternative bulkier Meyerasaurus model with 50% greater soft-tissue mass around the base of the limbs was also created to test the effect of a bulkier body outline . Our physics simulator ( based on previous methods [31] ) represents the plesiosaur as a collection of rigid body parts that meet at points of articulation . Specifically , we model the body ( torso , neck , and tail ) as one rigid component and the four limbs as additional rigid parts . In life , the plesiosaur torso was a rigid structure , since a sturdy trunk is a prerequisite for purely paraxial underwater locomotion . The neck and tail in plesiosaurs were flexible in life to variable degrees [32] , so they may have had a relatively minor role in propulsive locomotion . However , since our focus is on the question of limb-based propulsion in a four-winged paraxial swimmer , we kept the neck and tail immobile in the simulation to allow us to focus solely on the movement of the limbs . Each limb is joined to the body by a three-degree-of-freedom joint ( Fig 2 ) . These four joints can be actuated internally to generate motion . Each swimming motion is represented as a sinusoidal function at each joint degree of freedom , and the actuators track these desired motions by applying torques at the joints . In turn , the motion of the body and limbs affects the simulated fluid that surrounds the animal . Our simulator resolves the motion of the animal body and the fluid simultaneously , so that the final motion is due to the interaction between the body and the fluid . This is in contrast to studio-created computer animation , where the motion of the animal through the fluid is scripted by an artist , and may not obey the governing laws of physics . To study the forward swimming motion across a wide range of periodic swimming strokes of the limbs , we require a motion representation that is expressive but that is also biologically plausible . We decouple the degrees of freedom at a joint into a dorsal/ventral component , an anterior/posterior component , and a pronate/supinate component ( the rotational angle of the limb ) . We specify the limb motions of the plesiosaur by describing sinusoidal patterns for each of the three degrees of freedom at a given joint . The limb motion for a given component is specified by three values: the minimum and maximum value of the sinusoid , and the phase of the sinusoid . We use the same frequency ( 0 . 5 Hz ) for all of the sinusoids across all of the limbs . Since the motion for each degree of freedom is given by three values ( maximum and minimum range , and phase ) , nine numbers fully describe the motion of a single limb . We assume that the left and right limbs move in synchrony while the animal is swimming straight , as is the case for penguins , sea lions , marine turtles , and nothosaurs [33] . However , we specifically allow the front and back limbs to follow different patterns of motion: the minimum and maximum angles , and the phase of the sinusoid for the front and back limbs can be set differently . This allows us to test , for instance , the possibility that the front and back limbs move together or with offset phases . To specify both front and back limb sinusoidal motion , we require a total of 18 parameter values . Table 1 shows the optimized minimum/maximum ranges for all limbs , as well as the average travelling velocity and distance traveled for all of our experiments . Although using sinusoidal motions of various angle ranges and phases gives a wide range of possible swimming strokes , the motions of modern-day swimming animals depart from pure sinusoidal motion in at least two ways . Animals such as penguins that use an underwater flight stroke [12] hold the angle of rotation steady during the downstroke , quickly rotate the limb at the bottom of the stroke , and then hold the angle of rotation steady again during the upstroke . We allow for this possibility by using one additional degree of freedom for the rotation of a limb that specifies the duration of a motionless interval during which the limb maintains a zero rotational velocity . This interval can be set to zero , which indicates pure sinusoidal motion , or it can be non-zero to hold the angle of rotation steady through a portion of the stroke ( Fig 3A ) . This gives us two additional motion parameters , one for the forelimbs and one for the hindlimbs . We also allow for the possibility that the animal’s downstroke and upstroke take different amounts of time ( Fig 3B ) . This is in recognition of the observation that plesiosaurs may have had stronger musculature governing the downstroke of their limbs [7 , 9 , 14] . We add one more degree of freedom for each sinusoid to specify its degree of time asymmetry . This gives us six additional motion degrees of freedom , bringing the total number of parameters that describe a periodic swimming motion to 26 . Table 2 shows the time asymmetry as well as motionless portion in pronate/supinate direction optimized in all of our experiments . Plesiosaur limbs contain a single mobile joint located between the propodial and the girdle: the glenohumeral joint in the forelimb and the acetabulum-femoral joint in the hindlimb . The articulation points for these joints in the model are located in the anatomically-correct positions ( Fig 2 ) . Neutral limb positions were derived from existing estimates for Plesiosaurus sp . [3] . In the forelimb the neutral position is -15 degrees from the horizontal and -16 degrees from a line drawn perpendicular to the long axis of the body . In the hindlimb the neutral position is -30 degrees from horizontal and -27 degrees from a line drawn perpendicular to the long axis of the body ( Fig 2 ) . In forelimb only and hindlimb only optimizations , the static limbs are locked into these neutral positions and the active limbs are initiated in the fully abducted positions . While the available range of rotation along the long axis of the limb was identical in all limbs and optimizations: up to 30 degrees supination and 45 degrees pronation ( Fig 2C ) , we tested three different ranges of motions in the dorsal/ventral and anterior/posterior directions ( Fig 2A–B , D ) . The degree of freedom in the joints of living plesiosaurs was dependent on the extent and thickness of the cartilage that covered the head of the propodials and lined the glenoid and acetabulum . To account for possible differences in cartilage thickness and to investigate stroke efficiency and gait under different specified parameters , we performed optimizations under three different ranges of joint freedom: ‘narrow’ , ‘medium’ , and ‘wide’ ( Fig 2 ) . The narrow range was taken directly from conservative estimates of degrees of freedom in Plesiosaurus sp . [3] . The wide range provides an expanded degree of freedom that possibly exceeds the biologically possible range in the living animal , and the medium range represents a realistic compromise between the conservative narrow range and generous wide range . In life , rotation of these joints was complicated , but for simplicity , they pivot around a single point in the model . Although there are no additional mobile joints in plesiosaur limbs , cartilage and tendons would have allowed dorsoventral flexibility and twisting along the long axis of the limb [5] . This could have resulted in an increased range of motion at the tip of the limb compared to the range of motion at the joint , and may have affected water flow and minimized drag . One limitation of our method is that it does not currently replicate flexibility of this kind—the limbs are rigid elements in our simulations . We accounted for this , in part , by providing simulations with wider ranges . To fully address limb flexibility would require an entirely different simulator that uses the finite element method to allow limb deformations . This would also require a different approach to solid/fluid coupling in the simulator . Since a single swimming motion requires the specification of many different sinusoidal parameters ( 26 , as described above ) , we use numerical optimization to explore the range of possible plesiosaur swimming motions . Specifically , we use the sample-based method called Covariance Matrix Adaptation [34] which has been used to investigate eel swimming [19] and animal walking gaits [35] . Our optimization process runs several thousand different simulations with different joint motions , narrowing in on the set of motions that produces the fastest swimming motion . Note that for such a large parameter space , CMA is not guaranteed to find the global optimum . However , we observed only small variations in the final results of different CMA runs with the same parameter settings . Fig 4 shows the resulting optimal swimming motions for each of the three joint ranges , where both the front and back limbs move in a manner that best propels the plesiosaur forward . The white paths in the figure show the distal tip traces . ( See the two accompanying S1 and S2 Videos for the detailed motions . ) The best strokes for the forelimbs in both the narrow and medium range is an underwater flight motion , in which the limbs move primarily in the dorsoventral direction , and only rotate at the top and bottom of the stroke ( Fig 4A and 4B ) . This pure flight stroke has been suggested as the most likely swimming stroke for plesiosaurs based on several anatomical lines of evidence [5] . In contrast , our optimization determined that the best forelimb stroke for the large range of joint motion is a modified U-shaped flying stroke ( Fig 4C ) . This is similar to a flight stroke , but with more posterior motion during the power stroke and with a partially feathered recovery , and has also been proposed for plesiosaurs [6 , 15 , 8] . Note , however , that our optimizations suggest that this modified flight stroke is only plausible under the most liberal of joint range assumptions , and such a wide range of motion is considered biologically impossible [3] . None of our optimizations produced a substantial rowing stroke , as had been suggested by early researchers [4] . S1 Video shows the highest quality swimming motions from each of our optimization runs . To produce each of these video segments , we re-computed the simulation that corresponded to the highest quality motion sample that was found during the given optimization run . There are nine motion clips in this video , corresponding to the wide , medium and narrow ranges of limb joints , with motion from both pairs of limbs , just the forelimbs , and just the hindlimbs . As in the optimization runs , the model plesiosaur is initially at rest and then begins to move . During the first stroke , the model plesiosaur sometimes turns upwards , but then moves straight during subsequent strokes . For this reason we do not include the motion of the first stroke in our quality assessment ( described in detail later ) . The simulations used to make these videos not only provide the motion of the plesiosaur model , but also give us the velocity field for the simulated fluid at each simulation time step . This allows us to show not only the plesiosaur motion , but also the accompanying motion of the water . In each video clip , we show the motion of the fluid using particle traces ( particle trajectories over the last few time steps ) from randomly positioned particles in the virtual fluid . These massless particles are passively advected through the fluid . The particle traces are drawn only at locations with large vorticity ( greater than 1 . 5 s-1 ) to concentrate them at regions of interest . S2 Video shows the plesiosaur limb stroke motions from these same nine simulations , and provide traces of the tips of each moving limb . These motion clips allow a clearer picture of the limb motions relative to the body . In these video clips , the camera moves together with the body so that the body appears to be stationary . Two views are provided for each motion: a lateral view and a posterolateral view . For clarity , particle traces have been omitted in these video clips . Fig 5 shows the best swimming speeds for each joint range . These swimming speeds are similar to prior estimates of an optimum swimming speed of 0 . 48 m/s for Meyerasaurus [36] . Because all of the strokes have a two second period , it is to be expected that the larger ranges of motion result in a faster speed , so no conclusions should be drawn from the relative speeds between the three different ranges of joint motion . There are several potential relative motions between the forelimbs and hindlimbs during swimming . One proposal is that asynchronous motion is the most likely way to produce continuous forward motion [9] . Synchronous ( and semi-synchronous ) motion has been deemed more likely by other researchers [15 , 8 , 3] . Our optimizations provided no clear answer to this question , which is significant in itself . In the narrow range the limbs move asynchronously , in the medium range they move semi-synchronously , and in the wide range they move synchronously . In order to deduce the separate contributions of the forelimbs and hindlimbs to propulsion we performed optimizations with forelimb-only motion and hindlimb-only motion . The results were strikingly consistent across all three joint ranges . The forelimb-only strokes from optimization were roughly as fast as the best gaits resulting from optimizations using all four limbs ( Fig 5 ) . In contrast , the hindlimb-only strokes from optimization provide a much slower forward motion , even for the widest joint range . This inability of the hindlimbs to generate thrust explains the lack of consistency in the relative motions of the forelimbs and hindlimbs in our optimization results . It does not matter whether the plesiosaur moves its hindlimbs in or out of phase with the forelimbs , since neither strategy will contribute substantially to forward motion . Our optimization results imply that plesiosaurs were forelimb-dominated swimmers , and that the hindlimbs contributed little to their forward motion . This is consistent with trace-fossil evidence for forelimb-dominated locomotion in nothosaurs [33] . It also corroborates other studies that concluded that the hindlimbs were used primarily for steering and stabilization during swimming [8] , and that two-flippered gaits serve well for low-cost cruising [17] . The plesiosaur hindlimbs , despite their wing-like shape and large size , played a diverse role in locomotion , but a relatively minor role in propulsion . They supplemented the forelimbs , which were the primary propulsive organs , by enhancing maneuverability and stability , possibly in conjunction with the tail [29] . To investigate the effect of small body changes to swimming speed , we constructed an alternative version of the Meyerasaurus model with 50% greater ‘muscle mass’ around the bases of the limbs . Fig 6 shows the original and modified body shapes . We ran three simulations with the bulkier model using the limb motion parameters taken from the results of the optimizations with the ‘slimmer’ model ( medium—all limbs , medium—forelimbs only , medium—hindlimbs only ) . Because it is unlikely that substantially different limb motion parameters would give a more effective swimming motion , we did not use optimization to search for new motion parameters . Table 3 shows a comparison of the original ( slim ) and modified ( muscle bulk ) model , and for each simulation case gives the swimming distance and vertical deviation from the horizontal . The results showed that the bulkier model was marginally slower in each case , probably due to the increased drag . With both models , however , the contribution of the hindlimbs to locomotion is small . This test shows that manipulation of the fine details of the model does not have a major impact on the swimming speed . Larger modifications to the body shape could have a more substantial effect , and is an area for future work . With larger changes to the body shape , it would be necessary to use optimization to search for the most effective limb motions . There is great variation in head and neck proportions , flipper aspect ratios and relative limb proportions within the clade [37] . For example , in Meyerasaurus and other Lower Jurassic plesiosaurs the fore- and hindlimbs are subequal in size , whereas in derived pliosaurids the hindlimbs are larger than the forelimbs , and in derived plesiosauroids the forelimbs are largest . Furthermore , Meyerasaurus has high aspect ratio flippers , whereas some genera ( e . g . Cryptoclidus ) have low aspect ratio flippers [37] . Head and neck proportions also vary considerably within Plesiosauria . ‘Plesiosauromorph’ taxa possess a long neck and small head , while ‘pliosauromorph’ taxa possess a short neck and large head [2] . The Meyerasaurus model used in our experiments possesses an intermediate morphology with a moderately long neck and moderately large skull , so the results represent a generalized plesiosaur morphotype . Variations in head and neck size could shift the center of mass relative to that of our model , possibly affecting the relative contributions of the forelimbs and hindlimbs during swimming . A longer neck and/or a larger head would increase drag and slow the forward motion in a manner similar to our test with increased limb muscle bulk . It is also possible that the tail contributed to forward thrust during swimming . Although we conservatively extend our general conclusions for Meyerasaurus to all plesiosaurs , our method could be sensitive to substantial changes in bodily proportions , and different plesiosaurs may have swam in different ways . Further experimentation is therefore required to assess how bodily variation might affect locomotion in other types of plesiosaurs .
We simulate the fluid dynamics based on Euler equation ( Navier-Stokes equation without the viscosity term ) , since the viscosity of water is negligibly small . ∂u∂t+u∙∇u+1ρ∇p = g ∇∙u = 0 where u is the velocity of fluids , ρ is the density , p is pressure and g is gravity . We simulate the fluid using a staggered MAC grid [38] based solver . We use BFECC [39] to integrate the advection term and use explicit Euler scheme to integrate the gravity force . We solve the incompressibility term along with the two-way coupling between fluids and solids ( See Section 2 . 3 for details ) . The dynamics of articulated rigid body systems ( the plesiosaur ) is described by the equations of motion in the generalized coordinate: M ( q ) q¨+C ( q , q˙ ) = τint+τext where q , q˙ and q¨ are positions , velocities and accelerations in the generalized coordinates respectively , M ( q ) is the mass matrix , C is the Coriolis and Centrifugal force , τext are the external generalized forces , including the fluid pressure and gravity , and τint are internal torques exerted by the actuated joints . Given a reference swimming stroke , we use Stable Proportional-Derivative controllers [40] to compute the internal joint torques τint to closely track the stroke . We build our two-way coupling solver based on Tan’s two-step procedure [31] . In the first step , both the fluid and the solid are simulated independently . In the second step , a linear system of the pressure field is formulated , taking into account both the incompressibility of the fluid and the dynamics of the solid due to the fluid pressure . Our simulator needs to voxelize the plesiosaur onto the grid at each time step , which marks the grid cells that are inside the animal as solid cells and the remaining as fluid cells . Due to the large computational requirement of the two-way coupling simulation , we use relatively coarse grid resolution ( 100x80x60 to represent a 6 . 6 by 5 . 28 by 3 . 96 m3 region of water ) , which will cause stair-step boundary artifacts at the interface between the fluid and the animal . This can result in seemingly higher viscosity near the animal’s body . For this reason , we incorporate the variational approach [41] into our two-way coupling simulation . This enables us to perform simulation with sub-grid resolution and gain smoother results at the solid-fluid interface with almost negligible additional cost ( Fig 7 ) . The traditional fluid simulation enforces the incompressibility by solving the following Poisson equation , with Neumann boundary condition at fluid-solid faces and Dirichlet boundary condition at free surface . Following Tan’s [31] derivation , at coupled faces ( solid-fluid interface ) , the total generalized force exerted by the fluid pressure to articulated rigid-bodies is an integral of pressure over the surface of the plesiosaur: τtotal = ∬Sτ = ∬SJTpn = ∑i = 1kJiT ( Δx ) 2pini where Ji is the Jacobian matrix at the ith fluid-surface interface , Δx is the length of a single grid cell , p is the pressure inside the fluid cell and n is the solid surface normal . Similar to Batty’s variational approach [41] , we apply divergence theorem to convert the surface integral to the volume integral . We dropped the second term in the integral because the divergence of J is always zero for rigid body motions , and since p is zero everywhere inside solids and equals to the pressure in fluid cell at fluid-solid interface , we can discretize the above integral to be τtotal = ∭VJT ( ∇p ) = ∑i = 1kvi ( Δx ) 3JiT ( piΔx ) ni = ∑i = 1kvi ( Δx ) 2JiTpini where vi is fluid volume fraction at the coupled interface . The acceleration in the generalized coordinate of the coupled interface is q¨ = M-1τtotal where M is the mass matrix of the articulated rigid-body . Therefore the acceleration for the coupled interface in Cartesian space would be a = nT ( J˙q˙ ) = nT ( Jq¨+J˙q˙ ) = nT ( JM−1∭vJT ( ∇p ) +J˙q˙ ) = nT ( ∑ ki = 1vi ( Δx ) 2JM−1JiTpini+J˙q˙ ) Intuitively , the variational approach adds fluid volume fraction information vi to the original coupled linear equation [31] , making use of volume fraction weighted stencils , instead of being discrete values of zero or one . To compute the fluid volume fraction vi , we perform inside-outside tests of the plesiosaur model at the center of every cell of a higher resolution grid . We shoot a ray at the cell center and count the number of intersections ( parity ) between the ray and the model . Since our model is watertight , if the intersection count is even , the cell is outside of the body of the plesiosaur , and thus occupied by the fluid . Otherwise , the cell is inside the creature , and thus it is solid . We implemented a Surface Area Heuristic Kd-Tree [42] to accelerate ray-model intersection performance . In our implementation , we use a sub-grid resolution of 3x3x3 to compute the fluid volume fraction , achieving 1/27 volume fraction precision in the pressure solve . Finally , we construct a sparse symmetric positive linear system with the following equations at the fluid-solid faces , D ( ∑i = 1kvi ( Δx ) 2JM-1JiTpini ) = D ( u*Δt-J˙q˙ ) where D is discretized volume-weighted divergence operator and u* is the intermediate velocity field before enforcing incompressibility . We use normal stencils of Laplacian and Divergence operators at fluid-fluid faces . We apply Preconditioned Conjugate Gradient solver to solve the system , after which we project the fluid velocity with the following equation u = u*-Δt∇pρ where u* and u represents the fluid velocity before and after projection . As a result , we simultaneously enforce incompressibility of the fluids , compute dynamics of the solids , satisfy boundary velocity constraints and achieve smoother fluid flow at the solids-fluids interface . Note that our simulated water does not behave exactly the same as real water . Two common issues in numerical fluid simulations , numerical viscosity and voxelization artifacts , affect the accuracy of our two-way coupled simulator . The numerical errors in the simulation can make the simulated fluid more viscous than its real counterpart . This is called numerical viscosity [43] . Even though we dropped the viscosity term from the Navier-Stokes equations and used a higher-order integration scheme , BFECC [39] , the numerical viscosity cannot be eliminated entirely . Voxelization artifacts are caused by converting the rigid bodies ( the plesiosaur ) into a regular grid of cells for fluid simulation ( Fig 8 ) . The streamlined body shape of the animal is lost in this voxelization process . As a result , the animal may swim slower than in real life due to the increased form drag [44] . The variational framework [41] that we used allows us to simulate at sub-grid resolutions , and thus ameliorates this problem . However , this issue cannot be completely eliminated . Given a plausible range of motions for the front and back limbs , we wish to find the swimming motion that propels the animal the farthest distance in a given amount of time . We also want to favor straight swimming . Since a single swimming motion requires the specification of 26 parameters we formulate our question in terms of optimization . Let us call a specific set of swimming motion parameters a motion sample , and we can think each such motion sample s as a single point in a 26 dimensional space . Given a motion sample , the plesiosaur/fluid simulator generates a swimming motion through the hydrodynamic interaction . We formulate the quality function q ( s ) that favors a swimming stroke that leads to a longer swimming distance , and we penalize deviation from swimming straight . We calculate the swimming distance as the displacement of the plesiosaur’s center of mass ( COM ) in one swimming cycle projected onto its initial heading direction . The deviation from straight swimming is decomposed into two components: Directional deviation is measured by the displacement of COM that is perpendicular to the initial heading direction . Orientation deviation is measured simply as the orientation change in one swimming cycle . Since the simulation starts with a static plesiosaur submerged in motionless water , we evaluate the objective function only after the first stroke cycle is completed , when the plesiosaur has reached a steady speed . Forelimb-only and all limb optimizations were initiated with the front limbs in the fully abducted position and the hindlimbs in the neutral position . Hindlimb-only optimizations were initiated with the hindlimbs in the fully abducted position and the forelimbs in the neutral position . This forced all of the gait samples to begin with a downstroke , so that no gait was penalized for a slow start that is mid-way through a stroke . We used Covariance Matrix Adaptation ( CMA ) [34 , 45] to search for the sample s that gives us the highest quality . CMA is a sample-based approach to optimization that uses a Gaussian distribution to guide its selection of motion samples to test . Each iteration of the CMA draws a new set of samples based on the current distribution , and evaluates the quality of these samples using q ( s ) . The mean and the covariance matrix of the Gaussian are then updated according to a subset of sample with higher quality . As more iterations are taken , the covariance matrix narrows down towards the best quality sample . In our case , the best sample is the swimming motion parameters that move the simulated plesiosaur the farthest in a straight line through the water . To converge to the optimal motion , CMA requires many samples to be evaluated . Each sample evaluation requires a full simulation of the specified motion , and each such plesiosaur/fluid simulation requires roughly one hour of computation . We used 31 samples per iteration of CMA , and we found that a typical optimization run converged in about 70 iterations , after which the quality of the motion samples does not improve . This means that a full optimization requires more than two thousand swimming simulations . We made use of a compute cluster with 32 nodes , so that the computations for a single iteration were all calculated in parallel on separate nodes . ( One of these computer nodes coordinates the simulations of the other 31 nodes . ) Even with the compute cluster , performing CMA optimization for a single set of joint ranges requires three days to run . Although CMA converges to high quality samples , there is no guarantee that it will find the best possible sample in our 26-dimensional search space . This explains why , for the wide joint range , the best forelimb-only swimmer that was found is slightly faster than the all-limbs swimmer ( Fig 5 ) . To verify that this was due to small variations between optimization runs , we ran a total of three optimization runs for each of these two cases for the wide joint range . The three optimization runs for all-limbs gave speeds of 0 . 745 , 0 . 75 , and 0 . 785 ( m/s ) , and the three runs for forelimbs-only resulted in speeds of 0 . 745 , 0 . 80 , and 0 . 84 . Note that these ranges overlap . Fig 5 reports the fastest speeds from each of these sets of three optimization runs .
|
Plesiosaurs are an extinct group of Mesozoic marine reptiles with a global distribution that spans 135 million years . They maintained a unique body plan with two pairs of large wing-like flippers throughout their long evolutionary history , but how plesiosaurs swam has remained a topic of debate for almost 200 years . We address the question of how plesiosaurs swam using a digital , three-dimensional , free-swimming model of a plesiosaur in a simulated fluid . We performed thousands of simulations under different parameters to investigate possible plesiosaur swimming patterns . Our simulations show that the forelimbs provide the majority of thrust , and that the thrust from the hindlimbs is weak . The plesiosaur swims primarily with its forelimbs using an underwater flight stroke , essentially the same as turtles and penguins .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] |
[] |
2015
|
Computer Simulations Imply Forelimb-Dominated Underwater Flight in Plesiosaurs
|
The APOBEC3 proteins form a multigene family of cytidine deaminases with inhibitory activity against viruses and retrotransposons . In contrast to APOBEC3G ( A3G ) , APOBEC3A ( A3A ) has no effect on lentiviruses but dramatically inhibits replication of the parvovirus adeno-associated virus ( AAV ) . To study the contribution of deaminase activity to the antiviral activity of A3A , we performed a comprehensive mutational analysis of A3A . By mutation of non-conserved residues , we found that regions outside of the catalytic active site contribute to both deaminase and antiviral activities . Using A3A point mutants and A3A/A3G chimeras , we show that deaminase activity is not required for inhibition of recombinant AAV production . We also found that deaminase-deficient A3A mutants block replication of both wild-type AAV and the autonomous parvovirus minute virus of mice ( MVM ) . In addition , we identify specific residues of A3A that confer activity against AAV when substituted into A3G . In summary , our results demonstrate that deaminase activity is not necessary for the antiviral activity of A3A against parvoviruses .
Eukaryotes have evolved numerous innate immune defenses against invading pathogens . The apolipoprotein B mRNA-editing catalytic polypeptide-like 3 ( APOBEC3 ) proteins comprise a family of seven cytidine deaminases [1]–[4] that may each form distinct intrinsic barriers to endogenous retrotransposons and invading viruses [5]–[7] . The most characterized member of the family is APOBEC3G ( A3G ) , which restricts Vif-deficient human immunodeficiency virus 1 ( HIV-1 ) [8]–[11] . In addition to HIV-1 , the APOBEC3 proteins inhibit a diverse array of viruses including simian immunodeficiency virus ( SIV ) , human T cell leukemia virus 1 ( HTLV1 ) , murine leukemia virus ( MLV ) , mouse mammary tumor virus ( MMTV ) , and hepatitis B virus ( HBV ) [6] , [7] , [10]–[13] . Interestingly , human APOBEC3A ( A3A ) lacks activity against retroviruses but dramatically inhibits replication of the parvovirus adeno-associated virus ( AAV ) [14] . The molecular mechanisms that govern specificity of APOBEC3 antiviral activity are not yet fully understood . The APOBEC3 proteins share structural and functional features with zinc-dependent deaminases and possess cytidine deaminase activity [1] , [4] . The cytidine deaminase domains ( CDDs ) of APOBEC3 proteins contain an active site with the conserved consensus motif H-X-E-X23–28-P-C-X2–4-C ( where X is any amino acid ) . It has been proposed that the histidine and the two cysteine residues coordinate a Zn2+ ion , while the glutamic acid residue serves an essential role in catalysis as a proton shuttle [15]–[18] . APOBEC3 proteins contain either a single CDD ( A3A , A3C and A3H ) or two tandem CDDs ( A3B , A3D/E , A3F and A3G ) . In the case of A3G , both domains contain an intact active site consensus sequence motif but only CDD2 appears to be catalytically active , while CDD1 is responsible for interaction with the HIV nucleocapsid proteins and packaging [19]–[23] . Sequence alignment shows that A3A is highly homologous to the C-terminus of A3B and A3G proteins [2] . The mechanism of the antiviral activity of APOBEC3 proteins against retroviruses has been studied extensively [5] , [6] , [24] . APOBEC3 proteins are incorporated into virus particles , and encapsidation is mediated via interactions with Gag , viral RNA and cellular RNAs [19] , [22] , [23] , [25]–[32] . Encapsidated A3G is delivered into target cells where it deaminates dC to dU on newly synthesized minus strand cDNAs during the process of reverse transcription [9]–[11] , [33]–[35] . Deaminated genomes can be degraded by the action of the cellular base excision repair machinery [36] , although recent reports suggest that uracil-DNA glycosylase 2 ( UNG2 ) is not required for the A3G antiviral function [37] , [38] . In addition , hypermutated proviruses will contain sequence changes that inactivate the virus by generating alternate splicing , premature translation , and nonfunctional proteins [39] . Whether cytidine deamination is the principal mechanism for the antiviral activity of APOBEC3 proteins remains controversial [39] . APOBEC3 chimeric proteins and catalytically inactive mutants demonstrate that inhibition of HIV-1 can be achieved by A3G that lacks deaminase activity [40]–[42] . A3G can inhibit lentivirus reverse transcriptase ( RT ) and prevent accumulation of reverse transcripts and viral cDNA in target cells in a deamination-independent manner [35] , [42]–[44] . Additionally , subsequent steps of viral integration have been shown to be affected by APOBEC3 proteins [37] , [45] . The idea of deaminase-independent antiviral activity has been challenged by others who have argued that cytosine deamination is required for efficient inhibition of retroelements at low levels of APOBEC3 expression [38] , [46] , [47] . In the case of HBV , the mechanism for inhibition by APOBEC3 proteins is also controversial but has been suggested to be deaminase-independent due to infrequent editing , and may be caused by blocking reverse transcription and expression of HBV antigens [12] , [48] , [49] . Although it lacks activity against retroviruses , A3A is a potent inhibitor of both parvovirus and the human transposon LINE-1 [14] , [50] , [51] . Parvoviruses are small eukaryotic viruses that infect humans and a variety of other animal species [52] . The parvovirus genome consists of a linear single-stranded DNA ( ssDNA ) molecule approximately 4 . 5 kb in length , with hairpin structures at both ends that function as origins for viral DNA replication . The genome contains two major ORFs that encode the nonstructural replication proteins ( NS or Rep ) and the structural capsid proteins ( Cap ) . The family is divided into autonomous parvoviruses and dependoviruses , which require a helper virus for efficient replication and progeny production . The minute virus of mice ( MVM ) is an autonomous parvovirus while adeno-associated virus ( AAV ) is a dependovirus that uses adenovirus as a helper virus . In our previous studies we found that despite the absence of detectable AAV editing , mutations in conserved active site residues of A3A abrogated the antiviral activity against AAV [14] . A3A inhibition of parvovirus provides an attractive system in which to decipher the relative contribution of deamination and deaminase-independent mechanisms to antiviral activity . Unlike the other viruses known to be inhibited by APOBEC3 proteins , the parvoviruses replicate exclusively in the nucleus , do not pass through RNA intermediates , do not have a reverse transcription step in their replication schemes , and use DNA hairpins for priming replication [52] . Moreover , A3A is a simplified system for probing APOBEC3 functions because it has a single CDD and is not restricted to a specific subcellular compartment [2] , [14] . To ascertain whether deaminase activity is required for inhibition of parvovirus replication , and to understand the functional significance of amino acid divergence between A3A and A3G , we analyzed the properties of a panel of A3A point mutants and A3A/A3G chimeras . The proteins were tested for deaminase activity in vitro and for antiviral activity in rAAV production assays . We identify mutants that lack deaminase activity but retain the antiviral effect , supporting the idea of a deamination-independent mechanism . In addition to AAV , we show that A3A inhibits DNA replication of the autonomous parvovirus MVM . Chimeric A3A/A3G proteins generated by exchanging divergent sequences demonstrate loss-of-function for A3A and gain-of-function for A3G . Our mutants also reveal residues in the linker and pseudoactive site domains that are important for deamination , target specificity and antiviral activities of A3A . Together , these studies reveal domains of A3A that are responsible for its distinct antiviral activity and suggest that A3A can inhibit parvoviruses through a mechanism separate from its function as a cytidine deaminase .
We previously showed that A3A antiviral activity was dependent on the integrity of conserved amino acids in the active site that are responsible for proton shuttling and zinc coordination [14] . These results suggested that the catalytic domain of A3A must be intact for antiviral activity against AAV . To extend these studies , we generated additional point mutations at conserved residues ( F75 and F95 ) previously shown to be required for deaminase activity of APOBEC1 [16] , [53] . We also mutated the 99SPC101 residues to AAA ( SPC ) in the active site domain of A3A [50] . The position of these mutants is indicated in Figure 1A . First , we determined the cellular localization of the mutant proteins in transfected U2OS and HeLa cells ( Figure 1B and data not shown ) . Wild-type A3A was located throughout the cell , while A3G was predominantly cytoplasmic ( Figure 1B ) . Most of the A3A mutants displayed cellular localization patterns similar to wild-type protein . The double mutant FF7595LL showed a nuclear punctate pattern ( also observed in some cells transfected with E72Q and SPC ) , which might reflect misfolded protein . We next measured the catalytic activity of the proteins using an in vitro deamination assay . We previously demonstrated deaminase activity for A3A packaged into HIV-1 virions [14] . To bypass the requirement for packaging , we directly immunoprecipitated A3A from lysates of transfected cells [47] . We further adapted the assay to control for different expression levels by generating A3A by coupled in vitro transcription/translation ( IVT ) using wheat germ extract . An advantage of this method is that it allows assessment of proteins that are unstable or difficult to express in cells , such as FF7595LL and A3A truncations ( Figure S1 ) . Cytidine deaminase activity of the immunoprecipitates was measured by incubation with a radiolabeled deoxyoligonucleotide containing a single deoxycytidine target site . Wild-type A3A protein generated by IVT had deaminase activity ( Figure 1C ) and demonstrated similar activity to A3A produced by transfection ( Figure S1 ) . These results showed that the IVT generated protein was catalytically active and also suggested that A3A does not require a mammalian cellular co-factor for its catalytic activity . Analysis of the active site mutants in the IVT system showed that , consistent with previous studies , mutants in conserved active site amino acids ( H70R , E72Q , C106S , and SPC ) had lost deaminase activity [14] , [50] . Although the F75 and F95 residues have both been shown to be required for APOBEC1 deaminase activity , the F95L mutant of A3A retained deaminase activity in our analysis , while the F75L mutant was inactive . The antiviral activity of A3A mutants was compared to wild-type A3A and A3G by transfection of 293T cells in the recombinant AAV ( rAAV ) production assay ( Figure 1D ) . The APOBEC3 expression vectors were cotransfected with plasmids required for rAAV replication and packaging . The Rep and Cap proteins were supplied in trans to allow replication of an AAV vector and packaging of the ssDNA genome into virus particles . In this study we employed rAAV expressing luciferase ( rAAVLuc ) , which allowed for quantitative assessment of virus production by transduction of target cells . Immunoblotting confirmed that proteins of the expected size were expressed ( Figure 1D ) . Wild-type A3A completely blocked rAAV production , as previously reported [14] . In contrast , neither A3G nor the active site A3A mutants H70R , E72Q , SPC and C106S inhibited rAAV production . The F95L mutant that retains deaminase activity was active against rAAV . Surprisingly , F75L also inhibited rAAV despite its lack of deaminase activity in the in vitro assay . The F75L protein is therefore a separation-of-function mutant of A3A that facilitates analysis of the relative contribution of the deaminase-dependent and -independent mechanisms to AAV inhibition . The double mutant FF7595LL did not inhibit rAAV production but was poorly expressed and showed altered cellular distribution . Together , these data demonstrate that deaminase activity is not required for the anti-AAV effect of A3A . To exclude differences in AAV inhibition due to disparities in protein expression levels , we compared the mutants F75L and F95L with wild-type A3A over a dose-response ( Figure 2A ) . Immunoblotting revealed that expression levels of the mutants F75L and F95L were approximately ten-fold lower than wild-type A3A when equal amounts of DNA were transfected . However , despite differences in the deamination ability of the mutants , they displayed similar antiviral activity to wild-type A3A when equivalent protein levels were compared . We tested the F75L mutant against a panel of oligonucleotides that contained a cytosine in each of the possible tri-nucleotide configurations , but found no evidence of deamination above background levels on any sequence ( Figure S2A and Table S1 ) . Therefore the lack of detectable deaminase activity for F75L in vitro was not caused by an altered target sequence preference . We also tested the deaminase activity of A3A mutants in cell lysates by adapting the quantitative fluorescence resonance energy transfer ( FRET ) assay recently developed for A3G [54] . This FRET assay measures cleavage of a target oligonucleotide dual-labeled with fluorophores ( Figure S2B ) . We observed dose-dependent deaminase activity with increasing amounts of cell lysates from 293T cells transfected with the A3A plasmid . Background levels of activity were obtained with the defective mutants E72Q and C106S . F95L showed deaminase activity in this assay , whereas F75L was not above background ( Figure S2B ) . This result supports the observations from the in vitro deaminase assay and the conclusion that AAV is inhibited in the absence of deamination . In our previous studies of AAV production in the presence of wild-type A3A , we found no detectable evidence of AAV sequence changes but viral replication was inhibited [14] . To detect effects of A3A on the accumulation of rAAV DNA , we used Southern blotting of low molecular weight DNA extracted from the transfected cells during rAAV production ( Figure 2B ) . As controls , we compared the A3A mutants to wild-type A3A and A3G . Replicated rAAV DNA was detected by hybridization with a luciferase probe . Although F75L was slightly less effective than F95L , both mutants inhibited the accumulation of rAAV DNA ( Figure 2B ) , suggesting that inhibition of AAV replication is not dependent on deamination . To examine the effect of A3A and mutants on replication of wild-type parvovirus genomes , we used two different viral systems . AAV2 depends upon helper virus for replication , while MVM replicates autonomously . First we used immunofluorescence to assess the effect of APOBEC3 proteins in cells infected with AAV2 and adenovirus helper virus . As previously shown [14] , replication centers detected by staining for the viral Rep protein were present in cells that expressed A3G but were absent in those with A3A ( Figure 3A ) . Advanced stage viral replication centers were detected in cells expressing inactive mutants . In contrast , A3A mutants that inhibited rAAV production ( F75L and F95L ) also blocked formation of viral replication centers . These data demonstrate that observations made with rAAV production also apply to inhibition of wild-type AAV replication . Replication of an infectious plasmid clone of MVM was also dramatically inhibited by co-transfection of wild-type A3A , but not the C106S mutant ( Figure 3B , lanes 1–3 ) . Interestingly , replication of an MVM genome bearing a large in-frame deletion within the capsid gene , and therefore unable to generate the wild-type capsid proteins necessary to produce single-stranded progeny DNA ( MVM-ΔBglII ) , was also inhibited by A3A ( Figure 3B , lanes 4–6 ) . In separate experiments , the expression of the full spectrum of MVM RNA and protein generated from a non-replicating full-length MVM plasmid was not affected by expression of A3A ( data not shown ) , suggesting that A3A directly affects parvovirus genome replication . To identify further residues required for A3A antiviral activity , we compared the sequence of A3A to the C-terminus of A3B and A3G . Alignment of the amino acid sequences showed that A3A is most closely related to A3B . Two main regions with variable sequences ( VS1 and VS2 ) differ from A3G ( Figure 4 ) . Using the structure of the C-terminal domain of A3G as a template [55] , we predicted the secondary structure of A3A ( Figure 4 ) and this suggested that the VS1 would be located in the loop between the β2 strand and the α1 helix , and that VS2 partially overlaps with the α4 helix . To define regions of A3A responsible for the antiviral activity against parvoviruses , we tested the contribution of the linker and pseudoactive site subdomains [2] . We first generated chimeric proteins between A3A and A3G ( Figure 5A ) joined at the shared PmlI site . The N-terminus of A3A ( residues 1–119 ) was fused to the C-terminal PmlI fragment of A3G ( residues 306–384 ) to form the chimera A3ApmlA3G . The reciprocal chimera A3GpmlA3A was also generated . We also assessed the activity of the C-terminus of A3G ( A3G-CT , residues M197 to N384 ) and a fusion with A3A at the PmlI site ( A3G-CTpmlA3A ) . Cellular localization of the mutants was tested by immunofluorescence in transfected cells ( Figure S3A ) . We found that localization of the chimeric proteins to the cytoplasm was determined by the N-terminal domain of A3G as previously reported [51] . We also measured the deaminase activity of chimeric proteins immunoprecipitated from transfected cells in the in vitro assay with the T28CCCGT28 deoxyoligonucleotide substrate ( Figure 5B , upper panel ) . Full-length A3A and A3G both produced robust deamination , but all of the chimeras were less active . The mutants were also tested for activity against rAAV in the transfection assay ( Figure 5C ) . Neither A3G nor A3G-CT had any inhibitory effect against rAAV despite being expressed well in transfected cells . Fusion of the linker and pseudoactive site subdomains of A3A onto A3G or A3G-CT was not sufficient to confer antiviral activity ( A3GpmlA3A and A3G-CTpmlA3A ) . The A3ApmlA3G chimera , which possesses the linker and pseudoactive site of A3G fused onto A3A , showed diminished antiviral activity . This reduction was confirmed in a dose-response titration ( Figure 5D ) . Together these data suggest that the linker and pseudoactive site regions of A3A are important for deamination and antiviral activity , but that these domains are not sufficient to confer activity to A3G . Careful inspection of the deaminase assay autoradiograms ( Figure 5B , upper panel ) revealed that the deaminated products for A3A and A3G have slightly different mobilities , which likely reflects differences in deamination target specificity [14] , [33] , [56]–[59] . The deamination product for A3GpmlA3A migrated similarly to that of A3A , suggesting it had gained the target site specificity of A3A . To test this possibility , we analyzed the deaminase activity on a deoxyoligonucleotide containing the specific A3A target sequence TCA ( Table S1 ) . While A3A was highly active on the TCA substrate , A3G was inactive ( Figure 5B , middle panel ) . Although the level of deamination by the A3GpmlA3A chimera was less than that of A3A , this mutant was similarly active on both substrates ( Figure 5B , compare upper and middle panels ) . These observations suggest that the region encompassing the linker and pseudoactive site is involved in target site selection [56] . In our assay the C-terminal fragment of A3G was inactive , although recent studies reported that a similar fragment of A3G ( residues 198–384 ) showed mutator activity in bacteria [17] , [60] and deaminase activity in vitro [55] . However , these studies used GST-tagged A3G-CT in the bacterial assays or purified recombinant protein in their in vitro assays , while we have analyzed protein from cell lysates . Furthermore , A3G-CT was shown to be significantly less active than the full-length protein [55] . Our results may also reflect differences in experimental conditions or that protein produced by transfection may be less active due to RNA inhibition or lack of dimerization . Based on the analysis of chimeric proteins , we generated further mutants in which non-conserved residues of A3A were substituted with those from A3G ( see Figure 4 ) . Amino acids that differed between the two proteins were changed throughout A3A ( Figure 6A ) . Most of the mutants consisted of A3A residues replaced with the analogous A3G sequence . Two residues lacking in A3G were deleted from A3A ( ΔWG ) , and in another mutant unique residues from A3G were inserted into A3A ( EPWVR ) . The two main variable regions that contain stretches of divergence ( VS1 and VS2 in Figure 4 ) were switched in two stages that changed 3 or 4 residues at a time ( Figure 6A ) . All mutants displayed the same pattern of cellular localization as wild-type A3A ( Figure S3B ) . Mutant proteins were synthesized by IVT and evaluated for deaminase activity in the in vitro assay ( Figure 6B ) . Mutants PT , MAK , and SK retained wild-type levels of activity . The EPWVR and ΔWG mutants had diminished activity compared to wild-type A3A . The chimeras in variable stretches VS1 and VS2 lacked detectable deaminase activity . When these mutants were included in the rAAV production assay , all of them retained the ability to inhibit AAV ( Figure 6C ) . The VS1 mutants with diminished antiviral activity ( GFLE and PHKHGFLE ) were also compared to wild-type A3A over a dose response ( Figure S4 ) . When similar levels of protein were compared for their effect on rAAV production , the VS1 mutants GFLE and PHKHGFLE showed less activity than wild-type A3A ( ∼75% inhibition compared to ∼95% ) ( Figure S4 ) . This observation suggests that the VS1 region in A3A contributes to the antiviral activity . Together these data demonstrate that residues outside of the putative enzymatic active site of A3A , in both the N-terminus and C-terminus , are required for efficient deamination but that this does not correlate with antiviral activity against parvovirus replication . Analysis of the A3A mutants suggested that the two stretches of residues divergent between A3A and A3G ( VS1 and VS2 ) are important for deamination , and may play a role in antiviral function . Therefore , we determined whether the reciprocal switch ( where residues in A3G were substituted with the sequences from A3A ) would generate a gain-of-function ( Figure 7A ) . Sequences from A3A were incorporated into constructs that express the C-terminal fragment of A3G which can localize in the nucleus ( Figure 7B ) . The mutant proteins were tested for their effect on AAV in the virus production assay ( Figure 7C ) . Proteins of the expected size were expressed at similar levels . The C-terminal fragment that contained the complete sequence from VS1 of A3A ( A3G-CT/KNLLCGFY ) acquired significant inhibitory activity against AAV . It was less active than wild-type A3A and reduced rAAV production by approximately 50% when equal levels of protein where compared ( Figure S5 ) . When incorporated into full-length A3G , the VS1 region of A3A increased deamination in vitro , whereas the VS2 sequences decreased activity ( Figure S6 ) . Incorporation of the A3A sequences into full-length A3G did not confer AAV inhibition , presumably due to cytoplasmic localization or interference by the N-terminus ( Figure S6 ) . Thus A3G-CT/KNLLCGFY provides the first gain-of-function mutant for A3G and demonstrates that the VS1 region of A3A ( residues 60 to 67 ) contributes to the antiviral activity against parvovirus .
In this report we provide multiple pieces of evidence to show that A3A inhibition of AAV can occur through a deaminase-independent mechanism . Mutation of A3A active site residues that are essential for catalytic activity ( H70R , E72Q , SPC99-101AAA and C106S ) led to the loss of activity against AAV . However , other mutants ( F75L and mutants in VS1 and VS2 ) separated the deaminase activity from the ability to inhibit AAV . Together these data indicate that the integrity of the active site is important but that deaminase activity is not required for AAV inhibition . In APOBEC1 , aromatic residues analogous to F75L and F95L of A3A are required for both deaminase activity and binding to nucleic acids [16] , [53] . In the case of A3A , we found that F95 is not required for deaminase activity . This result probably reflects the differences in structure and nucleic acid specificity between APOBEC3 proteins and APOBEC1 , as revealed by recent structural studies [17] , [55] . We found that the F75L mutant was deaminase-defective in our in vitro deaminase assay . The lack of deaminase activity on a panel of target oligonucleotides demonstrated that the F75L A3A mutant is truly deaminase-deficient , and has not simply changed its target site preference and eluded detection in the deaminase assay . Although the in vitro deaminase assay has limitations , we demonstrated that lysates containing F75L also lacked deaminase activity in the FRET assay [54] . Evidence that deaminase activity is not essential for inhibition of AAV by A3A is consistent with the absence of signs of deamination in AAV DNA in cells expressing A3A [14] . Deaminase-independent inhibition of ΔVif-HIV and retroelements by A3G has been controversial [35] , [39] . The deaminase-defective A3G mutants in E259 retains anti-viral activity when over-expressed [41] , [42] . However , when equivalent proteins levels are compared in transient transfections or in stable cell lines , the deaminase deficient mutant has significantly less potent antiviral activity than wild-type A3G [38] , [46] , [61] . In our studies we assessed the dose response of A3A mutants by comparing their antiviral activity against AAV in titration experiments . We identified deaminase-defective mutants ( F75L and mutants in VS2 ) that displayed similar activity against AAV as wild-type A3A when analyzed at comparable protein levels . In addition , the F75L and VS2 mutants displayed the same subcellular localization as the wild-type protein and thus their phenotype cannot be ascribed to protein mislocalization . In addition to providing evidence for deaminase-independent antiviral activity , our study also offers insights into the structural basis of APOBEC3 protein function . The linker and pseudoactive site domains in the N-terminus of A3G are required for HIV-1 virion incorporation [19] , [62] . We demonstrate that these domains also influence the target site specificity of APOBEC3 . A3G prefers the target site ( T/C ) CC [9] , [21] , [33] , [57] , [59] , while A3A is more flexible , showing preference for ( T/C ) CA [14] . Replacement of the linker and pseudoactive sub-domains at the end of A3G with those from A3A modified the target site preference towards the A3A-specific consensus target TCA . Thus , A3A residues in the C-terminus contribute to its target specificity , in agreement with data from chimeric and mutant proteins of other APOBEC3 family members [56] . Residues in the VS2 region have been implicated in target site specificity [1] , but in our hands the exchange of VS2 residues between A3A and A3G did not affect target specificity . It is unclear why A3A has more potent in vitro deaminase activity than other APOBEC3 proteins . Deaminase activity of A3G resides in the C-terminal CDD [20] , [21] , [28] , [41] , [60] , which shares 68% identity with A3A . An intriguing difference between A3A and A3G is the presence of two additional residues ( WG ) within the PCX2–4C motif of A3A . Deletion of these amino acids in the A3A/ΔWG mutant slightly reduced in vitro deaminase activity ( Figure 6 ) . The variable region VS1 is situated immediately upstream of the H-X-E-X23–28-P-C-X4-C conserved motif , and replacement of A3A sequences with those from A3G caused a decrease in deaminase activity . Substitution of the VS1 region in A3G with sequences from A3A increased deaminase activity compared to wild-type A3G ( Figure S6 ) . This suggests that the VS1 region ( residues 60–67 ) may contribute to the increased enzymatic activity of A3A [14] , [50] . Interestingly , the VS1 region in located in the active center loop 3 of the C-terminal domain of A3G and disruption of this loop results in greatly impaired A3G deaminase activity [55] . In the variable region VS2 , substitution of A3A amino acids with those from A3G also decreased deamination , suggesting that this region is important for catalytic activity . This observation is supported by the decrease in deaminase activity observed for the reciprocal A3G mutants ( A3G/YDP and A3G/YDPLYK ) ( Figure S6 ) . Together these results indicate that regions outside of the active site contribute to catalytic activity of A3A . In support of our observations , a recent mutagenesis study of A3G also suggested that C-terminal residues ( residues 276–384 ) are important for deaminase activity [60] . Multiple mechanisms have been proposed to explain the deaminase-independent inhibition of retroelements and HBV by APOBEC3G [35] , [39] . In the case of retroviruses , the APOBEC3 proteins have been suggested to inhibit RT , prevent accumulation of reverse transcripts and viral cDNA in target cells , and block integration [37] , [42] , [44] , [45] . Biochemical studies have shown that purified recombinant A3G inhibited RT-catalyzed DNA elongation in vitro and this was independent of its deaminase activity [63] . A3G can inhibit HBV in the absence of extensive editing [12] , and has been suggested to be due to inhibition of early steps in viral reverse transcription and strand elongation [48] . AAV inhibition differs from these other systems because it does not involve an RT-mediated step or any RNA substrates . Since A3A also inhibits MVM replication and does not affect production of viral proteins [14] , we favor the idea that A3A inhibits parvovirus replication through a direct interaction with the viral DNA or the replication machinery . Binding to ssDNA by A3G is proposed to inhibit RT processivity [21] , [63] , and binding of A3A to ssDNA in the parvovirus genome could physically block movement of the DNA polymerase along the viral template . Although this inhibition would be independent of catalytic activity , amino acids in the active site may be required for efficient nucleic acid binding , explaining the loss of antiviral activity for mutants such as E72Q or C106S . Preliminary results suggest that the F75L mutant retains its ability to bind nucleic acid ( data not shown ) . Potential binding sites could be the viral ITR or the ssDNA/dsDNA junction , which may reduce Rep binding and inhibit DNA synthesis . A3A is not found in high molecular weight complexes that have been reported to modulate A3G activity [64] , [65] , however we cannot exclude the possibility that AAV inhibition might be mediated through A3A interactions with AAV Rep or cellular proteins that are required for AAV replication [66] . It will be interesting to investigate whether recombinant A3A can block viral DNA replication in an in vitro replication assay where we could test the direct activity of A3A on AAV replication [67] . This approach , however , is currently limited by the requirement for purified recombinant A3A and its mutants . Although 293 cells are a standard system used to study parvovirus replication , it remains unclear whether endogenous A3A restricts parvovirus infection in vivo . It is unknown which cells represent the primary site of AAV infection and replication in vivo . It has been shown that expression of APOBEC3 proteins is induced in response to interferon-α ( IFNα ) [68] . The levels of A3A that achieve inhibition of AAV in transfected cells are within the range of endogenous A3A levels detected in peripheral blood mononuclear cells ( PBMCs ) and macrophages activated with IFNα ( Figure S7 ) . Therefore , it would be interesting to test whether cells refractory to parvovirus replication will allow AAV replication when endogenous A3A levels are reduced . In addition to parvoviruses , A3A is active against LINE1 and other retrotransposons [14] , [50] , [51] , [64] , [69]–[71] where hypermutation is not detected . It will be informative to determine whether this occurs by a mechanism similar to parvovirus inhibition . In summary , our study demonstrates that the DNA cytidine deaminase activity of A3A is not required for inhibition of parvovirus replication . The combination of the single-domain cytidine deaminase A3A with the simple model system of parvovirus replication provides a valuable tool to uncover new mechanisms for the antiviral activity of the APOBEC3 proteins .
293T , HeLa , A9 and human osteosarcoma U2OS cell lines were purchased from the American Tissue Culture Collection . Cells were grown as monolayers in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal bovine serum and antibiotics at 37°C in a humid atmosphere containing 5% CO2 . Expression plasmids encoding cDNAs for A3A ( NM_145699 ) and A3G ( NM_021822 ) and the A3A mutants H70R , E72Q , and C106S in the pcDNA3 . 1 ( + ) vector with a hemagglutinin ( HA ) tag at the C-terminus have been previously described [14] . New A3A and A3G mutants were generated by site-directed mutagenesis using the QuikChange kit ( Stratagene ) ( Table S1 ) . The truncated form of A3G ( A3G-CT , residues 197-384 ) was generated by PCR amplification of its C-terminus . Plasmids expressing AAV Rep/Cap proteins ( pXX2 ) and the adenovirus helper proteins ( pXX6 ) have been described [72] . The rAAV vector plasmid ( pACLALuc ) consists of the luciferase gene amplified from pGL3basic ( Promega ) cloned into an ITR-flanked expression cassette under the control of the CMV promoter and the BGH polyadenylation signal ( pACLA ) . The complete ITR-flanked expression cassette in pACLALuc is 4 . 3 Kb . Recombinant AAV production assays were performed as previously described [14] , [72] . Briefly , 293T cells were seeded at 0 . 5×106 cells/well in 6-well plates and the next day were co-transfected with pXX6 ( 2 . 25 µg ) , pXX2 ( 0 . 75 µg ) , pACLALuc ( 0 . 75 µg ) and APOBEC3 expression vector ( 1 µg unless otherwise stated ) or pcDNA3 . 1 ( + ) control vector ( 1 µg ) . Dose-response titrations maintained the total amount of effector DNA by addition of pcDNA3 . 1 ( + ) . Transfections were performed in duplicate or triplicate using polyethyleneimine ( PEI ) [73] . Cells were harvested 72 h post-transfection after two washes in ice-cold PBS , and one third of each sample was removed for immuno-blotting . The other two thirds of the cells were used to generate rAAVLuc virus lysates by freeze/thaw cycles followed by centrifugation . Virus lysates were used to transduce 293T cells in 48 wells in triplicate . Transduced cells were incubated with Steady-Glo luciferase substrate reagent ( Promega ) 48 h post–transduction and lucifierase activity was quantified in triplicate in 96 well Lumiplates ( Greiner Bio-One ) in a TopCount NXT scintillation and luminescence counter ( PerkinElmer ) . rAAV production experiments are presented as mean+SEM of the relative value ( % ) of at least three independent experiments , and compared to mock transfections with pcDNA3 . 1 ( + ) . Immunoblotting was performed essentially as described [14] . Cell pellets from rAAVLuc production assays were lysed in lysis buffer ( 137 mM NaCl , 2 . 68 mM KCl , 10 . 1 mM Na2HPO4 , 1 . 76 mM KH2PO4 , 1 mM NaO3V , 20 mM β-glycerol phosphate , 20 mM NaF , 0 . 1% NP40 , and 0 . 025% Triton-X 100 ) supplemented with Complete protease inhibitor cocktail ( Roche ) for 30 min on ice . The lysates were clarified by centrifugation at 10 , 000×g for 20 min . Protein concentrations from whole cell lysates were quantified by BCA assay ( Bio-Rad ) , and 20 µg of protein was loaded per well onto polyacrylamide gels . Proteins were separated in 4–12% or 12% Acrylamide Bis-Tris NuPage gels in MOPS buffer ( Invitrogen ) and transferred onto Hybond nitrocellulose membranes ( Amersham Biosciences ) . Membranes were probed with anti-HA 16b12 monoclonal antibody ( mAb ) ( Covance ) and anti-Ku86 mAb ( Santa Cruz ) . Bound antibody was detected by incubation with goat anti-mouse antibody conjugated to horseradish peroxidase ( Jackson ImmunoResearch ) , and the bands were visualized with enhanced chemiluminiscence reagent ( ECL Western Lightning Kit , PerkinElmer ) followed by autoradiography . APOBEC3 protein localization was determined by indirect immunoflorescence [14] . U2OS or HeLa cells were grown on glass coverslips in 24 well plates and transfected with 0 . 8 µg APOBEC3 expression vector using Lipofectamine 2000 ( Invitrogen ) . After 36–48 h , cells were washed with PBS , fixed with 3% paraformaldehyde for 20 min and extracted with 0 . 5% Triton X-100 in PBS for 10 min . Cells were incubated with 3% BSA for 30 min , followed by incubation with anti-HA mAb 16b12 ( 1∶2000 ) . A 1∶2000 dilution of goat anti-mouse conjugated Alexa Fluor 568 ( Invitrogen ) and DAPI ( Sigma Aldrich ) in 3% BSA in PBS was added to cells and samples were incubated for 1 h at room temperature . The coverslips were mounted in Fluoromount-G ( Southern Biotech ) and cells were visualized by fluorescence microscopy ( Diaphot 300 inverted microscope , Nikon ) . For AAV replication U2OS cells were seeded on glass coverslips and transfected with APOBEC3 expression vector , and infected 16 h post-transfection with wild-type AAV and/or adenovirus . After 24 hr , the cells were fixed and stained with anti-HA and anti-Rep antibodies and DAPI as described above . Rep from the input virus is undetectable in this assay , so positive Rep staining is indicative of AAV replication . Low molecular weight AAV episomal DNA was analyzed by Southern hybridization with a 32P-labeled luciferase probe as previously described [74] . Briefly , 293T cells grown in 6 well tissue culture plates to 95% confluency were co-transfected with plasmids for rAAV production using Lipofectamine 2000 following the manufacturer's protocol . The following effector plasmids were included: pcDNA3 . 1 ( + ) ( 1 µg ) , A3G ( 1ug ) , A3A ( 1 µg , 0 . 1 µg , 0 . 01 µg and 0 . 001 µg ) , F75L ( 1 µg ) and F95L ( 1 µg ) . After 48 h , the cells were collected and washed in PBS . One third of each sample was removed for immunoblotting . DNA was isolated from pellets by a modified HIRT protocol [74] and digested with DpnI ( New England Biolabs ) to remove input plasmid . DNA was processed by gel electrophoresis on a 1% agarose gel in TAE buffer . The pACLALuc plasmid was digested with SmaI as control . The gel was depurinated in 0 . 2 M HCl , denatured in 1 M NaCl , 0 . 5 M NaOH , and neutralized in 0 . 5 M Tris pH 7 . 5 , 1 . 5 M NaCl . DNA was then transferred to a Hybond XL membrane ( Amersham Biosciences ) and UV-cross linked . The membrane was hybridized with a 32P labeled luciferase probe generated by PCR using the primers described in Table S1 and labeled with 32P dCTP using Radivue II labeling kit ( Amersham ) , and visualized in a FLA-5100 phosphorimager ( Fuji ) . Southern blot replication assays of wild-type MVM and MVM-ΔBglII were performed as previously described [75] . 293T cells were seeded at 5×105 cells/well in 6-well plates and transfected after one day with 3 µg of APOBEC3 expression vector . Two days post-transfection , the cells were rinsed twice with cold PBS and lysed for 30 min on ice in lysis buffer ( 50 mM Tris , pH 8 . 0 , 40 mM KCl , 50 mM NaCl , 5 mM EDTA , 0 . 1% Triton X-100 , 10 mM DTT ) . Lysates were clarified by centrifugation at 13 , 000×g for 10 min and pre-cleared with 50 µl of High flow protein-G-Sepharose ( Amersham ) . The lysate was incubated with anti-HA mAb 3F10 ( Roche ) for 2 h at 4°C . The lysate-antibody was then incubated with High flow protein-G-Sepharose for 1-2 h at 4°C . The resin was washed three times with lysis buffer . One-fifth of the resin was removed for immunoblot analysis and the remainder was washed once with deaminase reaction buffer ( 40 mM Tris , pH 8 . 0 , 10% glycerol , 40 mM KCl , 50 mM NaCl , 5 mM EDTA , and 1 mM DTT ) . PAGE purified deoxyoligonucleotide ( Table S1 ) was 5′-end 32P labeled and added into 20 µl of deaminase reaction buffer . The reaction was incubated at 37°C for 20 h , stopped by heating to 90°C for 5 min , cooled on ice , and then centrifuged to collect the resin at the bottom of tube . The supernatant was incubated with uracil DNA glycosylase ( New England Biolabs ) in buffer containing 20 mM Tris , pH 8 . 0 , 1 mM DTT for 1 h at 37°C and treated with 150 mM NaOH for 1 h at 37°C . The samples were incubated at 95°C for 5 min , 4°C for 2 min and separated by 15% TBE/urea-PAGE . The gel was dried , exposed to a phosphorimager screen and analyzed using a FLA-5100 scanner ( Fuji ) . For in vitro synthesis of A3A and mutants we employed the TNT Coupled Wheat Germ Extract System ( Promega ) using T7 polymerase . Translation reactions were performed with non-labeled amino acids following the manufacturer's protocol in 50 µl final volume that included 1 µg of pcDNA3 . 1 ( + ) plasmid encoding APOBEC3 . Reactions were incubated for 90 min at 30°C . After incubation , 450 µl of TritonX-100 buffer ( 50 mM Tris , pH 8 . 0 , 40 mM KCl , 50 mM NaCl , 5 mM EDTA , 0 . 1% Triton X-100 , 10 mM DTT ) were added to the reactions and used to analyze deaminase activity after immunoprecipitation as described above . In order to measure deaminase activity directly from IVT reactions , after 90 min incubation at 30°C translation reactions were centrifuged at 10 , 000×g for 1 min . An aliquot of supernatant ( 5 µl ) was removed for immunoblot analysis . The reaction mixture ( 15 µl of supernatant ) was incubated with 5′-end 32P labeled deoxyoligonucleotide in 30 µl of deaminase reaction buffer and assayed for deaminase activity following the same procedure described above . To quantify deaminase activity from cell lysates , we used a modification of the FRET-based protocol described by Thielen et al . [54] . 293T cells were seeded at 1×106 cells/well in 6 cm plates and a day later transfected with 8 µg of APOBEC3 expression vector . After 36 to 48 h , cells were resuspended in lysis buffer ( 200 µl ) and lysates were obtained as described above . Cell lysate ( 10 µl ) was mixed with 70 µl of FRET deaminase buffer ( 40 mM Tris , pH 8 . 0 , 40 mM KCl , 50 mM NaCl , 5 mM EDTA ) containing 10 pmol of dual-labeled probe ( Table S1 ) and 0 . 4 units of UDG ( New England Biolabs ) . Reactions were incubated at 37°C for 90 min followed by addition of 4 µl of 4N NaOH and incubation at 37°C for 30 min . Reactions were neutralized with 4 µl of 4N HCl and 36 µl of 1 M Tris-HCl ( pH 8 ) . 6FAM fluorescence was measured at 25°C in an Mx30005P ( Stratagene ) . Two-fold serial dilutions of each lysate were analyzed in duplicate . Fluorescence detected in 293T cells transfected with pcDNA3 . 1 ( mock ) was substracted from all samples and deaminase activity is shown as relative fluorescence units ( RFU ) . Human monocytes were purified from leukocyte enriched blood samples ( New York Blood Center ) using CD14+ magnetic beads ( Miltenyi Biotec ) according to manufacturer instructions . The CD14+ monocytes were cultured with 50 ng/ml GM-CSF ( Invitrogen ) for 7 days in order to differentiate them into macrophages . The monocyte-derived macrophages were plated in 12 well plates at 106 cells per well and cultured with or without 2000U of Universal Type I Interferon ( PBL Biomedical Laboratories ) for 20 hours . The cells were then collected in lysis buffer . Cell lysates from human peripheral blood mononuclear cells ( PBMCs ) were obtained from F . Chisari ( The Scripps Research Institute ) [76] . PBMCs were treated with 1000U of IFN-α or kept untreated for 24 hrs before cell lysates were generated . Lysates of PBMCs ( 60 µg ) and macrophages ( 30 µg ) of each sample were run on 4–12% or 12% Bis-Tris gels and analyzed by immunoblotting as described above . Anti-A3A ( raised against an N-terminal peptide ) ( 1∶250 dilution ) or anti-recombinant A3A ( 1∶1000 dilution ) polyclonal rabbit sera were used for A3A detection as described .
|
The APOBEC3 proteins constitute a family of seven cytidine deaminases . Cytidine deaminases are editing enzymes able to remove the amine group from cytidine in single-strand DNA ( ssDNA ) and RNA , converting it to uracil . APOBEC3 proteins have potent antiviral activity against retroviruses , retrotransposons , and DNA viruses . APOBEC3 generated high interest because of the ability of APOBEC3G ( A3G ) to inhibit HIV . APOBEC3A ( A3A ) is a member of the family that inhibits the human parvovirus adeno-associated virus ( AAV ) and the retrotransposon LINE-1 . Parvoviruses are simple ssDNA viruses that do not require a retrotranscription step for their replication . In contrast to A3G , which is predominantly cytoplasmic , A3A is located in both the nucleus and cytoplasm . In addition , A3A consists of a single cytidine deaminase catalytic domain , whereas A3G has two . The dependence of the antiviral function on deaminase activity is controversial . In this study , we identify numerous A3A residues required for deaminase and antiviral activities . We show that A3A not only inhibits AAV but also the minute virus of mice ( MVM ) . Importantly , we demonstrate that A3A does not require its deaminase activity to block the replication of both parvoviruses . Thus , exploiting the simplicity of parvoviruses together with the single-domain cytidine deaminase A3A , we are able to demonstrate that cytidine deaminase activity is not required for APOBEC3 mediated viral inhibition .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry/replication",
"and",
"repair",
"molecular",
"biology",
"immunology/immunity",
"to",
"infections",
"virology/host",
"antiviral",
"responses"
] |
2009
|
Deaminase-Independent Inhibition of Parvoviruses by the APOBEC3A Cytidine Deaminase
|
Chronic helminth infections typically induce an immunoregulatory environment , with markedly reduced immune responses to both parasite-specific and unrelated bystander antigens . Here we tested whether these changes are also observed in human infections with Mansonella ozzardi , a neglected filarial nematode widely distributed across Latin America . CD4+ T cell populations from microfilaremic ( Fil+ ) and uninfected ( Fil- ) inhabitants in M . ozzardi-endemic riverine communities in Brazil were characterized by flow cytometry analysis . Plasma concentrations of a wide range of cytokines and chemokines were measured . We examined whether M . ozzardi infection is associated with suppressed in vitro lymphoproliferative and inflammatory cytokine responses upon stimulation with filarial antigen , unrelated antigens or mitogens . Fil+ subjects had lower plasma levels of selected inflammatory cytokines , such as TNF-α , IL-8 , and IL-6 , than their Fil- counterparts . However , we found no evidence for attenuated T-cell responses to filarial antigens or co-endemic pathogens , such as malaria parasites and Toxoplasma gondii . CD4+ T cells expressing CD39 , an ectonucleosidase involved in the generation of the anti-inflammatory molecule adenosine , were increased in frequency in Fil+ subjects , compared to uninfected controls . Significantly , such an expansion was directly proportional to microfilarial loads . Surprisingly , CD39 blocking with a neutralizing antibody suppressed antigen-driven lymphoproliferation in vitro , while decreasing inflammatory cytokine responses , in Fil+ and Fil- individuals . These findings suggest that circulating CD4+ CD39+ T cells comprise subsets with both regulatory and stimulatory roles that contribute to the immune homeostasis in chronic M . ozzardi infection .
Helminth parasites induce a wide range of mechanisms that downregulate immunity , limiting host’s inflammation-mediated tissue damage and favoring parasites’ persistence in typically asymptomatic carriers [1 , 2] . Nearly 40 years ago , T cell proliferation responses to parasite-specific antigens , but not to mitogens , were first shown to be severely dampened in chronic human infections with tissue-dwelling helminths , such as in schistosomiasis and lymphatic filariasis [3] . Likewise , reduced antigen-driven T-cell responses have also been documented in infections with lumen-dwelling intestinal nematodes [4] . The immunomodulatory environment in these long-lasting infections is characterized by cytokine responses skewed towards interleukin ( IL ) -10 and transforming growth factor ( TGF ) -β and the expansion of several regulatory T ( Treg ) and B cell subtypes , including the classical CD4+CD25hiCD127-FoxP3+ Treg cells . Immune responses to unrelated bystander antigens , such as co-occurring pathogens , allergens , and vaccines , are also often suppressed—a phenomenon termed spillover suppression [5] . Because major infectious diseases such as HIV infection , tuberculosis , and malaria are endemic to areas where several helminth infections are highly prevalent , spillover suppression has major public health implications [5] . However , it remains unclear to what extent similar regulatory networks operate in all chronic helminth infections of humans . Three filarial nematodes of the genus Mansonella infect humans: M . streptocerca , which is endemic to Africa; M . perstans , which is commonly found in Africa but also occurs in South America [6]; and M . ozzardi , which is found exclusively in Latin America and the Caribbean islands [7] . M . ozzardi has a patchy geographic distribution from southern Mexico to northwestern Argentina , with overall prevalence ranging from <1% to 46% in affected communities [8 , 9] . Adult worms have been recovered from subcutaneous tissues of experimentally infected patas monkeys [10] , but their habitat in human hosts remains uncertain . Microfilariae circulate in the peripheral blood all day long , with very little , if any , periodicity . Most human infections cause little or no disease and may remain undiagnosed and untreated for several years [8] . Infections with the filarial nematodes Wuchereria bancrofti , Brugia malayi , and M . perstans typically reduce immunity to malaria parasites and other pathogens , with an IL-10-dependent decrease in IL-12p70 , interferon ( IFN ) -γ and IFN-γ-induced protein 10 ( IP-10 ) responses upon T-cell stimulation [11–14] . Nevertheless , the immunomodulatory potential of M . ozzardi infections and their spillover effects remain largely unexplored—even in the Amazon Basin , where this filarial nematode and malaria parasites are endemic and co-infections with them are very likely . Here we hypothesized that immune responses might be also downregulated in subjects chronically harboring M . ozzardi microfilariae . We first characterized peripheral blood mononuclear cell ( PBMC ) populations from people living in M . ozzardi-endemic communities of Brazil and found an expansion of CD4+ T cell subtypes expressing CD39 in infected subjects , compared to local uninfected controls . CD39 is an ectonucleosidase that catalyzes the phosphohydrolysis of extracellular ATP and ADP to AMP , which is in turn used by CD73 to synthesize the immunosuppresive molecule adenosine [15] . We next examined whether M . ozzardi infection is associated with suppressed T-cell responses driven by filarial and unrelated antigens and whether this is mediated through the induction of high levels of regulatory cytokines . Finally , we explored the effects of in vitro neutralization of CD39 expression on antigen-driven lymphoproliferative responses and cytokine production .
Study protocols were approved by the Institutional Review Board of the Institute of Biomedical Sciences , University of São Paulo , Brazil ( 1133/CEP , 2013 ) . Written informed consent was obtained from all patients or their parents or guardians , if participants were minors aged <18 years . The field study was carried out in six villages—Monte Verde , Valparaíso , Boa Vista , Retiro , Nova Vida , and São Pedro—situated along the banks of Purus River , in northwestern Brazil ( S1 Fig ) . These villages are located in the municipality of Boca do Acre ( 8°45’19"N , 67°23’50"W ) , southern Amazonas state , with a combined population of 1 , 300 inhabitants . They had previously been shown to be endemic for M . ozzardi , with an average prevalence of infection estimated at 27 . 3% [16] . Although M . ozzardi vectors have not been characterized in this area , simuliid black flies of the species Simulium amazonicum , S . argentiscutum , S . oyapockense s . l . , and S . roraimense are believed to transmit this helminth across the Amazon Basin of Brazil [17] . Low-level malaria transmission is recorded year-round in these communities , with Plasmodium vivax accounting for over 95% of the infections in 2013 . Other tissue-dwelling helminths such as Schistosoma mansoni and W . bancrofti are not endemic to this region . During a pilot study in March 2013 , we carried out house-to-house visits in the target communities and randomly screened 287 inhabitants for M . ozzardi microfilariae . By combining thick-smear microscopy and polymerase chain reaction ( PCR ) on finger-prick blood samples , we found 41 ( 14 . 3% ) subjects harboring M . ozzardi microfilariae , with prevalences of infection ranging between 6 . 1% and 30 . 4% across communities . As described in similar endemic settings [18] , prevalence of infection increased with age , from 4 . 4% in children below 10 years to 57 . 1% in adults over 50 years . During the next visit to the target communities , in September 2013 , we invited subjects previously found to harbor microfilariae , as well as a subsample of individuals found to be uninfected or not screened during the pilot survey , to contribute 40 mL of venous blood for microfilariae detection and PBMC and plasma separation . Because most infected subjects were adults , we prioritized adults as uninfected controls in order to have a similar age composition in the comparison groups . Blood was drawn between 9:00 am and 3:00 pm . Infections diagnosed during the pilot study and those newly diagnosed during this second survey were treated with a single dose of 0 . 2mg/kg of ivermectin [19] after blood sampling . No post-treatment samples were obtained . Study subjects were given plastic containers containing 10% formalin and asked to provide a stool sample . Stool specimens were examined for eggs , cysts and larvae of intestinal parasites with a standard sedimentation method [20] , which was preferred over the standard Kato-Katz method because we used formalin-diluted stool samples . No study participant had intestinal parasites detected by stool examination; furthermore , none of them had malaria diagnosed by using a quantitative real-time PCR targeting the 18S rRNA gene [21] . Plasma samples from all study participants were screened for IgG antibodies to Toxoplasma gondii by using the Serion ELISA classic IgG kit ( Institut Viron/Serion , Würzburg , Germany ) ; all of them were seropositive . Laboratory diagnosis of M . ozzardi was based on microscopic examination of thick smears and quantitative PCR on all samples . Subjects positive by either method were defined as microfilaremic ( Fil+ ) and those negative by both methods were defined as Fil- . Thick blood smears were stained with Giemsa; at least 100 fields were examined for microfilariae , under 1000 × magnification , before a slide was declared negative . We standardized an in-house , SYBR green-based real-time PCR approach to quantify the number of copies of amplicons , which was used as a proxy of microfilarial density . DNA templates for PCR amplification were isolated from 200 μL of whole venous blood on a QIAcube automated platform ( Qiagen , Hilden , Germany ) , using QIAamp DNA blood kits ( Qiagen ) , and eluted in 200 μL for water . Each 20-μL reaction mixture contained 2 μL of sample DNA ( corresponding to 2 μL of whole blood ) , 10 μL of 2× Maxima SYBR Green qPCR master mixture ( Fermentas , Burlington , Canada ) and 0 . 5 μM of each primer ( forward , 5’-CTT ATC ATC AGG TGA TAT TAA T-3’; reverse , 5’-TTA GTT TCT TTT CCT CCG CT-3’ ) . These primers allow the amplification of a M . ozzardi-specific 295-base pair ( bp ) fragment of the internal transcribed spacer ( ITS ) -2 of the ribosomal DNA ( rDNA ) gene [22] . Standard curves were prepared with serial tenfold dilutions of the target sequence , cloned into pGEM-T Easy vectors ( Promega , Madison , WI , USA ) , to allow for calculating the number of ITS-2 amplicons/μL of blood . We used a Mastercycler gradient real-time PCR cycler ( Eppendorf , Hamburg , Germany ) for PCR amplification with a template denaturation step at 95°C for 10 min , followed by 40 cycles of 15 sec at 95°C and 1 minute at 55°C , with fluorescence acquisition at the end of each extension step . Amplification was immediately followed by a melting program consisting of 15 sec at 95°C , 15 sec at 55°C , and a stepwise temperature increase of 0 . 2°C/sec until 95°C , with fluorescence acquisition at each temperature transition . No-template controls ( containing all reagents for amplification except for the DNA template ) were run for every qPCR microplate . Filaria-specific IgG4 antibodies were determined by ELISA as described [23] , using a B . malayi adult worm extract ( BmA ) as a solid-phase capture antigen . We compared plasma levels of the following cytokines in Fil+ and Fil- subjects: epidermal growth factor ( EGF ) , eotaxin , fibroblast growth factor ( FGF ) -basic , granulocyte-colony stimulating factor ( G-CSF ) , granulocyte-macrophage colony-stimulating factor ( GM-CSF ) , hepatocyte growth factor ( HGF ) , IFN-α , IFN-γ , IL-1ra , IL-1β , IL-2 , IL-2r , IL-4 , IL-5 , IL-6 , IL-7 , IL-8 , IL-10 , IL-12 ( p40/p70 ) , IL-13 , IL-15 , IL-17 , IP-10 , monocyte chemoattractant protein ( MCP ) -1 , monokine induced by IFN-γ ( MIG , also known as CXCL9 ) , macrophage inflammatory protein ( MIP ) -1α , MIP-1β , CCL5 ( also known as RANTES , Regulated on Activation , Normal T Cell Expressed and Secreted ) , TNF-α , and vascular endothelial growth factor ( VEGF ) . Samples were tested using the Human Cytokine Magnetic 30-Plex Panel ( Invitrogen ) , and data were acquired with a Bio-Plex 200 System ( Bio-Rad ) using Luminex 100 xMAP technology ( Luminex , Austin , USA ) . PBMC were separated by gradient centrifugation with Ficoll-Paque Plus ( GE Healthcare , Little Chalfont , United Kingdom ) , up to 6 hours after blood collection , cryopreserved in liquid nitrogen , and thawed as described [24] . PBMC were evaluated for viability and only samples with >90% viability were used in subsequent analyses . CD4+ T cell subtypes circulating in the peripheral blood of Fil+ and Fil-subjects were defined according to the ( co- ) expression of regulatory and activation markers . Regulatory molecules included: ( a ) the transcription factor FoxP3 , ( b ) the TNF receptor family costimulatory receptor OX-40 ( CD134 ) , ( c ) the glucocorticoid-inducible TNF receptor family-related protein ( GITR/CD357 ) , ( d ) the TNF receptor II ( TNFRII/CD120b ) , ( e ) the ectonucleosidase CD39 ( also known as NTPDase1 ) , ( f ) the programmed cell death protein 1 ( PD-1/ CD279 ) , ( g ) the lymphocyte-activation gene 3 product ( LAG-3/CD223 ) , ( h ) the membrane-bound C-terminal pro-region of TGF-β known as latency-associated peptide ( LAP-TGF-β ) , and ( i ) the primary inhibitory receptor CTLA-4 ( CD152 ) , which competes with CD28 for CD80 and CD86 binding on antigen-presenting cells . The activation markers were: ( a ) HLA-DR , which identifies a Treg cell subset involved in contact-dependent immune suppression , and ( b ) CD69 , an early activation marker with inhibitory properties . Treg cells were phenotypically defined as CD4+CD25hi cells that express FoxP3 but not the α chain of the IL-7 receptor ( CD127 ) . After thawing , 106 viable cells/mL were transferred to V-bottomed 96-well microplates with 200 μL of staining buffer ( PBS with 2% fetal bovine serum and 2 mm EDTA ) per well and centrifuged at 600 × g for 5 min . For surface staining , the cell pellet was incubated with directly conjugated antibodies at room temperature , in the dark , for 30 min and washed twice with 200 μL of staining buffer . To analyze the intracellular expression of FoxP3 and Ki67 , a nuclear protein that regulates cell division , cells were fixed and permeabilized using the Transcription Factor Fixation/Permeabilization kit of eBiosciences ( San Diego , CA , USA ) and then incubated with specific , labeled antibodies . In this context , we used intracellular Ki67 expression as an indicator of T-cell proliferation [25] . Monoclonal antibody panels are listed in S1 , S2 and S3 Tables . Samples were acquired on an LSR Fortessa flow cytometer ( BD Biosciences , San Jose , CA , USA ) using the FACSDiva software ( BD Biosciences ) . Anti-mouse IgG-coated beads ( BD Biosciences ) stained with each fluorochrome separately were used for software-based compensation . We used the Live/Dead Fixable Blue Dead Cell Stain for UV excitation ( Invitrogen , Carlsbad , CA , USA ) for dead cell exclusion and collected ≥ 300 , 000 events ( panels 1 and 2 , S1 and S2 Tables ) or ≥ 600 , 000 events ( panel 3 , S3 Table ) in each live gate . Only samples with >80% viability were analyzed . Fluorescence minus one ( FMO ) controls were used to control for spectral overlap . Boolean analysis was applied to evaluate co-expression of different molecules by the same cells . Data analysis was carried out using FlowJo software version 8 . 8 . 6 ( Tree Star , Ashland , OR , USA ) . We used BmA , prepared as described [26] , to elicit filarial-specific lymphocyte proliferation in vivo . As bystander antigens , we used: ( a ) commercially available Staphylococcus aureus enterotoxin B ( SEB; Sigma-Aldrich ) , ( b ) a soluble tachizoite extract prepared with the RH strain of T . gondii ( TgT ) [27] , and ( c ) red blood cells ( RBC ) infected with P . vivax schizonts ( PvS ) . To prepare PvS , patient-derived infected blood was filtered through Fresenius Kabi BioR01 Plus filters ( Bad Homburg , Germany ) to remove WBC and cultured for 44–46 h in McCoy’s 5A medium ( Invitrogen ) supplemented with 20% human AB serum until intracellular parasites reached the mature schizont stage ( ≥ 4 nuclei ) [28] . RBC infected with mature schizonts were enriched by using MACS separation columns ( Miltenyi , Auburn , CA , USA ) as described [29] . To examine whether M . ozzardi is associated with T-cell hyporesponsiveness , we stimulated PBMC from Fil+ and Fil- subjects with the antigens described above . To this end , 106 PBMC/well were labelled with CellTrace Violet ( Invitrogen ) and incubated in 96-well microplates with either BmA ( 13 μg/mL ) , SEB ( 10 μg/mL ) , TgT ( 12 μg/mL ) , or 105 P . vivax-infected RBC/well , to a final volume of 200 μL/well . BmA and SEB were used with 10 μg/mL of anti-CD28/CD49d co-stimulus ( BD FastImmune , San Jose , CA , USA ) . Cells were incubated for 72 h at 37°C in RPMI-1640 medium supplemented with 10% inactivated fetal bovine serum , 10 mM HEPES , 2 mM L-glutamine , 1 mM sodium pyruvate , 55 μM 2-mercaptoethanol , and a 1% ( vol/vol ) solution containing 100 U/mL of penicillin , 10 μg/mL of streptomycin , and 25 μL/mL of amphotericin B in a humidified chamber with 5% CO2 . Medium alone and 10 μg/mL of phytohemagglutinin ( PHA-P , Sigma-Aldrich , St . Louis , MO , USA ) were used as negative and positive controls , respectively . After acquisition of ≥ 300 , 000 events in each live gate on an LRS Fortessa or FACSCanto II flow cytometer ( BD Biosciences ) , the proportion of PBMC that had divided at least once during the culture period was calculated using the FlowJo software . Here , we subtracted the proportion of divided PBMC in medium-only wells ( “unstimulated” ) from that of antigen-containing wells to obtain net proportions of divided cells . Only samples with >80% viability were analyzed . To test the effect of in vitro CD39 blocking on lymphoproliferation , we compared the proportion of dividing PBMC after stimulation with SEB ( 10 μg/mL ) and anti-CD28/CD49d ( 10 μg/mL ) , as described above , in the absence ( medium alone ) or the presence of 2 μg/mL of anti-CD39 antibody ( Biolegend , San Diego , CA , USA ) [30] . At this concentration , this antibody was previously shown , by flow cytometry , to abolish surface CD39 recognition by the FITC-labeled anti-CD39 monoclonal antibody used in PBMC phenotyping . As a positive control , we used 2 μg/mL of a neutralizing anti-IL-10 antibody ( Biolegend ) . Separate lymphoproliferation assays were carried out in parallel with the addition of 2 mM ATP or adenosine ( both from Sigma-Aldrich ) to cells that had been pre-incubated with the blocking anti-CD39 antibody . We used standard intracellular cytokine staining to evaluate the production and accumulation , after antigen stimulation , of Th1- and Th2-type cytokines within the endoplasmic reticulum of CD4+ cells from Fil+ and Fil- subjects . To this end , 0 . 5 × 106 PBMC/well were cultured in U-bottomed 96-well microplates for 4 hours at 37°C in the absence ( medium alone ) or the presence of BmA ( 13 μg/mL ) or SEB ( 10 μg/mL ) combined with anti-CD28/CD49d co-stimulus ( 10 μg/mL ) . We next added 10 μg/mL of brefeldin A , to retain the cytokines within the PBMC , followed by further 19 hours at 37°C in a CO2 incubator . To evaluate how CD39 blocking affected cytokine production by CD4+ T cells , 2 μg/mL of purified anti-CD39 antibody ( Biolegend ) were added to selected wells shortly after antigen stimulation . Next , PBMC were transferred to V-bottomed 96-well microplates , for cell surface staining with anti-CD3 and anti-CD4 monoclonal antibodies for 30 min , followed by incubation with the permeabilization buffer supplied by eBioscience , for 1 hour at room temperature , and staining with monoclonal antibodies . The monoclonal antibody panel used for intracellular staining of IL-2 , TNF-α , IFN-γ , IL-4 , IL-5 , and IL-13 is listed in S4 Table . We used a FACSCanto II flow cytometer ( BD Biosciences ) to acquire ≥ 300 , 000 events in each live gate , in samples with >80% viability , and FlowJo software for data analysis . We also evaluated cytokine levels in culture supernatants harvested after 72-h incubation of PBMC with BmA ( 13 μg/mL ) or SEB ( 10 μg/mL ) combined with anti-CD28/CD49d ( 10 μg/mL ) , or medium alone . Again , 2 μg/mL of purified anti-CD39 antibody ( Biolegend ) were added to selected wells after antigen stimulation . We used the PeliKine compact ELISA kit ( Sanquin , Amsterdam , Netherlands ) to measure IL-6 , IL-13 , IL-10 , IL-4 , and IFN-γ levels . Because most continuous variables had an overdispersed distribution , results were summarized as medians and interquartile ranges ( IQR ) . Comparisons between samples from different subjects were done with nonparametric Mann-Whitney U tests ( for continuous variables ) or χ2 tests ( for proportions ) . Paired data were compared with Wilcoxon signed rank tests for continuous variables . Nonparametric correlation coefficients ( rs ) were estimated using Spearman rank correlation tests . All analyses were performed using SPSS 17 . 0 software ( SPSS , Chicago , IL , USA ) . Significance was defined at the 5% level . We used the Benjamini-Hochberg procedure [31] to control for the false discovery rate ( q ) when many comparisons were conducted in the same sample set and one or more of these tests resulted in a significant difference . We ranked all individual P values from the smallest to the largest and compared each of them to its Benjamini-Hochberg critical value , given by ( i/m ) q , where i is the rank and m is the total number of tests . We set q at 0 . 10 , meaning that we accept up to 10% of the associations with significant results being false positives . The largest P value that has P< ( i/m ) Q was considered significant , as well as all of the P values smaller than it . Calculations of the Benjamini-Hochberg critical value were done using the Excel spreadsheet available at: http://www . biostathandbook . com/multiplecomparisons . html .
Fifty microfilaremic ( Fil+ ) subjects aged between 10 and 98 years and 34 uninfected ( Fil- ) controls from the same communities aged between 7 and 84 years contributed plasma and PBMC samples . The number of ITS-2 amplicons ranged between 1 , 300 and 2 , 973 , 350 copies/μL in microfilaremics . Fil+ and Fil- subjects did not differ significantly according to their age and sex distribution , hemoglobin levels , total IgE levels and counts of white blood cells ( WBC ) , lymphocytes , T lymphocytes , and T CD4+ lymphocytes ( Table 1 ) . Moreover , clinical signs and symptoms were reported in comparable frequencies by Fil+ and Fil- individuals ( S5 Table ) . IgG4 antibodies to BmA were measured by ELISA in available plasma samples from 82 study participants . These were significantly higher in microfilaremics than in uninfected controls ( median , 3594 pg/mL vs . 1628 pg/mL , Mann-Whitney U test , P = 0 . 004 ) . Twenty-six subjects , hereafter termed IgG4H , had BmA antibody levels above the overall median; 25 of them were microfilaremic . The IgG4L group comprised of 56 subjects ( 25 of them infected ) with BmA antibody levels below the overall median . Therefore , half of the 50 Fil+ subjects but only one of the 32 Fil- subjects tested for antibodies were in the IgG4H group . Interestingly , the IgG4H and IgG4L groups had similar age and sex distribution , hemoglobin levels , and WBC counts , but the IgG4H population had significantly lower counts of lymphocytes , T lymphocytes , and T CD4+ lymphocytes ( S6 Table ) . Specific IgG4 antibody concentrations in microfilaremics correlated positively with ITS-2 amplicon copy numbers ( rs = 0 . 315 , P = 0 . 004 ) . Microfilaremics had significantly lower plasma concentrations of TNF-α ( median , 9 . 7 vs . 18 . 4 pg/mL , P = 0 . 012 ) , IL-4 ( median , 0 . 0 vs . 9 . 2 pg/mL , P = 0 . 008 ) , IL-8 ( median , 4 . 7 vs . 28 . 5 pg/mL , P = 0 . 005 ) , G-CSF ( median , 52 . 4 vs . 101 . 9 pg/mL , P = 0 . 016 ) , IL-6 ( median , 3 . 8 vs . 8 . 0 pg/mL , P = 0 . 021 ) , and MIP-1α ( median , 16 . 1 vs . 28 . 9 pg/mL , P = 0 . 016 ) , and significantly higher levels of the Th2-type mediator eotaxin ( median , 47 . 7 vs . 37 . 0 pg/mL , P = 0 . 006 ) , compared with uninfected controls ( S7 Table ) . However , the slight difference between Fil+ and Fil- subjects in IL-10 concentrations ( median , 17 . 4 vs . 13 . 7 pg/ml ) did not reach statistical significance ( P = 0 . 052 ) . Levels of eotaxin , but not those of other cytokines or chemokines , were significantly correlated to the proportion of circulating CD4+ cells that were CD39+ ( rS = 0 . 356 , P = 0 . 002 ) and to the proportion of circulating Treg cells that that were CD39+ ( rS = 0 . 274 , P = 0 . 022 ) . We found no evidence for lymphocyte hyporesponsiveness in M . ozzardi infection . Fig 1 shows comparable proliferation patterns of PBMC from microfilaremic and uninfected subjects that were stimulated with filarial ( BmA ) and two bystander antigens ( TgT and PvS ) , as well as with mitogen ( PHA-P ) . However , SEB-driven PBMC proliferation was significantly increased , rather than decreased , in microfilaremics , with median proportion of dividing cells of 45 . 2% vs . 31 . 7% ( P = 0 . 012 ) in Fil+ and Fil- subjects , respectively . Moreover , antigen-driven cytokine responses did not differ significantly between infected and uninfected subjects . We found comparable proportions of CD4+ T cells from Fil+ and Fil- individuals producing IFN-γ , IL-2 , IL-10 , TNF-α , and Th2-type cytokines ( IL-4 , IL-5 , and IL-13 ) in vitro following stimulation with BmA or SEB ( S8 Table ) . Finally , we found similar concentrations of secreted IL-6 , IL-13 , IL-10 , IL-4 , and IFN-γ in PBMC culture supernatants from Fil+ and Fil- subjects stimulated with BmA and SEB ( S9 Table ) . We compared the expression of regulatory and activation markers in circulating CD4+ T cells from Fil+ and Fil- subjects , using the gating strategy shown in S2 and S3 Figs . Fil+ and Fil- subjects had similar frequencies of CD4+ T cells expressing most regulatory and activation markers ( Table 2 ) . However , microfilaremics had a significantly higher frequency of CD4+ T cells expressing CD39 than uninfected controls ( median , 7 . 2% vs . 5 . 5% of all circulating CD4+ T cells ) . A similar difference in the frequency of CD4+ T cells expressing CD39 was observed when comparing IgG4H and IgG4L subjects ( S10 Table ) . We next examined the classical Treg cell compartment , phenotypically defined as CD4+CD25hiCD127-FoxP3+ cells , using the gating strategies shown in S4 and S5 Figs . Treg cells comprised identical proportions ( median , 1 . 5% ) of CD4+ T cells circulating in Fil+ and Fil- subjects , indicating that M . ozzardi infection did not induce an expansion of the Treg cell compartment . However , the analysis of surface marker expression on Treg cells revealed a significantly greater proportion of CD39+ Tregs in microfilaremics compared to uninfected controls ( median , 69 . 7% vs . 60 . 9% ) . CD39+ Treg cells accounted for a median of 1 . 1% and 0 . 9% of all CD4+ T cells circulating in infected and uninfected individuals , respectively . Therefore , ~15–16% of CD39+ cells within the CD4+ compartment were classical Treg cells . No significant difference was found in the frequencies of Treg cells expressing all other molecules listed in Table 3 , except for the CD39-PD-1+ subset ( increased in Fil+ subjects ) . Overall , similar results were obtained when comparing IgG4H and IgG4L subjects ( S11 Table ) . Interestingly , CD39+ Treg cells from microfilaremics and uninfected controls more often co-express a wide range of other regulatory molecules ( CTLA-4 , LAP-TGF-β , LAG-3 , TNFRII , GITR , and OX-40 ) and activation markers ( HLA-DR and CD69 ) than do CD39- Treg cells ( S6 Fig ) . These findings are consistent with the notion that CD39 expression characterizes a functionally more “suppressive” Treg subtype [32] . Because the frequencies of CD39+ CD4+ T cells and CD39+ classical Treg cells correlated positively with the number of M . ozzardi-specific ITS-2 amplicon copies in microfilaremics ( Fig 2 ) , we conclude that the M . ozzardi-associated expansion of CD39+ T-cell populations is directly proportional to the microfilarial load . To explore the immunomodulatory consequences of surface expression of CD39 in circulating CD4+ T cells , we used an anti-CD39 antibody to neutralize this molecule [30] . We observed that SEB-driven lymphocyte proliferation decreased significantly in the presence of CD39-blocking antibody ( Fig 3A ) , although proliferation increased , as expected [2] , in the presence of neutralizing anti-IL-10 antibody ( Fig 3B ) . SEB-induced lymphocyte proliferation in the presence of anti-CD39 antibody was further inhibited by adding 2mM adenosine to the culture medium ( Fig 3C ) , being partially restored by adding 2mM ATP ( Fig 3D ) . Interestingly , comparable inhibitory effects of CD39 blocking on cell proliferation in vitro were observed in Fil+ and Fil- subjects . Antibody-mediated CD39 neutralization also exerted profound effects on SEB-driven cytokine production . We observed increased proportions of CD4+ T cells producing IFN-γ , IL-2 , TNF-α , and Th2-type cytokines , but a decreased proportion of CD4+ T cells producing IL-10 , in the presence of anti-CD39 antibody ( Fig 4 ) . These changes were reversed in the presence of 2mM adenosine ( S7 Fig ) . Accordingly , levels of IL-4 , and IFN-γ in culture supernatants increased , while those of IL-13 , and IL-10 significantly decreased , following CD39 blocking in vitro ( Fig 5 ) . CD39 blocking caused similar changes in SEB-driven cytokine responses in Fil+ and Fil- subjects ( Figs 4 and 5 ) , although not all comparisons reached statistical significance due to the relatively small sample sizes . Intracellular Ki67 expression was measured to explore the effects of CD39 on T-cell proliferation . We first observed an increased expression of Ki67 by CD39+ CD4+ T cells , compared to their CD39- counterparts . This change was statistically significant in microfilaremics but not in uninfected controls , although the trends were similar ( Fig 6A ) . We also found an increased expression of Ki67 in the CD39+ subset of Treg cells , compared to CD39- Treg cells , in microfilaremics ( Fig 6B ) . These and previous findings [32] indicate that CD39 expression delineates CD4+ T cell subsets with enhanced proliferative ability . Indeed , CD39 blocking significantly reduced the proportion of Ki67+ CD39+ CD4+ T cells ( Fig 6C ) and Ki67+ CD39+ Treg cells ( Fig 6D ) . The trends were similar regardless of the infection status , but only reached statistical significance among microfilaremics . This further confirms the suppressive effect of CD39 neutralization on T-cell proliferation .
The filarial nematode M . ozzardi challenges the common view that chronic helminth infections will necessarily elicit immunomodulatory responses such as those thoroughly described in schistosomiasis , lymphatic filariasis , and intestinal nematode infections [1 , 3 , 4] . We observed decreased plasma levels of some inflammatory cytokines in microfilaremics , compared to uninfected controls living in the same communities , but Fil+ and Fil- groups had similar proportions of CD4+ T cells producing IFN-γ , IL-2 , IL-10 , TNF-α , and Th2-type cytokines , as well as similar levels of secreted IL-6 , IL-13 , IL-10 , IL-4 , and IFN-γ in culture supernatants , following antigen stimulation . A concomitant increase in plasma concentrations of both inflammatory ( IL-6 ) and regulatory ( IL-10 ) cytokines was recently described in M . ozzardi infections in Brazil [33] , but further comparisons with our data are limited by the lack of concurrent analyses of antigen-driven cytokine production in vitro [33] . Moreover , T-cell responses to filarial and unrelated antigens are not attenuated in subjects harboring M . ozzardi microfilariae . Our data suggest that M . ozzardi infection , which is highly prevalent in riverine communities across the Amazon Basin [8] , is unlikely to suppress T-cell responses to co-occurring pathogens , such as malaria parasites , in endemic populations . How Treg cells exert their suppressive effects remains incompletely understood , but CD39 is thought to play a crucial immunoregulatory role in these cells [15] . Indeed , surface expression of CD39 appears to confer enhanced proliferative and suppressive ability to induced Treg cells [32 , 34] . Moreover , CD39 boosts the differentiation of type 1 regulatory ( Tr1 ) cells , characterized by the production of IL-10 and lack of FoxP3 expression , which are able to limit inflammation and favor immune tolerance [34] . CD39 hydrolyses extracellular ATP ( eATP ) and ADP into AMP , while another ectonucleotidase , CD73 , dephosphorylates AMP to adenosine . eATP activates P2 purignergic and pyrimidinergic receptors expressed by T cells , B cells , dendritic cells , macrophages , and neutrophils , triggering a range of proinflammatory responses . Adenosine , in contrast , is a labile molecule that binds to the adenosine receptor 2A expressed on effector T cells and dampens cell proliferation and inflammatory cytokine secretion [15] . Adenosine concentrations are further regulated by adenosine deaminase ( ADA ) , which catalyzes the deamination of adenosine generating inosine and ammonia . Human ADA1 binds CD26 on T-cells , favoring adenosine turn-over [35] . Therefore , the interplay of CD39 with CD73 and CD26 regulates the levels of eATP , ADP , AMP , and adenosine , with major consequences for the purinergic control of inflammation and adaptative immune responses [34] . Increased expression of CD39 by circulating Treg cells has been associated with disease progression in tuberculosis [36] and HIV/AIDS [37–39] . Moreover , CD39 is overexpressed by several types of cancer cells , in addition to tumor-infiltrating T cells , suggesting that CD39 and purinergic signaling can directly modulate tumor growth , metastasis and angiogenesis , in addition to inducing immune tolerance , further favoring cancer progression [40] . Indeed , blocking CD39 in vivo with specific antagonists or monoclonal antibodies currently in preclinical development has been suggested as an immunomodulatory intervention to enhance effector T-cell responses in HIV infection [37 , 38] and in cancer [40] . However , the pool of circulating CD4+ T cells expressing CD39 comprises not only Treg cells with enhanced co-expression of regulatory markers and greater suppressive ability . In fact , 5–7% of all circulating CD4+ T cells expressed CD39 in our study subjects , and CD39+ Treg cells represented only 15–16% of the pool of circulating CD39+CD4+ T cells in microfilaremic and uninfected study subjects . Moreover , CD39 is constitutively expressed by the vast majority ( >90% ) of B cells and monocytes and by small proportions ( 5% or less ) of CD8+ T cells [40] . A particularly interesting subset of CD39+CD4+ T cells comprises the partially characterized CD25-FoxP3- “inducer” T ( Tind ) cells that promote and potentiate , rather than suppress , T-cell proliferation and inflammatory cytokine production [41 , 42] . Tind cells appear to be functionally similar to newly described activated effector CD4+ T cells that co-express CD39 and CD26 , have no suppressive effect , and are prone to apoptosis [43] . Co-culture with Tind cells enhances the proliferation of CD4+CD25-CD39- “responder” T cells ( Tres ) from healthy donors , an effect that can be partially reversed by anti-CD39 monoclonal antibodies [41] . Moreover , co-culture of Tres with Tind cells enhances the production of IFN-γ , TNF-α , GM-CSF , IL-6 , and IL-10 [41] . These findings imply that increased CD39 expression and regulatory ability are not necessarily coupled in all circulating CD4+ T cell subsets . The dramatic decrease in antigen-driven cell proliferation observed in both microfilaremic and uninfected individuals in the presence of CD39-blocking antibody ( Fig 3A ) is consistent with a predominance of Tind cells in the CD39+CD4+ pool of circulating T cells from our study population . Therefore , the relative proportions of CD39+ Treg and Tind cells may determine the patterns of immune homeostasis observed in human populations exposed to different environmental antigens and pathogens . In addition , this balance may determine the outcome of immunomodulatory interventions based on CD39 blocking . The main limitation of this study is its cross-sectional design . Therefore , we were able to identify statistically significant associations between current infection and the proportion of certain CD4+ T cell populations circulating in the peripheral blood , but our study design does not allow us to infer causal relationships . Moreover , phenotypic characterization of CD4+ T cells is restricted to the circulating compartment , which may not be representative of the T cell population in infected subjects . Analyses of post-treatment samples could theoretically help to delineate infection-related changes in T-cell responses that can be reversed in the absence of helminth-derived antigenic stimulation . However , given that ivermectin , the first-line treatment for M . ozzardi infection , is highly efficacious against microfilariae but does not appear to kill adult worms [8 , 19] , PBMC samples collected after treatment may still be stimulated by circulating excreted-secreted soluble antigens that are chronically released by adult worms . In conclusion , it is tempting to speculate that the balance between suppressive CD4+CD39+ Treg cells and immunostimulatory CD4+CD25-FoxP3-CD39+ Tind cells is a key factor contributing to the unexpected patterns of immune regulation found in M . ozzardi infection . We suggest that an increased frequency of Tind cells in our microfilaremics might have prevented lymphocyte hyporesponsiveness despite the concomitant expansion of the CD4+CD39+ Treg subset . However , the CD39+ cell pool remains uncharacterized in other chronic helminth infections that induce typical immunomodulatory responses . Therefore , further studies on CD39+ cells may help to unveil some of the biochemical and molecular pathways whereby different helminths manipulate their hosts’ immunity .
|
Helminth infections downregulate immunity and reduce host’s inflammatory responses , but the filarial nematode Mansonella ozzardi , which is widely distributed across Latin America , appears to represent an exception to this rule . We found similar lymphoproliferative responses to filarial and unrelated antigens and comparable regulatory cytokine responses in subjects harboring M . ozzardi microfilariae , compared to local uninfected controls . The proportion of CD4+ T cell subtypes expressing CD39 was significantly increased in infected subjects and correlated positively with their microfilarial density . However , antibody blocking of CD39 , an ectonucleosidase involved in the synthesis of the immunosuppresive molecule adenosine , paradoxically reduced , rather than promoted , antigen-driven lymphoproliferation in vitro . We suggest that CD39+ CD4+ T cells circulating in microfilaremics comprise both regulatory and stimulatory cell subsets that are concomitantly expanded . The balance between these cell subsets with opposing regulatory functions may be crucial to maintain immune homeostasis during chronic M . ozzardi infections .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"glycosylamines",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"immunology",
"parasitic",
"diseases",
"nematode",
"infections",
"developmental",
"biology",
"molecular",
"development",
"adenosine",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"lymphocytes",
"immune",
"system",
"biochemistry",
"helminth",
"infections",
"cell",
"biology",
"physiology",
"nucleosides",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"regulatory",
"t",
"cells",
"glycobiology"
] |
2018
|
CD39 and immune regulation in a chronic helminth infection: The puzzling case of Mansonella ozzardi
|
A role for natural selection in reinforcing premating barriers is recognized , but selection for reinforcement of postmating barriers remains controversial . Organisms lacking evolvable premating barriers can theoretically reinforce postmating isolation , but only under restrictive conditions: parental investment in hybrid progeny must inhibit subsequent reproduction , and selected postmating barriers must restore parents' capacity to reproduce successfully . We show that reinforced postmating isolation markedly increases maternal fitness in the fungus Neurospora crassa , and we detect the evolutionary genetic signature of natural selection by quantitative trait locus ( QTL ) analysis of the reinforced barrier . Hybrid progeny of N . crassa and N . intermedia are highly inviable . Fertilization by local N . intermedia results in early abortion of hybrid fruitbodies , and we show that abortion is adaptive because only aborted maternal colonies remain fully receptive to future reproduction . In the first QTL analysis of postmating reinforcement in microbial eukaryotes , we identify 11 loci for abortive hybrid fruitbody development , including three major QTLs that together explain 30% of trait variance . One of the major QTLs and six QTLs of lesser effect are found on the mating-type determining chromosome of Neurospora . Several reinforcement QTLs are flanked by genetic markers showing either segregation distortion or non-random associations with alleles at other loci in a cross between N . crassa of different clades , suggesting that the loci also are associated with local effects on same-species reproduction . Statistical analysis of the allelic effects distribution for abortive hybrid fruitbody development indicates its evolution occurred under positive selection . Our results strongly support a role for natural selection in the evolution of reinforced postmating isolation in N . crassa .
The geographical ranges of N . crassa and N . intermedia are broadly overlapping , and individuals of both species can be collected from the same site [15] , [17] , [18] . Both species are largely outbreeding , and outbreeding is confirmed by population genetic analysis [19]–[22] . Hybrids of the two species can be obtained in laboratory crosses , but natural hybrids have not been encountered [15] . This absence may reflect the rarity of hybridization in nature , the low viability of hybrids , or both . Nevertheless , phylogenetic conflict between some gene trees and the species tree of the Neurospora genus indicates historical hybridization and introgression between N . crassa and N . intermedia [23] . In Neurospora , mating can occur only between individuals having different alleles at the mating-type locus ( mat a or mat A ) . Under nutrient limited conditions , a haploid colony of Neurospora differentiates female reproductive structures ( protoperithecia ) . Fertilization occurs when a specialized hypha ( trichogyne ) growing from a protoperithecium fuses with a cell from a colony of the opposite mating type . The attraction of trichogynes to fertilizing cells is mediated by mating-type specific pheromomones . After fertilization , nuclei from the fertilizing strain travel through the trichogyne to the protoperithecium , where karyogamy eventually occurs . A series of independent meiotic events give rise to the sexual progeny ( ascospores ) , which develop within flask shaped fruitbodies ( perithecia ) on the maternal thallus . Upon maturity , the ascospores are forcibly ejected from the fruitbody . In Neurospora , evolution of premating isolation is apparently constrained because the sequences of the mating-type–specific , peptide pheromones controlling attraction between trichogynes and fertilizing cells are conserved throughout the genus ( as determined by BLAST analysis of the N . crassa , N . tetrasperma , and N . discreta genomes [24] ) and even beyond [25] . Evolution of the extracellular , ligand-binding portions of mating-pheromone receptor proteins is also comparatively constrained [26] . In Neurospora , sex cells of mating-type–compatible partners usually fuse before incompatibilities are expressed , and incompatibility arises either prezygotically in the fusion cell and the subsequent dikaryotic cells that proliferate from it , or postzygotically during the meiosis that directly follows karyogamy and during the formation and development of the ascospores [17] . Since Neurospora progeny develop within fruitbodies composed entirely of maternal tissue , the maternal colony ( mycelium ) bears virtually the entire cost of reproduction . Because 98% of N . crassa×N . intermedia hybrid progeny are inviable [15] , and because allocation of resources to developing fruitbodies on one part of the colony abolishes the fertility of uncrossed regions of the colony [27] , hybridization is severely maladaptive . The key questions are: 1 ) Does abortion of hybrid fruitbodies by N . crassa make subsequent reproduction possible for the maternal colony , thereby conferring a fitness advantage ? and 2 ) Did this postmating barrier evolved by natural selection ? First we show that hybrid fruitbody abortion is adaptive , because it preserves the fertility of maternal N . crassa . Then by quantitative trait locus ( QTL ) analysis of hybrid fruitbody abortion and statistical analysis of the allelic effects distribution for the detected loci , we show that the genetic architecture that we observe is consistent with evolution by positive natural selection .
We tested whether female fertility of colonies was preserved after abortion of sympatric hybrid fruitbodies in sequential mating experiments between the two Neurospora species . N . crassa colonies were grown in Petri plates on synthetic cross agar medium , which promotes sexual reproduction . One half of the receptive N . crassa colony was fertilized by either an allopatric or sympatric N . intermedia strain , and the effect of allopatric vs . sympatric fertilization on the reproductive success of subsequent conspecific fertilization on another portion of the maternal colony was assayed . Initial fertilizations by a conspecific strain or with water ( pseudo-fertilizations ) were also performed as controls . Each of the four initial-fertilization treatments ( allopatric male , sympatric male , conspecific male , pseudo-fertilization ) was replicated three times . Following Dettman , et al . , reproductive success was scored on a seven-category scale incorporating fruitbody development and quality of ejected ascospores , if any [15] . The N . crassa and N . intermedia strains used are described in the Materials and Methods and listed in Table 1 . In all sequential matings , both the maternal strain and conspecific fertilizing strains were N . crassa from the NcC clade endemic to India , hereafter referred to as NcC-India [15] . Fruitbody development on portions of the maternal colony fertilized by allopatric N . intermedia strains is normal and results in ascospore ejection , although a majority of the hybrid ascospores are unmelanized and inviable ( reproductive success score ( RSS ) = 4 . 33±0 . 14 , Figure 1 ) . Following allopatric hybridization , response to conspecific fertilization at the second time point is completely inhibited ( RSS = 0 . 00±0 . 00 ) . In contrast , fertilization by sympatric N . intermedia strains at the first time point yields only aborted fruitbodies ( RSS = 1 . 00±0 . 00 ) , but the subsequent conspecific matings are fully fertile ( RSS = 6 . 00±0 . 00 , Table 2 ) . We performed semi-parametric regression using a proportional hazards survival model [28] to examine the effects of fertilization at the first time point ( water control , conspecific control , or allopatric or sympatric heterospecific ) on progression through the sexual cycle after fertilization at the second time point ( measured as RSS ) . The first–time-point fertilization treatment had a significant effect ( P<0 . 0028 ) , whereas the nested effects of geographic origin and strain identity of the first–time-point male do not have significant effects ( Table 3 ) . We conclude that abortion of hybrid fruitbodies is selectively advantageous because abortion preserves maternal fertility of the colony after hybridization . Previous research on the genetics of reinforcement focused on premating barriers in animals [29]–[31] . Here we investigate whether the genetic architecture of a reinforced , microbial , postmating barrier is consistent with evolution by directional natural selection [32] , [33] . A 500-member , N . crassa mapping population was derived from an intraspecific , inter-clade cross between the NcC-India strain FGSC 8866 , and a Louisiana , USA , strain , FGSC 8903 , a member of the NcA clade , hereafter referred to as NcA-Louisiana . Neurospora are hermaphroditic so that the parental strains could be mated reciprocally . Based on the identity of the maternal parent we can infer that the mapping population contains both individuals with NcA-Louisiana cytoplasm ( 57% ) and individuals with NcC-India cytoplasm ( 43% ) . The N . crassa mapping strains were crossed maternally and paternally to N . intermedia strains from Tamil Nadu , India , which are sympatric to the NcC-India parent and allopatric to the NcA-Louisiana parent . NcC-India aborts fruitbodies after fertilization by N . intermedia from India , but NcA-Louisiana does not [15] , [16] . The mapping strains were also crossed maternally and paternally to African N . intermedia strains , which are allopatric to both the NcC-India and NcA-Louisiana parents . Neither the NcA-Louisiana parent nor the NcC-India parent aborts fruitbodies after fertilization by the African N . intermedia strain . The four traits that we studied were fruitbody development in the four different types of matings . The four types of matings are defined by the parental role of the mapping strain ( maternal or paternal ) and the geographic relationship of the N . intermedia strain ( Indian and therefore sympatric to the NcC-India parent of the mapping population , hereafter termed “sympatric”; or African , and therefore allopatric to both the NcC-India parent and the NcA-Louisiana parent , hereafter termed “allopatric”; see Table 4 ) . Therefore each member of the mapping population was crossed twice to an Indian N . intermedia strain and twice to an African N . intermedia strain , once with the mapping strain in the maternal role and once with the mapping strain in the paternal role , for a total of four crosses per mapping strain . We examined fruitbody development in each cross , recording its fruitbody development score ( FDS ) 10 days after fertilization ( see Materials and Methods ) . The mapping strains were genotyped at 69 AFLP ( Table 5 and Table 6 ) and 28 microsatellite ( Table 7 ) markers as well as the mat locus . A genetic map containing seven linkage groups ( LG ) reflecting the seven chromosomes of Neurospora was estimated , with a total map length of 837 . 9 cM and an average intermarker distance of 9 . 2 cM ( Figure 2 ) . Non-Mendelian segregation and non-random associations ( NRA ) among alleles at multiple loci can reflect genetic incompatibilities between the NcC and NcA genomes . Individuals in the mapping population inherited 53 . 6% of genotyped alleles from the NcA-Louisiana parent , and this is significantly higher than the 50% expected under Mendelian segregation ( t Test , Pt = 6 . 8721 , DF = 93<0 . 0001 ) . The proportion of NcA-Louisiana alleles inherited varied across linkage groups ( ANOVA , PF ratio = 23 . 34 , DF = 93<0 . 0001 ) , with linkage groups II , VI , and VII showing below 50% inheritance of NcA-Louisiana alleles ( Figure 3 ) . The skew towards NcA alleles was strongest on linkage group I , with 59 . 0% of marker alleles inherited from that parent . Of 22 markers showing significant segregation distortion ( χ2 , P<0 . 05 ) , only one marker on LG IV ( nc4L4 ) was significantly skewed in favor of the NcC-India allele , while the 21 other significantly distorted markers favored the NcA-Louisiana and were located on LG I ( 16 markers ) and LG IV ( 5 markers ) . Seven marker pairs representing four pairs of linkage groups exhibit significant non-random associations ( Table 8 ) . Positive non-random associations , reflecting an overrepresentation of parental haplotypes , are consistent with the existence of negative epistasis among clade specific alleles , as predicted by the Dobzhansky-Muller ( D-M ) model of the evolution of genetic incompatibilities first articulated by Bateson [34] . Per this model , incompatibilities between two loci , X and Y , must arise in in a two-step fashion , as follows: Consider that at the time that the NcA and NcC clades diverged , both populations contained ancestral ( anc ) alleles at every locus . The first fixation of a derived allele , e . g . , Xanc→XNcA , will not cause incompatibility , because XNcA arises in an ancestral genetic background and must also be compatible with the ancestral alleles that comprise the NcC genome . However the next derived allele to be fixed by either population can be a source of incompatibility because the genetic backgrounds of the two clades are no longer identical . For example , if we consider fixation of a new Y allele in the NcC population , Yanc→YNcC , we see that YNcC evolves in the presence of Xanc , and may not be compatible with the previously derived allele XNcA in the NcA clade . Conversely , fixation of a new Y allele in the NcA population , Yanc→YNcA can also give rise to incompatibilities , since YNcA need not be compatible with the Xanc allele , which is still fixed in the NcC population . ” Given partial intersterility between NcA-Louisiana and NcC-India and the segregation distortion that we observed , we predicted that observed significant non-random associations would be positive , indicating a deficit of recombinant haplotypes . However , only one significant marker pair ( markers nc2L1 and tnf223 from LG II and IV , respectively ) showed positive non-random associations ( D' = 0 . 19 ) . Surprisingly , the loci on the three other significant linkage-group pairs ( III and V , V and VI , and VI and VII ) showed negative , non-random associations , with D' values ranging from −0 . 35 to −0 . 15 , implying overrepresentation of recombinant haplotypes . Overrepresentation of recombinant haplotypes could be evidence that an optimal N . crassa genome would include some mixture of clade-specific alleles . Indeed , a population genomic study of N . crassa NcA from the Caribbean Basin and the south-eastern United States identified a haplotype in the Louisiana population consisting of a four-gene “migrant tract” originating from an unidentified , genetically diverged population or species , concluding that this tract was fixed in the Louisiana population via a selective sweep [35] . Alternatively , if alleles are not fixed in the clades , overrepresented recombinant haplotypes could be analogous to selectively advantageous , intrapopulation haplotypes . We used composite interval mapping to identify QTLs for fruitbody development . The genetic basis of postmating reinforcement was revealed by mapping loci for maternal influence on sympatric fruitbody development ( trait A , see Table 4 ) . We identified 11 additive-effect QTLs for this trait ( Figure 4a and Table 9; complete CIM scans for all traits are in Figure 5 ) . Seven of the QTLs are located on LG I , including one of large effect , while LG II and V each contain a single broad QTL region of weak effect , and the left arm of LG VI harbors two other QTLs of large effect . The detected QTLs account for roughly 61% of trait variance , with the three loci of large effect accounting together for roughly 30% of trait variance . The inferred cytoplasm type ( NcA-Louisiana or NcA-India ) of the mapping strains did not significantly affect this trait . For 10 of the 11 QTLs , the allele from the sympatric NcC-India parent has a negative effect on sympatric fruitbody development . Only the weak QTL on LG II has the opposite effect . The prevalence of negative alleles in the NcC-India background is consistent with evolution of abortion via directional natural selection . This inference can be statistically tested by the QTL sign test , which tests the null hypothesis that the observed genetic architecture was generated during neutral trait evolution , i . e . , without selective advantage for negative alleles [32] . The QTL sign test assumes that all QTL effects are additive . We note that the accepted model of intrinsic , postzygotic isolation barriers involves negative among species-specific alleles in hybrids . However , our experiments were not designed to , nor can they , interrogate the genetics of hybrid dysfunction , but rather the genetics of evolutionary response to maladaptive hybridization . The genetic architecture of reinforcement may or may not involve epistatic effects . However , in contrast to the case of hybrid dysfunction , epistasis would involve within-genome , interaction effects among loci contributing to the reinforced barrier . To determine whether or not epistasis plays a role in the genetics of reinforced sympatric barriers in N . crassa , and to determine whether or not the genetic data conform to the assumptions of the QTL sign test , we performed a two-dimensional genome scan for interacting QTLs . No significant interaction effect was detected . Moreover , the genome-wide maximum LOD score for any interaction effect was determined to be 16 . 7 , well below the critical LOD score of 37 . 4 ( for a Type I error of 0 . 05 ) , which was estimated from 1000 permutations of the data . Because the two-way scan for genetic interactions among loci failed to find any significant or marginally significant interaction effects , the data are consistent with an additive genetic model and conform to the assumption of additivity required for the QTL sign test . Given the observed number of positive and negative QTLs and the distribution of their effect magnitudes ( gamma b = 0 . 034 , c = 13 . 5 ) , and conditioned on the parental difference in fruitbody development after sympatric fertilization ( F . D . S = 2 ) , the null hypothesis of neutral trait evolution for abortion in NcC-India is rejected ( QTL sign test , P = 0 . 0099 ) ( Figure 4b ) . This result implies that fruitbody abortion in sympatric NcC-India–maternal×N . intermedia-paternal crosses ( trait A ) evolved under positive natural selection via a reinforcement mechanism . One of the major QTLs on LG VI is flanked by two microsatellite loci , nc6L15 and nc6L16 , and can therefore be physically located to a 135 , 874 bp region of the N . crassa genome containing 24 ORFs ( Table S1 ) [36] . We also investigated the genetics of fruitbody development in crosses not showing reinforced isolation ( Traits B , C , and D , see Table 4 ) . Note that NcC-Indian strains show enhanced isolation from Indian N . intermedia only when the NcC-Indian strain performs the maternal role [16] . We did not detect any QTLs for paternal influence on the development of sympatric fruitbodies ( trait B ) , but in crosses to allopatric N . intermedia strains we detected two QTLs affecting maternal influence and three QTLs affecting paternal influence on fruitbody development ( traits C and D , respectively; Figure 4a ) . All five of these loci are located on LG I . Four of the five allopatric QTLs co-localize with three of the sympatric QTLs on LG I , which could either indicate the presence of genes with pleiotropic effects or linked genes with trait-specific effects .
Linkage group I represents less than 24% of the genome , and it is striking that 75% of our QTLs map here . Interestingly , 73% of loci showing non-Mendelian segregation also map to linkage group I , so that every QTL on this linkage group is flanked by at least one marker showing segregation distortion . In all cases , the NcC alleles of the linkage group I QTLs have a negative effect on sympatric hybrid fruitbody development , and the NcC alleles of the linkage group I markers are underrepresented in the NcA×NcC N . crassa mapping population . It is not known why one-fifth of genetic markers are distorted in favor of the NcA allele . It is possible that , because laboratory strains of N . crassa have historically been derived from the NcA clade , our crossing and progeny-isolation methods , which were developed for NcA-clade N . crassa , have inadvertently created a selective environment favoring NcA alleles at loci linked to distorted markers . It is also possible that distorter loci present in the NcA background are normally repressed through the action of NcA modifier loci , but become unrepressed and active in some recombinant NcA×NcC genotypes . Although segregation distortion can be caused by nuclear-cytoplasm incompatibilities , it is unlikely to be the cause in this case . The Neurospora mapping population comprises a mixture of individuals with NcA-Louisiana cytoplasm and NcC-India cytoplasm . Moreover , even in individuals with NcC-India cytoplasm , linkage group I markers are distorted in favor of the NcA allele , with an overall NcA allele frequency of 0 . 61 for markers on this linkage group . Another hypothesis is that reinforcement alleles themselves can pleiotropically cause ascospore inviability in conspecific , inter-clade crosses . Laboratory crosses between members of the NcA and NcC clades are partially intersterile , usually resulting in <50% ascospore viability [15] . In Neurospora , all products of meiosis contribute to the ascospore cohort , so segregation distortion most likely results from inviability of hybrid ascospores carrying the disfavored NcC allele ( s ) . Pleiotropy for reinforcement and reproductive isolation between allopatric conspecifics has previously been observed in animals [37] , [38] . Reduced conspecific fertility can present a challenge to the evolution of reinforced barriers , since the fitness advantage of avoiding hybridization must outweigh the cost of lower conspecific fertility . However , restricted migration between conspecific populations should reduce the incidence of interpopulation mating and reduce the fitness costs associated with pleiotropic effects on conspecific fertility . The NcC and NcA clades are geographically separated , so limited interclade migration would reduce the fitness cost to NcC of lower fertility with NcA and facilitate the spread of reinforcement alleles in the NcC clade . If the pleiotropy hypothesis is correct , the evolution of reinforcement QTL on linkage group I could be a partial explanation for the evolution of incomplete reproductive isolation between the NcC and NcA clades . Notably , QTLs on other linkage groups ( II , V and VI ) are flanked by markers showing Mendelian segregation , so that for these QTLs there is no suggestion of pleiotropic negative effects on within-species , inter-clade reproduction . Moreover , the two reinforcement QTLs on linkage group VI lie in the vicinity of three markers ( tnc088 , nc6L13 , nc6L2 ) , which participate in non-random associations with loci on other linkage groups , such that recombinant , non-parental haplotypes are overrepresented in the mapping population . Therefore the patterns of marker inheritance near QTLs on these other linkage groups do not suggest any connection between reinforcement QTLs and isolation of conspecific allopatric N . crassa populations . We note that linkage group I contains the mating-type locus of Neurospora , and that some studies have found that reproductive isolation loci are more prevalent on sex-determining chromosomes than on autosomes [39] , [40] . It is true that in a closely related species , N . tetrasperma , recombination is suppressed over a large region of Linkage Group I in a process considered analogous to an early stage of sex-chromosome evolution [41] . However , no recombination block exists on linkage group I of N . crassa . Additionally , Neurospora species are hermaphroditic , so the mating-type locus determines mating compatibility , rather than sexual role . We therefore consider it unlikely that the same forces that cause reproductive isolation loci to preferentially accumulate on sex chromosomes can account for the prevalence of the observed QTLs on linkage group I . Earlier genetic studies of reproductive isolation in Neurospora identified a QTL on linkage group I as the N . crassa member of a Dobzhansky-Muller incompatibility locus-pair responsible for a severe defect in hybrid perithecial development between allopatric N . crassa and N . intermedia when N . crassa acts as the male partner [42] , [43] . These incompatibility loci were first identified in populations of N . crassa×N . intermedia hybrids evolved under divergent environmental conditions in a test of the hypothesis that ecological adaptation can incidentally drive reproductive isolation [42] . Subsequent mapping determined that the incompatibility was caused by interactions between an N . crassa locus ( dma on linkage group I ) and an N . intermedia locus ( dfe on linkage group V ) [43] . Considering that both the geographic relationship of the species and gender role of N . crassa differ between this study and ours , it is very interesting that the N . crassa dma locus maps to a region of linkage group I that potentially coincides with the locations of QTLs identified in our study . Direct comparison between mapping results is prevented by the absence of sequence anchored , microsatellite markers in this region of our map . Sexual microbes are likely to have simple premating recognition mechanisms , but will nevertheless experience selective pressure to avoid maladaptive hybridization . When evolution of premating barriers is constrained , microbial reinforcement may be more likely to involve non-premating-recognition mechanisms , including differentiated substrate or host fidelities [10] or evolution of divergent mating kinetics [44] . Here we have shown that selective abortion of hybrid fruitbodies by N . crassa fertilized by sympatric N . intermedia had the potential to evolve by natural selection by demonstrating that maternal colonies that abort hybrid fruitbodies are subsequently able to mate normally with conspecifics and can have higher reproductive fitness . We then show that the genetic architecture of hybrid fruitbody abortion is consistent with evolution via directional natural selection . Plants and animals are known to sometimes selectively abort otherwise viable embryos , thereby restricting parental investment to offspring with higher potential fitness [45] . Our data show that microbes like Neurospora , which provide costly parental care and are capable of multiple matings , are capable of undergoing reinforcement selection for selective abortion of hybrid offspring . Further studies on the evolution and genetics of reproductive isolation in microbial eukaryotes will be needed to challenge this hypothesis .
The biology of Neurospora , the evolutionary relationships among species and clades , and the biogeography of reproductive isolation between N . crassa and N . intermedia have been described previously [15] , [16] . Culturing , crossing , and isolation of ascospore progeny were performed as previously described [15] , [17] . Table S1 lists the wild-collected Neurospora strains used in this study . The QTL mapping population created for this study has been deposited with the Fungal Genetics Stock Center , Kansas City , Missouri . Sequential fertilization was performed according to the methods of Howe and Prakash [27] , except that at the first mating time point ( 5 days after inoculation of the NcC-India maternal strain ) , the conidial suspension of one fertilizing strain ( either an NcC-India strain as conspecific positive control; an allopatric N . intermedia ( African ( n = 2 ) or Caribbean ( n = 2 ) ) ; sympatric N . intermedia ( Indian ( n = 2 ) ) ; or water negative control ) was applied to 50% of the plate , while at the second time point ( 10 days after maternal inoculation ) , the fertilizing suspension of NcC-India was applied to two 1 cm2 spots located 1 . 5 cm–2 cm from the edge of the first fertilization . Three replicates were performed for a total of 24 plates . Fertility was scored 20 days after maternal inoculation using a 0–6 reproductive success scale ( RSS ) [15] . The effects of cross type and geographic origin and strain identity of the first–time-point fertilizing males on reproductive success of the second–time-point crosses were analyzed using a semi-parametric , proportional hazards model , with nested effects , as implemented by JMP 5 . 0 . 1a . We obtained genomic DNA for each member of the QTL mapping population following the protocol of Dettman et al . [15] . AFLP and microsatellite primer sequences are shown in Table 5 , Table 6 , and Table 7 . The mat-a1 and mat-A1 loci were amplified by multiplex PCR with the following primers: Ba1-5 , AAGAAGAAGGTCAACGGCTTCATG; Ba1-3 , CCAGAGCCATGTTCTAGGAATCATT; Sa1-5 , CGTCGATGGCAATCTTTTCTGGAA; and Sa1-3 , ATTGGCATCGTAGTTGAGAAGCTT [46] . The mat-a1 and mat-A1 fragments were distinguished by agarose gel electrophoresis . Genomic DNA was prepared for amplification of AFLP loci with the Invitrogen AFLP Core Reagent Kit . Selective “E” primers were 5′-modified with either 6-FAM or HEX fluorescent dye , and products of selective amplification were electrophoretically separated on an ABI 3100 genetic analyzer , and data were collected and analyzed with the ABI software GeneScan and Genotyper ( Applied Biosystems , Inc . , Carlsbad , CA ) . Microsatellite loci in targeted genomic regions ( e . g . , chromosome ends and QTL regions ) were selected from a published list of SSR in N . crassa [47] , and primers were designed with Primer3 on the web [48] . Forward primers were 5′-fluorescently labeled with NED , 6-FAM , or HEX dyes , and size data were collected as for AFLP markers , above . Linkage analysis was performed with MAPMAKER/EXP 3 . 0 [49] . Loci were sorted into sets of linked markers iteratively at a linkage threshold of 3 . 0 LOD and 50 cM ( Kosambi ) and then 6 . 0 LOD and 30 cM using the “group” command . For each linkage group , the “order” command was used to identify the best marker order with a window of 7 markers and a log-likelihood exclusion threshold of 2 . 0 , followed by attempted placement of excluded markers at log-likelihood 1 . 0 . Marker order was confirmed using the “ripple” command to permute five neighboring loci at a time and flag any alternative orders within a log-likelihood of 2 . 0 of the best order . Markers that could not be placed confidently according to these criteria were discarded . Linkage groups were assigned to the seven chromosomes of N . crassa based on inclusion of multiple microsatellite markers targeted to that chromosome . Mendelian segregation of markers was checked in R/qtl [50] with the “geno . table” command . Linkage disequilibrium in pairs of physically unlinked markers was tested in Genepop 3 . 4 [51] , using option 2 , which uses a Fisher exact test implementing a Markov chain to estimate an unbiased P-value . Experiment wide significance threshold ( Type I error α = 0 . 05 ) was determined by Bonferroni correction for the 21 non-identical linkage-group pairs . We investigated the genetics of hybrid fruitbody development in four kinds of N . crassa×N . intermedia matings ( Table 4 ) . Crosses were performed on synthetic cross media in 16×100 mm tubes . At 10 days post-fertilization , fruitbody development scores ( FDS ) were recorded following a four-category scale: 0 , no fruitbody development; 1 , early abortion ( small fruitbodies without apical pores ( ostioles ) ) ; 2 , late abortion ( larger fruitbodies with ostioles , but lacking “beaks” ) ; and 3 , fully developed ( large flask-shaped fruitbodies with ostioles and “beaks” ) . Some crosses resulted in a mixture of two consecutive fruitbody development stages and were recorded as half steps in the scale ( e . g . , crosses with early- and late-aborted fruitbodies were recorded as 1 . 5 ) . Quantitative trait loci ( QTLs ) were identified by composite interval mapping ( CIM ) using Windows QTL Cartographer , v2 . 5 [52] . CIM was performed under model 6 with 5 control markers and a window size of 20 cM using a 1 cM walk speed . At each step the likelihood ratio statistic ( LR ) testing the hypothesis that a QTL exists versus the null hypothesis that no QTL exists was determined . For each trait , a critical LR threshold reflecting a Type I error of 0 . 05 was estimated by permuting the data 1000 times . Significant QTLs were CIM maxima whose LR exceeded the critical threshold and whose 95% confidence intervals were discontinuous with those of other CIM maxima . Ninety-five percent support intervals were estimated as the area bounded by 1-LOD drops in the LR where LOD = log10 ( LR/2 ln 10 ) . The null hypothesis of neutral trait evolution for sympatric hybrid fruitbody abortion ( the reinforcement trait ) was tested by subjecting the genetic effects data to a QTL sign test [32] , as implemented by the QTLsigntest software provided by H . A . Orr . Because the QTL sign test assumes an additive genetic model , we first scanned for epistatic loci using “scantwo” of R/qtl using the expectation-maximization , interval mapping algorithm and multipoint genotype probabilities calculated using the “calc . genoprob” command with a step size of 2 . 5 and an error probability of 0 . 01 . For each chromosome position the likelihood ratio statistic comparing the full epistatic model to the two-locus additive model was determined For each trait , a critical likelihood ratio statistic threshold reflecting a Type I error of 0 . 05 was estimated by permuting the data set 1000 times . QTLsigntest determines how likely the proportion of loci with positive vs . negative additive effects is under a neutral model of complex trait evolution , when conditioned on the magnitude of the trait difference in the parent strains , the number of detected QTLs , the threshold of detection , and distribution of the absolute value of additive effects , which are all empirically determined . QTLsigntest was parameterized as follows: Parental RSS difference = 2; number of QTLs = 11; detection threshold = 0 . 25; effects distribution gamma ( shape = 13 . 5 , scale = 0 . 034 ) .
|
Although Darwin believed that natural selection could not drive intersterility between species , it is now well established that there is a role for natural selection in the evolution of premating discrimination that reinforces barriers to hybridization . However , natural selection for postmating barriers , like hybrid inviability , is still controversial , because it can only occur when overall maternal fitness is increased by the inviability of hybrid offspring . Constraint on adaptive evolution of postmating barriers poses a problem when organisms without premating preferences must adapt to the presence of related species and ensure that reproduction occurs only between members of the same species . We studied the evolutionary genetics of a reinforced , postmating barrier between two species of mold , Neurospora crassa and N . intermedia . Although hybrids have low fitness , Neurospora females do not discriminate against different-species sex partners before mating . Instead , N . crassa has adapted to the presence of the N . intermedia in its range by selectively aborting hybrid fruitbodies . We show that abortion increases maternal fitness because N . crassa can mate again after hybridization only if fruitbodies abort . Abortion is controlled by 11 loci , whose genetic effects are consistent with an adaptive evolution model , confirming that abortion evolved via natural selection against hybridization .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"mycology",
"fungi",
"adaptation",
"natural",
"selection",
"speciation",
"biology",
"evolutionary",
"biology",
"hybridization",
"microbiology",
"evolutionary",
"processes"
] |
2011
|
Genetic Architecture of a Reinforced, Postmating, Reproductive Isolation Barrier between Neurospora Species Indicates Evolution via Natural Selection
|
The Saccharomyces cerevisiae Dun1 protein kinase is a downstream target of the conserved Mec1-Rad53 checkpoint pathway . Dun1 regulates dNTP pools during an unperturbed cell cycle and after DNA damage by modulating the activity of ribonucleotide reductase ( RNR ) by multiple mechanisms , including phosphorylation of RNR inhibitors Sml1 and Dif1 . Dun1 also activates DNA-damage-inducible genes by inhibiting the Crt1 transcriptional repressor . Among the genes repressed by Crt1 are three out of four RNR genes: RNR2 , RNR3 , and RNR4 . The fourth RNR gene , RNR1 , is also DNA damage-inducible , but is not controlled by Crt1 . It has been shown that the deletion of DUN1 is synthetic lethal with the deletion of IXR1 , encoding an HMG-box-containing DNA binding protein , but the reason for this lethality is not known . Here we demonstrate that the dun1 ixr1 synthetic lethality is caused by an inadequate RNR activity . The deletion of IXR1 results in decreased dNTP levels due to a reduced RNR1 expression . The ixr1 single mutants compensate for the reduced Rnr1 levels by the Mec1-Rad53-Dun1-Crt1–dependent elevation of Rnr3 and Rnr4 levels and downregulation of Sml1 levels , explaining why DUN1 is indispensible in ixr1 mutants . The dun1 ixr1 synthetic lethality is rescued by an artificial elevation of the dNTP pools . We show that Ixr1 is phosphorylated at several residues and that Ser366 , a residue important for the interaction of HMG boxes with DNA , is required for Ixr1 phosphorylation . Ixr1 interacts with DNA at multiple loci , including the RNR1 promoter . Ixr1 levels are decreased in Rad53-deficient cells , which are known to have excessive histone levels . A reduction of the histone gene dosage in the rad53 mutant restores Ixr1 levels . Our results demonstrate that Ixr1 , but not Dun1 , is required for the proper RNR1 expression both during an unperturbed cell cycle and after DNA damage .
Cells experiencing DNA damage or replication blocks activate stress response pathways , or checkpoints , that arrest the cell cycle and facilitate DNA repair . In budding yeast , the key checkpoint protein kinases are Mec1 ( homolog of human ATR ) and Rad53 ( homolog of CHK2 and functional homolog of CHK1 in human ) , reviewed in [1] , [2] . In human cells , ATR and CHK2 are upstream regulators of p53 and are inactivated in many cancers . In Saccharomyces cerevisiae , Mec1 and Rad53 are essential and control phosphorylation and activation of the checkpoint kinase Dun1 [3] ( Figure 1 ) . Dun1 shares homology with Rad53 and Chk2 and is required for the DNA damage response . The essential function of the Mec1-Rad53-Dun1 pathway is to maintain an adequate supply of dNTPs by regulating the activity of ribonucleotide reductase ( RNR ) during the normal cell cycle [4]–[6] . RNR catalyzes the rate-limiting step in the biosynthesis of all four dNTPs and maintains both their balance and appropriate concentration . Full activation of RNR in S . cerevisiae by the Mec1-Rad53-Dun1 checkpoint in response to DNA damage results in a 6- to 8-fold increase in dNTP concentration [7] . Such increases in dNTP concentration during DNA damage correlate with DNA damage tolerance . Four genes encode yeast RNR: RNR1 and RNR3 encode the large subunit [8] , [9] , and RNR2 and RNR4 encode the small subunit [10]–[13] . The three key targets of the Mec1-Rad53-Dun1 pathway are Sml1 , a protein inhibitor of RNR; Crt1 ( Rfx1 ) , a transcription factor; and Dif1 , a protein that regulates the nuclear retention of Rnr2 and Rnr4 ( Figure 1 ) . Phosphorylation of Sml1 during S phase or after DNA damage by Dun1 targets Sml1 for proteolysis , which relieves the inhibition of RNR activity [14] . Phosphorylation of Dif1 releases Rnr2 and Rnr4 into the cytoplasm , where they combine with Rnr1 to form an active RNR complex [15] , [16] . Crt1 blocks transcription at target promoters through recruitment of the general repressors Tup1 and Ssn6 [17] . Phosphorylation of Crt1 in a Mec1-Rad53-Dun1–dependent manner after DNA damage or replication stress promotes its dissociation from target promoters and activation of transcription . Crt1 represses RNR2 , RNR3 , and RNR4 [12] , [17] . RNR3 is not essential and is normally expressed at very low levels , but is highly induced by DNA damage and has been used in genetic screens for the identification of both DUN1 and CRT1 [3] , [17] . Interestingly , the fourth RNR gene , RNR1 , is also DNA damage inducible but does not contain the Crt1-binding sites in its promoter and consequently is not repressed by Crt1 . The mechanism of RNR1 activation by DNA damage remains unknown [17] , [18] . The lethality of mec1 and rad53 mutants can be rescued either by deletion of SML1 ( suppressor of mec1 lethality ) [5] , CRT1 [17] , DIF1 [15] , [16] , or by overexpression of RNR1 or RNR3 [6] , all resulting in increased RNR activity . In contrast , the deletion of DUN1 is not lethal and does not cause any obvious proliferation defects except for a slightly prolonged S phase , defects in mitochondrial propagation and decreased dNTP levels [3] , [14] , [19] . It is therefore possible that another pathway exists downstream of Mec1 and Rad53 , functioning in parallel with Dun1 . Alternatively , the elevation of dNTP levels by the Mec1-Rad53-Dun1 pathway is essential only in mec1 or rad53 mutants , because Mec1 and Rad53 , in contrast to Dun1 , are involved in a plethora of important chromosomal transactions , reviewed in [1] . Others have performed large-scale analyses of synthetic genetic interactions , in which the DUN1 gene was one of the baits . In two of such screens , DUN1 mutants were found to be synthetic lethal with a gene encoding the intrastrand cross-link recognition protein ( Ixr1 ) , but the reason for this synthetic interaction remains unknown [20] , [21] . Ixr1 is a high mobility group ( HMG ) transcription factor first identified by its ability to bind DNA modified by the anticancer drug cisplatin ( cis-diamminedichloriplatinum ( II ) ) [22] . Very little is known about the cellular function of Ixr1 . In addition to the two HMG boxes , Ixr1 has several polyglutamine regions , important for protein–protein interactions . Its closest homolog in yeast is Abf2 ( TFAM in human ) , a mitochondrial DNA-binding protein important for replication and transcription [23] . Earlier , Ixr1 was implicated in aerobic transcriptional repression of COX5b , which encodes a subunit of mitochondrial cytochrome c oxidase [24] . Here we demonstrate that Ixr1 is required for the maintenance of the Rnr1 levels . In the absence or Ixr1 , Rnr1 levels are decreased and became even lower after DNA damage , instead of increasing as in the wild-type . This observation explains the sensitivity of ixr1 mutants to hydroxyurea ( HU ) , an inhibitor of RNR . In contrast , the levels of Rnr3 and Rnr4 in ixr1 mutants are increased due to the activation of the Mec1-Rad53-Dun1-Crt1 pathway , and increase even further after DNA damage and replication blocks , similar to wild-type . We show that deletion of SML1 or overexpression of RNR1 or RNR3 elevates dNTP pools and rescues the ixr1 dun1 synthetic lethality . The requirement for RNR activation in ixr1 via Dun1 explains why Dun1 is indispensible in ixr1 mutants .
Earlier screens for synthetic genetic interactions between dun1Δ and other genes used a collection of yeast strains with null alleles in all nonessential genes . To facilitate identification of synthetic genetic interactions of DUN1 with essential genes we performed a colour-based synthetic lethal screen using a dun1Δ strain as described in the Materials and Methods . Briefly , the ade2 ade3 yeast strains are white unless a plasmid with ADE3 is present , conferring red color . Three mutations resulting in synthetic lethality with dun1Δ ( first designated as mut1 , mut2 , and mut3 ) were isolated based on the inability of the ade2 ade3 dun1Δ mut strains to lose the pK503 plasmid carrying DUN1 and ADE3 . MUT1 was identified as RAD53 , MUT2 as WHI3 , and MUT3 as IXR1 . Sequencing of the rad53 mutant allele identified a single point mutation changing His 622 to Tyr . The H622 residue of Rad53 was identified before as crucial for the interaction of Rad53 with Rad9 [25] , but the synthetic lethality of rad53-H622Y was not known . Sequencing of whi3 identified a single point mutation changing Gln in position 481 to a stop codon . Consistent with this observation , deletion of WHI3 showed a synthetic growth defect with dun1Δ in a large-scale analysis [20] . Finally , sequencing of the ixr1 mutant allele identified a single point mutation that changed Ser 366 to Phe in the first of the protein's two HMG boxes ( Figure 2A ) . This highly conserved serine residue ( Figure 2B ) forms water-mediated hydrogen bonds to DNA bases and interacts with DNA [26] . In this study , we concentrated our efforts on the genetic interaction of DUN1 with IXR1 . A well-established role of Dun1 is to increase RNR activity by targeting Sml1 and Dif1 for degradation and by transcriptional activation of RNR2 , RNR3 and RNR4 . Therefore , we asked whether deletion of SML1 or elevated expression of RNR genes rescues the lethality of dun1Δ ixr1 . The dun1Δ ixr1-S366F [pK503] strain , originally identified in the screen , was crossed with a dun1Δ sml1Δ strain . Both strains are ade2 ade3 mutants . The unstable pK503 plasmid was lost in dun1Δ sml1Δ ixr1-S366F colonies , based on their white color , but not in dun1Δ ixr1-S366F colonies ( Figure 2C ) . Next , we crossed ixr1Δ sml1Δ with dun1Δ . Tetrad analysis confirmed that dun1Δ ixr1Δ sml1Δ spores were viable while dun1Δ ixr1Δ were unable to germinate ( Figure 2D ) . An additional copy of RNR1 rescues the dun1 ixr1 synthetic lethality , as demonstrated by sporulation and tetrad analysis of the ixr1/IXR1 dun1/DUN1 diploid strain transformed with a centromeric plasmid pBJ6 containing RNR1 under the control of the native promoter ( Figure 2E ) . Finally , after transformation with plasmids overexpressing RNR1 ( Figure 2F ) or RNR3 ( Figure 2H ) or with the centromeric pBJ6 plasmid containing an additional copy of RNR1 ( Figure 2G ) , we also observed the loss of the pK503 plasmid from the dun1Δ ixr1-S366F strain , leading to the sectoring phenotype . The requirement of Dun1 for ixr1 viability suggests that the Mec1-Rad53-Dun1 pathway is activated in ixr1 strains . A highly sensitive readout of this pathway's activation is induction of Crt1-controlled RNR2 , RNR3 , and RNR4 . Indeed , Rnr3 and Rnr4 levels were higher in ixr1-S366F and ixr1Δ strains compared to wild-type ( Figure 3A , lanes 1–3 ) . Rnr2 levels were not significantly changed in ixr1Δ compared to wild-type ( Figure 3B ) . Activation of the Mec1-Rad53-Dun1 pathway in ixr1 strains was not maximal; exposure to DNA damaging agents further increased Rnr2 , Rnr3 and Rnr4 levels ( Figure 3A and 3B ) . We were able to detect an increase in Rnr3-HA levels in the undamaged ixr1Δ strain only with the more sensitive anti-HA antibodies , but not with the polyclonal anti-Rnr3 antibodies , also indicating that the Mec1-Rad53-Dun1 pathway activation is low ( Figure S1A ) . Importantly , the elevation of Rnr2 , Rnr3 and Rnr4 in response to DNA damage was identical in the ixr1 and wild-type strains , indicating that the Mec1-Rad53-Dun1 pathway was not compromised in ixr1 mutants . Sml1 levels were decreased in ixr1 , again suggesting that the Mec1-Rad53-Dun1 pathway was activated ( Figure 3C ) . We did not observe a mobility shift of the Rad53 band in ixr1Δ , indicating low activation of the Mec1-Rad53-Dun1 pathway ( Figure 3D ) . The full activation of the checkpoint by DNA damage resulted in hyperphosphorylation of Rad53 both in ixr1Δ and wild-type , leading to a shift of the Rad53 band ( Figure 3D ) . Nevertheless , the increased Rnr3 and Rnr4 levels in ixr1Δ are Rad53 dependent . Deletion of RAD53 , but not of DUN1 , abolished upregulation of Rnr3 and Rnr4 in the ixr1Δ strain ( Figure 3E ) . We interpret this result to mean that Rad53 can take over the Dun1 function and activate Rnr3 and Rnr4 expression in ixr1Δ when DUN1 is deleted , but not vice versa because Dun1 functions downstream of Rad53 . A DUN1-independent pathway for RNR transcriptional induction was observed earlier , and it was suggested that Rad53 can directly recognize Dun1 substrates [12] . Alternatively , Rad53 , and not Dun1 , could be the main activator of Rnr3 and Rnr4 expression in the ixr1Δ strain in the absence of DNA damage . Because the levels of several RNR proteins are increased ( Rnr3 , Rnr4 ) or unchanged ( Rnr2 ) in ixr1Δ , while Sml1 levels are decreased , the dNTP levels might have been elevated in the ixr1 strains; however , dNTP levels were lower in ixr1Δ compared to wild-type ( Figure 4A ) . Similarly , dNTP levels in ixr1Δ sml1Δ were lower than in sml1Δ alone ( Figure 4A ) . DNA damage induction by 4-nitroquinoline 1-oxide ( 4-NQO ) resulted in an increase in dNTP concentration in ixr1Δ , but to a lower level compared to wild-type ( Figure 4B ) . The ixr1Δ mutant exhibited an increased frequency of petite formation ( Figure 4C ) and sensitivity to HU ( Figure 4D ) , two phenotypes associated with decreased dNTP production [5] . Interestingly , although the dNTP levels were higher in ixr1Δ sml1Δ than in wild type , Rnr3 and Rnr4 in ixr1Δ sml1Δ were still elevated to the same levels as in the ixr1Δ single mutant ( Figure 3E ) . This observation indicates that activation of the Mec1-Rad53-Dun1 pathway in ixr1Δ is not only due to a decreased dNTP production , but also due to some other defects . The paradoxical finding that dNTP pools decreased despite the activation of the Mec1-Rad53-Dun1 pathway in ixr1 strains indicates a deficiency in another component ( s ) of the RNR machinery . Analysis of Rnr1 steady state levels demonstrated moderately decreased levels in ixr1Δ and in ixr1-S366F compared to wild-type ( ∼64% , and ∼62% respectively ) ( Figure 5A , 5B , and 5C ( 0 min lanes ) ) . As expected , incubation of wild-type cells in the presence of 4-NQO or HU for two hours led to an increase in Rnr1 levels ( ∼39% and 57% , respectively ) ( Figure 5A and 5B ) . Interestingly , the same treatment of ixr1-S366F and ixr1Δ led to a further reduction of Rnr1 to ∼19% and ∼37% , respectively , after 4-NQO treatment , and to ∼56% and ∼46% , respectively , after the HU treatment ( Figure 5A and 5B ) . This reduction was continuous; Rnr1 levels remained lower in ixr1Δ throughout a 12-hour incubation with 4-NQO ( Figure 5C ) . Based on flow-cytometric analysis , the ixr1Δ strain had a slightly greater proportion of cells in S phase compared to wild-type both before and during 4-NQO treatment ( Figure 5C ) , although the overall proliferation rate was similar between ixr1Δ and wild-type ( Figure 5D ) . The decreased Rnr1 levels caused by ixr1Δ provide an explanation for the synthetic lethality displayed by the ixr1Δ dun1Δ double mutant strain: the dun1Δ strains are defective in relieving the inhibition of RNR imposed by Sml1 , Dif1 , and Crt1 . Using a β-galactosidase assay we demonstrated that the decreased Rnr1 levels in ixr1Δ are due to a lower RNR1 promoter activity , which indicates that Ixr1 directly or indirectly regulates RNR1 transcription ( Figure 6A ) . Deletion of IXR1 also caused a concomitant increase in RNR3 and RNR4 promoter activities ( Figure 6A ) , in agreement with the observed increase in Rnr3 and Rnr4 protein levels ( Figure 3A ) . To gain further insight into the mechanism of RNR1 regulation by Ixr1 , we performed chromatin immunoprecipitation ( ChIP ) experiments followed by qPCR using Ixr1-9xMyc fusion protein and 9E10 antiserum . We analyzed binding of Ixr1 to the RNR1 promoter ( pRNR1 ) region . In addition , we analyzed the DSF2 promoter ( pDSF2 ) region earlier identified as an Ixr1-interacting locus [27] and the actin ( ACT1 ) open reading frame , a commonly used negative control . As another control , we used an untagged congenic IXR1 strain . Ixr1 interacted with all three loci ( Figure 6B ) : relative to the input DNA we recovered 0 . 46% , 0 . 35% and 0 . 79% of pRNR1 , pDSF2 and ACT1 loci , respectively , in the IXR1-9Myc strain . The interaction of Ixr1 with the RNR1 promoter did not change after the treatment of cells with 4-NQO ( 0 . 46% and 0 . 48% , respectively ) . In the untagged strain , DNA recovery was at background levels as judged by the ChIP samples where 9E10 antiserum was omitted . The precipitation of the ACT1 ORF locus indicates that Ixr1 binds to many loci in the genome . The DNA-damage-inducible genes become damage uninducible in dun1 mutants , because Dun1 is required to relieve the inhibition imposed by the transcriptional inhibitor Crt1 . Indeed , induction of Rnr4 in response to 4-NQO is less pronounced in the dun1Δ sml1Δ strain compared to wt ( Figure 7A ) . The RNR1 promoter , however , does not contain Crt1 sites and the expression of RNR1 is not affected by the CRT1 deletion [17] , [18] . In Figure 7A , we demonstrate that the elevation of Rnr1 levels in response to DNA damage does not depend on DUN1 , but does depend on Rad53 and Mec1 . All checkpoint mutant strains in this experiment contained sml1Δ , because mec1Δ and rad53Δ are inviable otherwise . As a control , we demonstrate that the deletion of SML1 by itself has little effect on Rnr1 and Rnr4 levels in the wild type and ixr1Δ strains treated by 4-NQO ( Figure S1C ) . Because Mec1 , Rad53 and Dun1 are protein kinases , it was possible that they directly phosphorylate Ixr1 and modulate its function . We analyzed the phosphorylation status of Ixr1 and Ixr1-S366F proteins and found that Ixr1 is a phosphoprotein most likely phosphorylated at several residues ( Figure 7B ) . Serine 366 is important for Ixr1 phosphorylation , as Ixr1-S366F separated by SDS-PAGE as several bands with higher mobility compared to Ixr1 . Treatment of Ixr1 with λ-phosphatase increased the mobility of Ixr1 to that of Ixr1-S366F . However , we did not observe changes in Ixr1 mobility in mec1 , rad53 , or dun1 mutants ( Figure 7C ) . Interestingly , Ixr1 levels were significantly lower in rad53Δ , but not in mec1Δ or dun1Δ , compared to wt ( Figure 7C ) . Rad53 is known to regulate histone levels , and rad53 strains have increased amount of histones [28] , [29] . Lowering histone dosage in the rad53 mutant strains by deleting copies of histone 3 and histone 4 genes ( HHT2 and HHF2 ) restored Ixr1 to wild-type levels ( Figure 7D ) .
The Dun1 kinase is a downstream target of the Mec1-Rad53 checkpoint , which monitors the genome integrity . In S . cerevisiae , the Mec1-Rad53-Dun1 pathway also regulates the RNR activity both during the normal cell cycle and after DNA damage [4] . RNRs are instrumental in controlling dNTP balance and concentration [30] . Deletion of DUN1 is synthetic lethal with the deletion of many genes involved in DNA replication and DNA repair [20] , [21] . Synthetic lethality of dun1 with a number of other genes remains unexplained . Here , we demonstrate that IXR1 , deletion of which is synthetic lethal with dun1 , is required for the normal expression of the RNR1 gene and maintenance of the dNTP pools . In the absence of DNA damage , the deletion of IXR1 leads to a moderate decrease of Rnr1 and dNTP levels . This decrease is partially compensated by the activation of the Mec1-Rad53-Dun1 pathway . We base this conclusion on the following observations . First , Rnr3 and Rnr4 , whose levels are controlled by the Mec1-Rad53-Dun1 pathway , are upregulated in ixr1 , and Rad53 is required for this upregulation . Second , RNR inhibitor Sml1 , whose levels are also controlled by the Mec1-Rad53-Dun1 pathway , is downregulated in ixr1 . Third , DUN1 is indispensable in ixr1 , but ixr1 dun1Δ synthetic lethality is rescued by elevated dNTP levels . We note that elevation of dNTP levels in ixr1Δ caused by the SML1 deletion does not eliminate the checkpoint activation ( Figure 3E ) , indicating that deletion of IXR1 leads to replication stress not only because of the decreased dNTP levels expression but also because of other defects . It is conceivable that in addition to RNR1 some other genes involved in the DNA biosynthesis are regulated by Ixr1 . The reported synthetic lethality of ixr1Δ with the origin recognition complex mutant orc2-1 [31] and synthetic sickness with the thymidylate kinase mutant cdc8-2 [32] indicate the importance of Ixr1 for the processes involved in DNA replication . RNR expression increases in response to DNA damage in most organisms . In E . coli , nrdA and nrdB ( encoding the large and the small RNR subunits , respectively ) are among the most highly induced genes following UV exposure ( induced ∼20- and ∼7-fold , respectively ) [33] , [34] . In mammalian cells , DNA damage induces the p53R2 protein , an alternative small RNR subunit , about 4-fold in a p53-dependent manner [35]–[37] . Similarly , the Drosophila large RNR subunit , RnrL , is induced by ionizing radiation in wild-type , but not p53-deficient strains [38] . In the yeast Schizosaccharomyces pombe , RNR genes are among the most robustly induced genes following DNA damage [39] . All four S . cerevisiae RNR genes are activated by DNA damage and replication blocks [8] , [10] , [12] . The pathway involved in the activation of RNR2 , RNR3 and RNR4 is well understood and requires the Mec1-Rad53-Dun1 kinase cascade , which targets Crt1 , transcriptional inhibitor of DNA-damage-inducible genes ( Figure 1A ) . Here we demonstrate that elevation of Rnr1 in response to DNA damage requires Mec1 and Rad53 , but not Dun1 ( Figure 7A ) . Earlier , it has been shown that RNR1 expression does not depend on Crt1 [17] , [18] . Thus , the downstream Dun1-Crt1 part of the Mec1-Rad53-Dun1-Crt1 pathway , which is known to control the DNA-damage-inducible genes in yeast , is not involved in the regulation of Rnr1 . Instead , the elevation of Rnr1 levels in response to DNA damage requires Ixr1 ( Figure 7E ) . In addition to Crt1 , transcription of RNR2 , RNR3 and RNR4 genes is also controlled by Rox1 and Mot3 , the DNA binding proteins that repress the hypoxic genes by recruiting the Ssn6/Tup1 general repression complex . Again , in contrast to RNR2 , RNR3 , and RNR4 genes , no Rox1 or Mot3 sites are present in the RNR1 promoter [18] . Transcription of RNR1 is controlled by MBF , a dimeric transcription factor composed of Swi6 and Mbp1 [40]–[42] . Interestingly , Swi6 is directly phosphorylated by Rad53 in response to DNA damage [43] . It will be interesting to investigate whether Ixr1 is important for MBF-dependent regulation of the RNR1 promoter . Earlier , Ixr1 was implicated in controlling the levels of the hypoxic gene COX5b [24] . Currently , we do not know whether Ixr1 is involved in the activation of RNR1 expression in response to oxygen deprivation . In contrast to many other HMG-box proteins , Ixr1 is rather large ( 68 kDa ) and contains several polyglutamine repeats , which are often involved in protein-protein interactions and are present in many transcription factors . The HMG box is a conserved domain of ∼80 amino acids , binding to the minor groove of DNA . Proteins containing two or more HMG boxes usually recognize structural features of DNA without sequence specificity , while proteins containing one HMG box can recognize DNA in a sequence specific manner . In S . cerevisiae , there are two proteins containing two HMG boxes ( Ixr1 and Abf2 ) and five proteins containing one HMG box ( Nhp6A , Nhp6B , Nhp10 , Hmo1 and Rox1 ) . The closest homolog of Ixr1 , Abf2 , binds to many loci in the mitochondrial genome [44] . Yet , the HMG-box proteins with two or more HMG boxes can bind to specific loci in the genome . For example , human transcription factor UBF , which has 6 HMG boxes and belongs to the sequence-nonspecific class of HMG-box proteins , binds specifically to rDNA or to heterologous UBF-binding sequences from Xenopus integrated into ectopic sites on human chromosomes [45] . Our ChIP analysis of Ixr1 identified the RNR1 promoter as a binding locus . However , Ixr1 bound equally well at two other tested loci , the DSF2 promoter and the ACT1 open reading frame . Still , it is possible that , in the context of the RNR1 promoter , Ixr1 together with other proteins directly regulates RNR1 gene expression . Interestingly , the mutation in the Ixr1 S366 residue that is important for interaction of HMG boxes with DNA results in the same phenotype as the deletion of IXR1 gene . We show that this serine residue is required for the phosphorylation of several amino acid residues in Ixr1 . Currently we do not know the phosphorylation status of S366 . It is possible that S366 itself is not phosphorylated , but its interaction with the DNA or other proteins is required for the phosphorylation of other residues in Ixr1 . Multiple phosphorylation of Ixr1 causes an increase in the apparent molecular weight of the protein: Ixr1 separates by SDS-PAGE as a ∼85 kDa protein ( not as predicted 68 kDa ) . We demonstrate that neither Mec1 , nor Rad53 , nor Dun1 are responsible for the phosphorylation of Ixr1 , as its mobility is not affected in the respective mutants . The region of the HMG domain around the Ser366 residue has been shown to affect DNA binding specificity . All sequence-specific HMG proteins have an asparagine at this position , whereas all non-sequence-specific HMG proteins have a serine ( e . g . , Ser10 in the D . melanogaster HMG-D box co-crystallized with DNA ) [26] . To our knowledge , crystal structures analyzing the interaction of HMG boxes and DNA were solved with the non-phosphorylated proteins . It would be interesting to investigate whether S366 is phosphorylated in Ixr1 , whether the corresponding serine residues are phosphorylated in other HMG proteins in other species , and whether Ser366 phosphorylation affects DNA binding and/or makes binding of the HMG box sequence specific . Although the mobility of Ixr1 is not changed , its levels are significantly reduced in the rad53Δ strain ( Figure 7C ) . rad53 mutant strains are known to have increased histone levels due to a defect in histone degradation [28] . Increased histone levels presumably lead to decreased Ixr1 levels , because we show that decreasing histone dosage in the rad53 strain restores Ixr1 levels ( Figure 7D ) . There are at least two possibilities explaining this interplay between Ixr1 and histone levels . Because Ixr1 contains two HMG boxes and therefore binds DNA presumably without sequence specificity , it might compete with histones for DNA binding . Increased histones in rad53Δ might displace Ixr1 from the IXR1 promoter , where it was shown to bind and regulate its own expression [46] . Alternative , but not exclusive possibility is that Ixr1 displaced by histones from DNA undergoes degradation . In summary , we identify Ixr1 as a novel factor involved in regulation of dNTP pools and RNR1 , a gene that , in contrast to all other known DNA-damage inducible genes , is not controlled by Dun1 and Crt1 .
All yeast strains used in this study are congenic to W1588-4C [5] . Table 1 gives only the allele ( s ) that differ from the W1588-4C genotype . Table S1 lists primers used for strain construction . DUN1 was deleted using the KanMX4 cassette PCR-amplified with primers F_Dun1 and R_Dun1 from the dun1Δ::KanMX4 Y03798 strain ( Euroscarf ) . The CY1263 ade3::HISG strain [47] was crossed with W1588-4C to select ade2 ade3 clones ( TOY502 ) . The resulting strain was crossed with dun1Δ::KanMX4 to create the strain used for the synthetic lethality screen ( TOY527 ) . The TOY836 ( IXR1-9MYC ) strain was generated by amplifying and introducing the 9MYC-TRP1 cassette from the Z1580 strain [48] into W1588-4C . To overexpress Rnr1 or Rnr3 , the previously described pESC-pGAL1-RNR1 or pBAD79 plasmids were used [6] , [49] . To express RNR1 under its own promoter from a low-copy centromeric vector , the pBJ6 ( pRS316-RNR1 ) plasmid was used ( gift of Anders Byström , Umeå University ) . To construct pK503 ( ura3Δ::LEU2 , ADE3 , DUN1 ) , the DUN1 gene including the promoter region was PCR-amplified using primers Dun1_F and Dun1_rev . The PCR product was cloned into the SalI site of p2013 [50] , and the URA3 gene in the resulting plasmid was then replaced by LEU2 . To construct pK521 ( TRP1 , DUN1 ) a SalI/SalI fragment of DUN1 from pK503 was cloned into the SalI site of pRS414 [51] . To construct plasmids for the β-Galactosidase assay , the RNR1 , RNR3 and RNR4 promoters were PCR amplified from the W1588-4C genomic DNA using primers pR1-F , pR1-R , pR3-F , pR3-R , pR4-F and pR4-R . The RNR3 promoter was cloned in the BamHI site of pJO20 , and the RNR1 and RNR4 promoters were cloned in the BamHI site of pJO21 [52] , resulting in plasmids pK505 , pK504 and pK506 , respectively . β-Galactosidase levels were assayed as described [52] . To identify mutations synthetic lethal with dun1Δ , we used a color-based synthetic lethal screen [53] . The TOY541 strain carrying pK503 was grown in selective medium to ∼2×107 cells/ml . Cells were spun down , resuspended in water , plated onto YPD plates at 1500 cells/plate , and UV-mutagenized with a dose of 150 J/m2 , resulting in 30% survival . Plates were placed in the dark and incubated for 3 days at 30°C . Non-sectoring red colonies were re-streaked twice on YPD , and those retaining the red color were selected for further analysis . Candidate mutants were transformed with pK521 to exclude mutants synthetically lethal with the plasmid-borne ADE3 or LEU2 genes . Transformants were grown on –Trp medium , and strains with a sectoring phenotype were selected . The candidate mutants were crossed with TOY566 to test recessiveness/dominance , and tetrad analysis was performed to select mutations with monogenic inheritance . Selected strains were mated in all possible combinations to establish complementation groups . In total , we isolated 4 mutants falling into three complementation groups and identified them as one ixr1 , one rad53 and two whi3 mutants as outlined below . One strain from each complementation group was transformed with a pRS314-based yeast genomic DNA library [54] , and transformants were selected on –Trp plates . Clones that showed a sectoring phenotype were re-streaked onto –Trp plates , and plasmids were isolated from these clones and partially sequenced using T3 and T7 standard primers . The obtained sequences were subjected to BLAST homology searches using the S . cerevisiae genome database , and genomic regions were retrieved . One of the regions contained the IXR1 ORF ( TOY544 ) . A TRP1 cassette was inserted downstream of the IXR1 ORF creating TOY598 , which was crossed with TOY566 to verify the co-segregation of a genomic marker and the non-sectoring phenotype . Then , the genomic region retrieved from the DNA library in pRS314 was shortened , and resulting plasmids were re-transformed in TOY544 . Plasmids lacking the full-length IXR1 ORF failed to recover the sectoring phenotype . IXR1 and ixr1 ORFs were PCR amplified from the genomic DNA of TOY502 and TOY544 , respectively , using primers F_ixr and R_ixr , and sequenced using the same primers . The mutations in RAD53 and WHI3 genes were identified by the same procedure and PCR amplified followed by sequencing using primers F_rad53 , R_rad53 , F_whi3 and R_whi3 . Protein samples for Western blotting were prepared as described [55] . Proteins were separated by SDS-PAGE and transferred to a nitrocellulose membrane ( Protran BA 85 , Whatman , USA ) using the Minigel System ( C . B . S . Scientific Co . , USA ) . Rabbit polyclonal anti-Rnr1 ( AS09 576 ) , anti-Rnr2 ( AS09 575 ) , anti-Rnr3 ( AS09 574 ) , and anti-Sml1 ( AS10 847 ) antibodies were produced by Agrisera , Sweden ( peptides used for immunization are listed in Table S2 ) . For the detection of Ixr1 we used rabbit polyclonal antibodies produced by Agrisera , Sweden ( Table S2 ) . For the detection of the HA-tag , mouse monoclonal 12CA5 antibodies were used ( 1∶5000 ) . For the detection of both Rnr4 and α-tubulin [56] , we used YL1/2 rat monoclonal antibodies ( Sigma ) at 1∶2500 . These antibodies recognize C-termini of α-tubulin and small RNR subunits from different species . The absence of the Rnr4 band on a Western blot with an extract from an rnr4Δ strain ( CUY995 , [11] ) confirmed that YL1/2 antibodies specifically recognize yeast Rnr4 ( Figure S1B ) . For the detection of Rad53 , we used yC-19 goat polyclonal antibodies at 1∶2000 ( Santa Cruz Biotechnology , USA ) . Quantification of protein levels was performed using ImageJ software ( http://rsbweb . nih . gov/ij ) . Protein levels were calculated as relative units ( RU , levels of the particular protein divided by the levels of tubulin in corresponding sample ) . To quantify Rnr1 levels three independent clones were analyzed on the same membrane . Chromatin immunoprecipitation followed by qPCR was performed as previously described ( Barsoum et al . , 2010 ) . DNA damaging agent 4-NQO was added to the cells to final concentration 0 . 25 mg/L at OD ∼0 . 5 and cells were grown 2 hours to OD ∼1 . 2–1 . 5 . To amplify RNR1 promoter , DSF2 promoter and ACT1 open reading frame ChIP_pRNR1 , ChIP_pDSF2 and ChIP_ACT1 primers were used ( Table S1 ) . 9×107 cells were collected , vortexed with glass beads in 10% w/v trichloroacetic acid and spun down 10 min in microcentrifuge in cold room . Pellet was re-suspended in 150 µl of λ-Phosphatase buffer , pH was adjusted to 7 . 5 with basic 1 M Tris and 15 µl of 10× Complete Protease Inhibitor Cocktail ( Roche Applied Biosystems ) was added to the samples . 60 µl of 10× PhosStop Phosphatase Inhibitor Cocktail ( Roche Applied Biosystems ) or 6 µl of λ Phosphatase ( New England Biolabs ) was added to the respective samples and all samples were incubated 1 hour at 30°C . Then , samples were boiled 10 min with Laemmli buffer and analyzed by SDS PAGE followed by the Western blotting . NTP and dNTP extraction and quantification were performed as previously described [7] . Nucleotides were analyzed by HPLC on a Partisphere SAX-5 HPLC column ( 4 . 6 mm×125 mm , Whatman International Ltd . ) using a UV-2075 Plus detector ( Jasco , Tokyo , Japan ) . Mid-log phase cells were collected , sonicated , and plated at appropriate dilutions . For spot assays , 2 µl of 10-fold serial dilutions were spotted onto YPD plates or YPD plates containing 200 mM HU . Cells were grown at 30°C for 3 days . Measurement of the frequency of petite formation was done as described before [5] .
|
Dun1 is a non-essential protein kinase important for the maintenance of genome stability in budding yeast . Earlier studies found that simultaneous deletion of DUN1 and IXR1 results in lethality , but the reason for this so-called synthetic lethality is not clear . Ixr1 is implicated in DNA repair based on its ability to bind to DNA modified by the anticancer drug cisplatin . Here , we investigated the mechanism behind the ixr1 dun1 synthetic lethality . We demonstrate that yeast strains lacking Ixr1 have decreased amounts of dNTPs , the building blocks of DNA . This is because Ixr1 is required for the normal expression of Rnr1 , one of the essential subunits of the enzyme ribonucleotide reductase ( RNR ) , which catalyzes the rate-limiting step in the production of all four dNTPs . Cells lacking Ixr1 compensate the decreased expression of Rnr1 by the increased expression of other RNR genes and degradation of RNR inhibitors . These compensatory processes require Dun1 . Hence , cells lacking both Dun1 and Ixr1 have dNTP pools that are too low for survival . Our work identifies a new important player in the synthesis of the building blocks of DNA .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"cell",
"biology",
"dna",
"replication",
"nucleic",
"acids",
"genetics",
"molecular",
"genetics",
"biology",
"dna",
"dna",
"repair",
"nucleotides",
"molecular",
"biology",
"genetics",
"and",
"genomics",
"gene",
"function"
] |
2011
|
Ixr1 Is Required for the Expression of the Ribonucleotide Reductase Rnr1 and Maintenance of dNTP Pools
|
In Trypanosoma brucei , glycosylphosphatidylinositol phospholipase C ( GPI-PLC ) is a virulence factor that releases variant surface glycoprotein ( VSG ) from dying cells . In live cells , GPI-PLC is localised to the plasma membrane where it is concentrated on the flagellar membrane , so activity or access must be tightly regulated as very little VSG is shed . Little is known about regulation except that acylation within a short internal motif containing three cysteines is necessary for GPI-PLC to access VSG in dying cells . Here , GPI-PLC mutants have been analysed both for subcellular localisation and for the ability to release VSG from dying cells . Two sequence determinants necessary for concentration on the flagellar membrane were identified . First , all three cysteines are required for full concentration on the flagellar membrane . Mutants with two cysteines localise predominantly to the plasma membrane but lose some of their flagellar concentration , while mutants with one cysteine are mainly localised to membranes between the nucleus and flagellar pocket . Second , a proline residue close to the C-terminus , and distant from the acylated cysteines , is necessary for concentration on the flagellar membrane . The localisation of GPI-PLC to the plasma but not flagellar membrane is necessary for access to the VSG in dying cells . Cellular structures necessary for concentration on the flagellar membrane were identified by depletion of components . Disruption of the flagellar pocket collar caused loss of concentration whereas detachment of the flagellum from the cell body after disruption of the flagellar attachment zone did not . Thus , targeting to the flagellar membrane requires: a titratable level of acylation , a motif including a proline , and a functional flagellar pocket . These results provide an insight into how the segregation of flagellar membrane proteins from those present in the flagellar pocket and cell body membranes is achieved .
The external cell surface of the mammalian bloodstream form of African trypanosomes is covered with a densely packed protein coat predominantly composed of a single polypeptide species , the variant surface glycoprotein ( VSG ) [1] , [2] . The VSG is central to the interaction with the host: it functions in both a population survival strategy through antigenic variation and an individual cell survival strategy though rapid endocytosis , removal of bound antibody , and recycling back to the cell surface [3] , [4] , [5] . The VSG coat is essential and loss of VSG synthesis causes a growth arrest in culture and cell death in an animal model [6] . The VSG is attached to the external face of the plasma membrane through a C-terminal glycosylphosphatidylinositol ( GPI ) anchor [7] . Bloodstream form trypanosomes contain a GPI-phospholipase C ( GPI-PLC ) [8] , [9] , [10] and when plasma membrane integrity is compromised , for example by hypotonic lysis , hydrolysis by GPI-PLC releases all VSG from the plasma membrane within five minutes [10] , [11] . However , in a population of cells , the half-life of VSG is ∼30 hours , equivalent to several cell generations [12] , [13] , indicating that there is very tight regulation of GPI-PLC activity and/or access to VSG in live cells The GPI-PLC gene is not essential but acts a virulence factor as a null ( −/− ) mutant was attenuated in mice [14] . The attenuation may be caused by the failure to release VSG from trypanosomes killed by the host immune response . Release of the VSG from dying trypanosomes into the host bloodstream may cause the VSG antibody response to be directed towards novel epitopes on the released VSG and away from expanding populations of cells expressing novel VSGs [15] . However , no definitive function for GPI-PLC has been identified . GPI-PLC behaves as an integral membrane protein [9] , [16]: it has neither an N-terminal signal peptide nor a transmembrane domain [17] but contains a short motif , 268 ACCGACP 274 ( abbreviated to CCGAC ) , which contains three cysteine residues that were shown to be modified by palmitoylation when GPI-PLC was expressed in Xenopus oocytes [18]; furthermore , native GPI-PLC is acylated [19] . In trypanosomes , when acylation was prevented through expression of a GPI-PLC mutant transgene containing the sequence 268 ASRGARP 274 in a trypanosome GPI-PLC −/− background , there was no release of VSG on hypotonic lysis [18] . The subcellular localisation of GPI-PLC has been investigated and two overlapping but distinct results were obtained [20] . Immunofluorescence localised GPI-PLC to a linear array along the flagellum between the paraflagellar rod and flagellar attachment zone ( FAZ ) and evidence was presented for localisation to the external face of the plasma membrane . In contrast , GPI-PLC tagged at the C-terminus with eYFP localised to the plasma membrane , being more concentrated on the flagellar membrane than on the cell body [20] . In trypanosomes , the flagellar membrane is one of three discrete domains of the plasma membrane , the other two being the cell body and the flagellar pocket ( FP ) [21] . The protein complement in these domains is dominated by the VSG , but each domain also contains a set of unique proteins [20] , [22] , [23] . The three domains are demarcated by two structures: firstly by the flagellar pocket collar ( FPC ) , a ring at the neck of the FP that marks the boundary between the cell body and FP membranes; and second by the collarette , which marks the boundary between the flagellar and FP membranes at the point at which the flagellum enters the FP [21] . The FP is the only site of exocytosis and endocytosis [24] and all components of the flagellar and cell body membranes added through vesicular transport pass through the FP . Subsequent sorting in the FP must ensure that components reach their correct destination . The FPC may act as a diffusion barrier to maintain the distinct membrane composition of the flagellum , for example the higher concentration of GPI-PLC [20] . One component of the FPC , BILBO1 , has been characterised: knockdown results in the loss of both the FPC and the FP at the newly synthesised flagellum and subsequent cell death [25] . Acylation is a common but not universal theme in membrane proteins that localise to flagella or cilia [26] , when acylation occurs it is necessary for localisation [27] , [28] , [29] , [30] . However , in some proteins acylation is not sufficient and other amino acid motifs are also required for efficient targeting [28] . Here , the relationship between subcellular localisation and access to the VSG substrate has been investigated through expression of GPI-PLC mutants both to provide information about the regulation of GPI-PLC and also to identify determinants necessary for concentration on the flagellar membrane that may be applicable to a wider range of proteins . When all the cysteines within the CCGAC motif are mutated to serines , GPI-PLC is still enzymically active but is now cytoplasmic and unable to release the VSG coat on hypotonic lysis . The correct pattern of modifications of the CCGAC motif is required for successful trafficking of GPI-PLC to the flagellar membrane: as the number of cysteines is reduced , GPI-PLC becomes localised to the endomembranes between the nucleus and flagellar pocket . However , acylation of the cysteine residues is not sufficient for flagellar localisation: mutation of a single proline results in the failure to concentrate on the flagellar membrane . The concentration on the flagellar membrane requires a functional FPC but does not require flagellar attachment to the cell body .
Hydrolysis of GPI-anchors by GPI-PLC results in the release of diacylglycerol and the formation of inositol 1 , 2 cyclic phosphate on the GPI-glycan . The cyclic phosphate is part of the cross-reacting determinant ( CRD ) , an epitope formed by GPI-anchor hydrolysis by GPI-PLC [31] , [32] , [33] . In vitro , two assays are used to analyse GPI-PLC activity in trypanosomes . First , after detergent lysis , activity is determined by following the appearance of the CRD epitope on the VSG: this assay is independent of any regulation based on spatial segregation . In the second assay , hypotonic lysis is used to rupture the plasma membrane and hydrolysis of the GPI-anchor is detected by separating the reaction into pellet and supernatant fractions; before GPI-hydrolysis membrane-attached VSG is in the pellet fraction and afterward in the supernatant fraction . In this assay , only VSG present on the plasma membrane is hydrolysed; newly synthesised VSG en route to the plasma membrane is not a substrate , which is assumed to be due to lack of access as all VSG is hydrolysed on detergent lysis [34] . The hypotonic lysis assay measures not only the enzymic activity of GPI-PLC , but also its access to plasma membrane VSG . Electron micrographs of hypotonically lysed cells showed cell ghosts with clear holes in the plasma membrane [35] , and the assay is used as the best approximation of immune system-mediated rupture . Using these assays the GPI-PLC −/− cell line is unable to hydrolyse the GPI-anchor [14] . To determine whether the C-terminal eYFP tag affected the activity of GPI-PLC , the relative rate of VSG GPI-anchor hydrolysis after detergent and hypotonic lysis was determined in two cell lines , GPI-PLC/− and GPI-PLC-eYFP/− . The cell lines were made by first deleting one allele [14] to make the GPI-PLC/− cells and then modifying the remaining allele to make GPI-PLC-eYFP/− cell lines [36] . Expression was examined by Western blotting using GPI-PLC antiserum: both proteins had the expected relative molecular mass and similar expression levels ( Figure S1A ) . Both were expressed at lower levels than the parental GPI-PLC +/+ cell line [37] , [38] . The rate of hydrolysis of the VSG GPI-anchor was similar in both detergent and hypotonic lysis assays of both cell lines ( Figure 1 ) , indicating that the eYFP tag did not affect the ability of GPI-PLC to access or hydrolyse its VSG substrate . The membrane topology of GPI-PLC has been the subject of much discussion , with some evidence presented that some or all of the GPI-PLC is on the external face of the plasma membrane despite the absence of an obvious signal sequence [20] . A comparison of the membrane topology of tagged and untagged GPI-PLC was performed by determining the sensitivity to trypsin after addition to live cells expressing both wild type and eYFP tagged forms of the protein ( Figure 2 ) . The cell line for these experiments was made by modifying one of the GPI-PLC alleles with eYFP in wild type cells [36] . Expression was dependent on endogenous transcription and total GPI-PLC protein levels were similar to wild type . Two proteins present on the external face of the plasma membrane were used as controls: VSG and ISG65 [39] . The cytoplasmic controls were the paraflagellar rod proteins PFR1 and PFR2 [40] as well as the cytoplasmic RNA helicase DHH1 [41] . VSG and ISG65 were digested rapidly whereas the cytoplasmic proteins were stable until the endpoint at 18 minutes . Almost complete digestion of ISG65 was unexpected as a fraction resides in the endosomal membrane system [42]; it is possible that either recycling of ISG65 back to the cell surface continued during the trypsin digest and/or trypsin gained access to the endosomal compartment . Both forms of GPI-PLC behaved exactly the same as cytoplasmic proteins , suggesting that the majority of the protein is on the internal , not external , face of the plasma membrane . However , this assay would not have detected a small fraction of externally disposed GPI-PLC . All proteins were digested within three minutes if the plasma membrane was disrupted with detergent . GPI-PLC-eYFP was used as a reporter for mutational analysis of GPI-PLC . Each of the three cysteines in the CCGAC motif was mutated to serine both individually and in combination to make a total of seven mutants . The wild type and each mutant were expressed as a GPI-PLC-eYFP transgene from the endogenous locus in a −/− background [37]; therefore any GPI-PLC activity originated from the transgene . These transgenes used the endogenous 3′ untranslated region and resulted in expression levels lower than wild type cells ( Figure S1B ) . The insertion of the transgene had no obvious effect on growth or morphology ( data not shown ) . All the GPI-PLC-eYFP variants had the expected relative molecular mass and similar expression levels ( Figure 3A ) . Wild type and the single cysteine mutants appeared as two polypeptides with slightly different electrophoretic mobilities , whereas the double and triple cysteine mutants appeared homogeneous and with the same mobility as the faster migrating form of the wild type and the single cysteine mutants ( Figure 3A ) . The presence of fatty acids can reduce the rate of protein migration during SDS-PAGE [43] and the slower migrating form might represent more acylated variants . The doublet has been observed previously in both native [14] [44] and recombinant [45] GPI-PLC protein . The activity of the wild type and each mutant was examined using the detergent lysis assay ( Figure 3B ) . Two independent clones were analysed for each mutant and samples collected 20 minutes after addition of detergent . This time point was chosen as it was close to the end point of hydrolysis in a cell line with a single copy of wild type GPI-PLC ( 15 to 20 minutes as shown in Figure 1B ) . The wild type and all the mutants were active as the CRD epitope was produced , indicating that the cysteine motif was not required for GPI-PLC activity . These data confirmed earlier work showing no individual cysteine was necessary for activity in the recombinant protein [45] . Next , the ability of each mutant to access plasma membrane VSG was determined using the hypotonic lysis assay ( Figure 3C ) . One clone of each cell line was examined and samples collected 20 minutes after hypotonic lysis . The vast majority of VSG was released from cells expressing wild type or any of the three single cysteine mutants , indicating that the mutants were able to access VSG GPI-anchors on hypotonic lysis . Only a fraction of the VSG was released from one double cysteine mutant , CSGAS , the remainder pelleting with the cell bodies . Little or no VSG was released on hypotonic lysis of either of the two other double cysteine mutants or the triple cysteine mutant . These data indicate that the cysteine motif has a role in the ability of GPI-PLC to access the VSG GPI-anchors on hypotonic lysis and gives a more precise definition of the requirement than earlier work that showed that all three cysteines are required [18] . The inability of some cysteine mutants to hydrolyse the VSG anchor was investigated further by examining the subcellular localisation of GPI-PLC-eYFP variants by fluorescence microscopy ( Figure 4A ) . The wild type was localised to the plasma membrane , and was more concentrated on the flagellar and FP membranes than on the membrane of the cell body . The same localisation was observed when the eYFP tag was located at the N-terminus ( Figure S2 ) . This differs from the previously reported localisation to a narrow stripe on the outside of the flagellum membrane between the FAZ and the paraflagellar rod [20]; the possible reasons for the discrepancy are discussed below . The three single cysteine mutants also localised to the cell membrane in a similar pattern to the wild type with concentration along the entire length of the flagellar membrane . In addition to the plasma membrane localisation , the GPI-PLC variants were also present in the endomembrane system located between the nucleus and flagellar pocket in a fraction of cells . Wild type GPI-PLC-eYFP was associated with these membranes in 8/28 ( 29% ) cells , and the amount detected was low ( Table 1 ) . In cells expressing the single cysteine mutants SCGAC and CCGAS , all the cells examined ( n = 20 , n = 21 respectively ) had GPI-PLC-eYFP associated with the endomembranes . In addition , the signal in the mutants was stronger than in the wild type suggesting a larger fraction of the total was membrane associated . In cells expressing the single cysteine mutant CSGAC , 7/22 ( 32% ) contained GPI-PLC-eYFP associated with the same internal membranes , but again the endomembrane signal was stronger than in wild type cells . In the case of the wild type , there was no obvious morphological difference , such as cell cycle stage , between cells with and without internal membrane localised GPI-PLC-eYFP . The GPI-PLC-eYFP signal intensity from the flagellum was quantified to assess the impact of the mutations . The ratio of fluorescence intensity between a region of the flagellum that had extended beyond the cell body and a region of the anterior cell body was determined for the variants ( Figure S3 ) ; this ratio provides a semi-quantitative measure of relative abundance . The ratio of intensity ( flagellum/cell body ) in cells expressing wild type GPI-PLC-eYFP was 3 . 8±0 . 3 ( ±s . e . m ) ( n = 28 ) . This was reduced in the mutants: SCGAC 1 . 2±0 . 1 ( n = 20 ) , CSGAC 2 . 0±0 . 2 ( n = 22 ) and CCGAS 2 . 1±0 . 3 ( n = 21 ) . Thus , loss of one cysteine within the motif diminished the ability of GPI-PLC to concentrate on the flagellar plasma membrane and increased the size of the pool present on internal membranes lying between the flagellar pocket and nucleus . In all cells ( n>40 ) expressing one of the double cysteine mutants , CSGAS , GPI-PLC-eYFP was localised to the cell and flagellar membranes but to a greatly reduced level compared to the wild type and single cysteine mutants: the majority was associated with the endomembrane system ( Figure 4 ) . In all cells ( n>40 ) expressing the two other double cysteine mutants , SCGAS and SSGAC , GPI-PLC-eYFP localised predominantly to the same endomembrane system with a fraction present in the cytoplasm ( Figure 4A ) . In all cells ( n>40 ) expressing the triple cysteine mutant SSGAS , there was no longer any obvious membrane association and the protein appeared to be cytoplasmic ( Figure 4 ) . To test this further , rapid hypotonic cell lysis was performed , the lysate was separated into a soluble and pellet fraction and the components present in each fraction were analysed by Western blotting using the GPI-PLC antiserum ( Figure 4B ) . In this type of experiment , some cytoplasmic components will be trapped inside cell ghosts but all membrane and cytoskeletal components will be in the pellet . The cytoplasmic protein DHH1 was present in similar amounts in the supernatant and pellet fractions whereas the paraflagellar rod proteins PFR1 and PFR2 were present solely in the pellet . All forms of GPI-PLC-eYFP were present solely in the pellet with the exceptions of the SSGAC and SSGAS mutants , which were distributed in a similar manner to DHH1 , providing evidence that a significant fraction was not membrane attached . These data show that the cysteine motif , CCGAC , is necessary for membrane association and has a role in trafficking of GPI-PLC to the cell membrane ( summarised in Table 1 ) . The localisation of some cysteine mutants of GPI-PLC-eYFP to structures lying between the nucleus and flagellar pocket is consistent with association to one or more of the secretory , recycling and lysosomal endosomal compartments . Clathrin localises across a substantial part of the endosomal system [46] and a cell line was made expressing both GPI-PLC-eYFP and a clathrin light chain with a double tomato fluorescent protein tag at the N-terminus ( dTomFP-CLC ) . The dTomFP-CLC was expressed in the GPI-PLC-eYFP double cysteine mutant cell lines ( Figure S4 ) . The GPI-PLC-eYFP double cysteine mutants were localised predominantly in the posterior of the cell and partially co-localised with the dTomFP-CLC , indicating that a proportion of the mutant proteins were associated with these endosomal systems . However the co-localisation was not complete and there were regions of GPI-PLC-eYFP signal without a corresponding dTomFP-CLC signal and vice versa , so a definite conclusion about the precise compartment ( s ) occupied by the internal GPI-PLC could not be drawn . Trypanosoma congolense is the species most closely related to T . brucei for which a genome sequence is available . The amino acid sequence of the T . congolense GPI-PLC ( TcGPI-PLC ) has 58% identity with the T . brucei GPI-PLC ( TbGPI-PLC ) ( Figure S5 ) , and the conserved residues include the three cysteine motif . To determine whether the targeting signals were conserved , a transgene encoding TcGPI-PLC with a C-terminal eYFP tag ( TcGPI-PLC-eYFP ) was introduced into a T . brucei GPI-PLC −/− cell line . Expression was analysed by Western blotting using a GFP antibody ( Figure S6A ) , Expression was readily detected but the level of TcGPI-PLC was lower than the equivalent cell line containing a TbGPI-PLC-eYFP transgene . Next , the activity of the TcGPI-PLC against the VSG GPI-anchor in two independent clones was analysed using the detergent and hypotonic lysis assays ( Figure S6B ) . On detergent lysis there was a faint CRD signal detected from both clones , indicating that VSG with a hydrolysed GPI-anchor was produced . On hypotonic lysis there was partial release of VSG into the supernatant fraction . These results indicate that TcGPI-PLC was active against the T . brucei VSG GPI-anchor and was able to access the GPI-anchors on hypotonic lysis . However , TcGPI-PLC was not as active as TbGPI-PLC; it is unclear whether this was due to less activity against the T . brucei VSG GPI-anchor and/or lower expression . The localisation of TcGPI-PLC-eYFP was examined by fluorescence microscopy ( Figure 5 ) : it was present in the cytoplasm with a fraction associated with the cell membrane as there was a sharp line defining the shape of the cell . TcGPI-PLC-eYFP was not observed on the flagellum . As TcGPI-PLC-eYFP did not localise to the flagellum , a flagellar concentration signal was hypothesised to be located in the regions of TbGPI-PLC that did not share identity with TcGPI-PLC . To test this theory , a series of TbGPI-PLC and TcGPI-PLC hybrids was constructed ( Figure S7 and S5 ) . The points of fusion were located in regions of identity between TbGPI-PLC and TcGPI-PLC and the hybrids were expressed with an eYFP tag at the C-terminus in the GPI-PLC −/− cell line . The expression of the hybrids was analysed by Western blotting with a GFP antibody ( Figure S6A ) : all were the expected relative molecular mass and the expression levels were all similar to that of wild type TbGPI-PLC-eYFP . The localisation of the hybrids was examined by fluorescence microscopy ( Figure 5 ) . Hybrid 1 consisted of the N-terminal region of TbGPI-PLC up to the CCGAC motif , after which the TcGPI-PLC sequence was present . Hybrid 1 localised to the cell and flagellar membranes but there was also a large proportion associated with the endomembrane system and possibly some present in the cytoplasm ( Figure 5 ) . In order to quantify the effects of the different hybrids on GPI-PLC localisation , the ratio of the fluorescence intensity from a region of the flagellum away from the cell body to the fluorescence intensity of a region of the anterior cell body was calculated as above . Wild type TbGPI-PLC had a ratio of 3 . 8±0 . 3 ( ±s . e . m ) ( n = 28 ) ; hybrid 1 had a ratio of 1 . 0±0 . 1 ( n = 40 ) . The N-terminal part of TbGPI-PLC up to and including the CCGAC motif was able to support membrane localisation but hybrid 1 did not concentrate in the flagellum . Hybrid 2 was derived from hybrid 1 by replacing the C-terminal 32 with the equivalent region from TbGPI-PLC . Hybrid 2 localised to both the cell and flagellar membranes and a fraction associated with the endomembrane system , and the overall appearance of the cells was similar to that of the wild type TbGPI-PLC ( Figure 5 ) . However for hybrid 2 the ratio of fluorescence intensity was 1 . 2±0 . 1 ( n = 14 ) , only marginally higher than that for hybrid 1 . Hybrid 3 consisted of TbGPI-PLC but with the C-terminal 32 residues replaced with the equivalent sequence from TcGPI-PLC . Hybrid 3 localised to the cell and flagellar membranes with a large proportion associated with the endomembrane system ( Figure 5 ) and a fluorescence intensity ratio of 1 . 0±0 . 1 ( n = 23 ) . Hybrid 3 was similar to hybrid 1 in terms of its localisation pattern and fluorescence intensity ratio , suggesting that the C-terminal 32 residues were necessary for the concentration of TbGPI-PLC on the flagellar membrane . The C-terminal 32 residues of TbGPI-PLC and TcGPI-PLC are similar with 50% identity . One of the differences is the location of a proline residue: the C-terminal 32 residues from both proteins contain a single proline but the location is not conserved ( Figure S5 ) . Hybrid 4 was derived from hybrid 3: it contained the C-terminal 32 residues from TcGPI-PLC altered to move the single proline to the equivalent position in TbGPI-PLC ( P333T and L340P ) . Hybrid 4 localised to both the cell and flagellar membranes and was concentrated on the flagellar membrane ( Figure 5 ) . The fluorescence intensity ratio for hybrid 4 was 1 . 9±0 . 2 ( n = 13 ) , an increase in the concentration on the flagellar membrane compared to hybrid 3 . To determine whether the P340 was necessary for concentration on the flagellar membrane , a point mutation of P340G in TbGPI-PLC was tested . The mutation resulted in a dramatic change in the localisation: GPI-PLC-eYFP P340G was evenly distributed over the cell and flagellar membranes , which was confirmed by the drop in the fluorescence intensity ratio to 0 . 4±0 . 03 ( n = 10 ) ( Figure 5 ) . This result showed that P340 was necessary for the flagellar membrane concentration of TbGPI-PLC but not for association with the plasma membrane . Two independent clones expressing GPI-PLC-eYFP P340G were assayed for VSG release after hypotonic lysis ( Figure S8 ) . The mutant was able to release VSG on hypotonic lysis , indicating that it was active and able to access its substrate . Further mutants designed to determine the full extent of the motif containing P340 were inconclusive . GPI-PLC mutants containing alterations to the C-terminal side of P340 ( M341A and 341 MNAV to 341 AAAA ) were able to concentrate on the flagellar membrane ( Fig . S9 ) . A mutation to the N-terminal side ( K339A ) destabilised the GPI-PLC-eYFP to the extent that localisation data was not obtained . The motif causing flagellar concentration does not extend to the C-terminal side of P340 but may extend to the N-terminal side . Little is known about how proteins are concentrated in different domains of the trypanosome plasma membrane . Components from the structures that are believed to delimit the different domains have been identified , such as BILBO1 , a component of the FPC , located at the junction of the cell body and FP plasma membrane domains . In proliferating cells , RNAi knockdown of BILBO1 results in a failure to form the new FPC and FP . In addition , the new flagellum is detached from the cell body and migrates to the posterior of the cell [24] . A feature of the trypanosome flagellum is that it is attached to the cell body as far as the anterior pole of the cell and then extends beyond the cell body . Components of the FAZ are necessary for the attachment: RNAi knockdown of FAZ components results in the newly synthesised flagellum being detached from the cell body , but the formation of the FP is unaffected . FLA3 is a FAZ component upregulated in bloodstream form trypanosomes and RNAi knockdown causes the new flagellum to be synthesised without attachment to the cell body but , importantly , there is no apparent effect on the FPC or FP [47] . Cell lines were constructed expressing a GPI-PLC-eYFP transgene and containing plasmids to direct tetracycline-inducible RNAi against BILBO1 or FLA3 . The population doubling times of the BILBO1 and FLA3 RNAi cells were measured as 12 and 7 hours respectively . RNAi was then induced for a single doubling time before the localisation of GPI-PLC-eYFP was determined . As the induced cells had undergone one cell cycle , the majority of the population of cells had one old flagellum , constructed before induction of RNAi , and one new flagellum displaying the RNAi phenotype . This time point allowed any changes in GPI-PLC localisation caused by the RNAi to be detected by comparing the new and old flagella . On induction of BILBO1 RNAi , the new flagellum was detached and had migrated to the posterior pole of the cell in the majority of cells as described previously [25] . In these cells , GPI-PLC-eYFP localised to the plasma membrane ( Figure 6A ) ; however , there was a reduction in GPI-PLC-eYFP concentration in the new flagellum and in some cells there was no difference between the intensity of GPI-PLC-eYFP in the new flagellum and the intensity on the cell body . In all the cells examined , the old flagellum retained a higher concentration of GPI-PLC-eYFP than the plasma membrane , as might be predicted since the loss of BILBO1 does not affect the existing FPC . The localisation of GPI-PLC-eYFP after induction of FLA3 RNAi is shown in Figure 6B . Cells were examined that had an attached old flagellum and a detached new flagellum . FLA3 knockdown had no apparent effect on GPI-PLC-eYFP localisation: GPI-PLC-eYFP localised to the plasma membrane and was concentrated to a similar extent in both the old and new flagella . To quantify the effect of BILBO1 or FLA3 knockdown on the ability of GPI-PLC-eYFP to concentrate in the detached new flagellum , the ratio of the GPI-PLC-eYFP fluorescence intensity between a region of the attached old flagellum that extended beyond the cell body and the detached new flagellum was calculated; both values had the background fluorescence intensity subtracted before the ratio was calculated ( Figure 6 ) . A value close to 1 would be expected , as the old and new flagella should contain the same set of proteins at similar levels . The ratio for the FLA3 knockdown cells was 1 . 2±0 . 1 ( ±s . e . m ) ( n = 8 ) , implying that the detachment of the flagellum by FLA3 knockdown does not greatly affect the ability of GPI-PLC to concentrate in the detached new flagellum . For BILBO1 knockdown cells , the ratio was 2 . 2±0 . 3 ( n = 18 ) . This value was significantly different ( p<0 . 005 ) to that found for the FLA3 knockdown , indicating that there was a loss of GPI-PLC-eYFP flagellar concentration on BILBO1 knockdown . Furthermore , the defect in GPI-PLC-eYFP flagellar concentration appeared to be related to the loss of the FPC and/or FP and not to the detachment of the flagellum , as there was no apparent loss of GPI-PLC-eYFP concentration in the flagellum on FLA3 knockdown .
Bloodstream form trypanosomes contain sufficient GPI-PLC to release the entire VSG coat within a few minutes , yet this release only occurs after the integrity of the plasma membrane has been compromised . Previous work on the regulation of GPI-PLC activity has shown that there is dynamic acylation [19] and that three cysteines present in the CCGAC motif are necessary for acylation and access to VSG substrate in ruptured cells [18] . The latter suggested that localisation to membranes was necessary for access to the VSG and was consistent with the subsequent finding that GPI-PLC is concentrated on the flagellar membrane [20] . Here , the determinants of this localisation have been identified and the effect of localisation on access to VSG substrate determined . The major findings are: ( i ) the cysteines in the CCGAC motif act in a dose-dependent manner in determining localisation to the plasma membrane; ( ii ) association of the GPI-PLC with the plasma membrane is necessary for VSG release on hypotonic lysis , association with the endomembrane system between the nucleus and flagellar pocket is not sufficient; ( iii ) a proline , residue 340 , close to the C-terminus is necessary for concentration of GPI-PLC on the flagellar membrane; ( iv ) GPI-PLC does not concentrate on a flagellum without a flagellar pocket collar and/or flagellar pocket . The CCGAC motif is acylated [18] , [19] and the cysteines are required for complete localisation to the plasma membrane ( Figure 4 ) . Once associated with the plasma membrane , GPI-PLC is predominantly on the cytoplasmic face as shown by the lack of proteolysis of GPI-PLC in live cells when treated with a high external concentration of trypsin ( Figure 2 ) . The reason for the discrepancy between this finding and the earlier finding that GPI-PLC is localised on the external face of the plasma [20] probably lies in the assay used . Here , trypsin access to the GPI-PLC in live cells assays for all the GPI-PLC in the cell , whereas the surface labelling used previously [20] did not as the fraction labelled was not determined . In addition Bülow and colleagues analysed the sensitivity of GPI-PLC activity to trypsin and found that its activity was insensitive to trypsin until detergent was added to the cells [48] . It is also worth noting that cytoplasmically acylated proteins have , to date , always been reported to be associated with the cytoplasmic face of membranes [23] , [49] , [50] , [51] , [52] , [53] . The subcellular localisation of GPI-PLC-eYFP to the plasma membrane and concentration on the flagellar membrane is distinct from that idenitified by previous studies which have used immunofluorescence and other assays to propose a range of localisations; the list is described and discussed in [20] . There is only one report [20] that demonstrated monospecificity of the antibody by Western blotting and specificity in immunofluorescence using GPI-PLC −/− cells . Deconvolution of confocal images of immunofluorescence experiments was used to report a localisation to a narrow stripe on the flagellar membrane near the FAZ . Here , the localisation of GPI-PLC-eYFP reported has a wider distribution , encompassing the whole plasma membrane , with concentration on the flagellar membrane . Why is there a discrepancy between the results here and the earlier report [20] ? There is a range of possibilities; the first is that the use of the eYFP tag alters the localisation . This must remain a possibility although it is unlikely as the same localisation was obtained with the eYFP at the N- and C-termini ( Figure S2 ) . An alternative is that the fixation used prior to the immunofluorescence-determined subcellular localisation restricted the access of the antibodies to a subset of the GPI-PLC and it is worth noting that fixation conditions did affect the results [20] . Further , there is a history of contradictory results on the subcellular localisation of the GPI-PLC determined by immunofluorescence suggesting that access to and/or detection of GPI-PLC is not facile . A second possibility for the difference is the deconvolution of the images that resulted in the original localisation of GPI-PLC-eYFP to the flagellum [20] . The deconvolution process removed all but the strongest fluorescent signal for GPI-PLC-eYFP as if the remainder were out-of-focus fluorescence and producing a bias towards a discrete structure . The result was an artefactual localisation to the region where GPI-PLC-eYFP was most concentrated , the flagellar membrane . To illustrate this point , a comparison PFR1-mCherryFP and GPI-PLC-eYFP localisation shows the dispersion of GPI-PLC-eYFP to the plasma membrane compared to the discrete localisation of PFR1 to the paraflagellar rod ( Figure S10 ) . The number of cysteine residues affected the mobility of GPI-PLC on SDS-PAGE . Forms with two or more cysteines appeared as a doublet whereas forms with one cysteine or none appeared as a single band , equivalent to the faster migrating form . This observation provides evidence that the doublet results from a mixture of differently acylated forms present in the cell at time of lysis and that the slower migrating band is multiply acylated . The number of acylated residues within the cysteine motif has not been quantified but as the loss of one cysteine residue reduces the flagellar concentration of GPI-PLC it appears that all three have to be modified for correct localisation . Acylation of GPI-PLC is a dynamic process and occurs in cells where protein synthesis has been inhibited with cycloheximide . It is likely that the acylation patterning on GPI-PLC can be continuously remodelled [19] . The ability to remodel the modifications on the protein is similarly found for palmitoylated proteins in other systems , such as SNAP25 in mammalian cells [54] . The identity of the acyl transferase acting on GPI-PLC is unknown: there are >10 in the trypanosome genome [23] . However , the recombinant protein produced in E . coli also runs as a doublet on SDS-PAGE [45] , which suggests that either the GPI-PLC is a substrate for a bacterial non-specific acyl transferase or that the acylation is spontaneous . In either case , no great specificity is required to acylate GPI-PLC . The cysteine motif CCGAC had previously been implicated in the regulation of the activity and localisation of GPI-PLC but the role of individual cysteines within the motif had not been studied [18] . All mutants retained activity for VSG GPI-anchor hydrolysis on detergent lysis of cells . Thus , acylation of the CCGAC motif is not necessary for activity in vitro and no large reduction in activity was visible in the absence of acylation . In contrast , the ability to access the VSG substrate on hypotonic lysis was strictly dependent on localisation . When a single cysteine was mutated there was a decrease in the concentration of GPI-PLC within the flagellar membrane and an increase in the association of GPI-PLC with the endomembrane system . However , there was no observable effect on access to VSG substrate on hypotonic lysis . When two cysteines were mutated , there was a dramatic redistribution of GPI-PLC from the flagellar membrane to the endomembrane system . There was also an effect on the ability of the mutants to access the VSG on hypotonic lysis . With the CCGAC→CSGAS mutant , there was only partial release of VSG and the SSGAC and SCGAS mutant were unable to release VSG from the plasma membrane . The mutation of all three cysteines caused the GPI-PLC to localise to the cytoplasm and it was unable to access the GPI-anchor on hypotonic lysis . These data provide evidence that GPI-PLC is only able to access the GPI-anchors on hypotonic lysis when localised to the plasma membrane and that the plasma membrane is the site of action of GPI-PLC . The concentration of GPI-PLC on the flagellar membrane was not required for the release of VSG on hypotonic lysis as the GPI-PLC P340G mutant was able to hydrolyse the VSG GPI-anchors , despite the mutation resulting in the redistribution of GPI-PLC . These observations are consistent with an earlier model in which osmotic rupture of the cells allowed VSG and GPI-PLC to mix through passive diffusion around the edges of discontinuities in the plasma membrane [35] . Since the VSG and GPI-PLC have to be in the same membrane for hydrolysis to occur [8] any vesicles derived from endosomes or free GPI-PLC in the cytoplasm presumably cannot access VSG [34] . The simplest model for GPI-PLC regulation is that it cannot access the VSG substrate in live cells as it is on the opposite side of the membrane . In proliferating bloodstream form trypanosomes , the half-life of the VSG is several cell generations [12] , [13] and thus this regulation is very effective . However , the VSG is shed during differentiation from bloodstream forms to insect infective procyclic forms . During the process , both GPI-anchor hydrolysis and proteolytic cleavage by MSP-B , a cell surface metalloprotease , result in VSG release: either enzyme is sufficient to complete the process [55] . This means that the GPI-PLC has regulated access to the VSG during the differentiation process in cells that remain viable and go on to divide . The mechanism remains to be investigated but might provide an explanation for the concentration of the GPI-PLC on the flagellum . In addition to acylation , proline 340 is necessary for concentration of the GPI-PLC on the flagellar membrane . GPI-PLC from the closely related T . congolense ( TcGPI-PLC ) has 58% identity with the T . brucei GPI-PLC ( TbGPI-PLC ) , including the cysteine motif . However , in T . brucei the localisation of TcGPI-PLC appeared to be mainly cytoplasmic with a small proportion localised to the cell membrane . TcGPI-PLC did not have the same pattern of localisation as TbGPI-PLC despite the presence of the cysteine motif , suggesting that the cysteine motif is not sufficient for the concentration of GPI-PLC within the flagellum . Other proteins that rely on acylation for flagellar localisation , such as rhodopsin , also have another flagellar localisation signal [28] . Therefore a flagellar concentration signal was hypothesised to be located in the regions of the TbGPI-PLC sequence that did not share identity with the TcGPI-PLC sequence ( Figure S4 ) . A single proline residue with this C-terminal region was shown to be a key GPI-PLC flagellar concentration signal and its mutation resulted in GPI-PLC failing to concentrate on the flagellar membrane . Two other flagellar membrane localisation motifs have been described in kinetoplastids . The flagellar calcium binding protein from Trypanosoma cruzi is myristoylated and palmitoylated at the N-terminus . These acylations are necessary but not sufficient for flagellar localisation , which also requires three lysine residues located between residues 13 and 22 [56] . Furthermore , residues 1 to 24 were sufficient for localisation of a GFP reporter to the flagellar membrane and this same reporter was localised to the flagellum in Leishmania amazonensis suggesting a conserved mechanism . The second motif was identified in the flagellar glucose transporter 1 from Leishmania mexicana ( LmxGT1; LmxM . 36 . 6300 ) [57] . LmxGT1 contains 12 membrane-spanning segments located after residue 131; the N-terminal 130 residues are present in the cytoplasm . A mutational analysis of the N-terminal 130 residues to locate sequences necessary for flagellar localisation identified a motif between residues 95–97 with the sequence asn-pro-met ( NPM ) . This is similar to the context of P340 in GPI-PLC lys-pro-met ( KPM ) , and it is possible that LmxGT1 and GPI-PLC are substrates for the same flagellar concentration mechanism . The proline may be part of a motif or may act as a substrate for a peptidyl prolyl isomerase that regulates accumulation on the flagellar membrane . The isomerisation of the proline could re-orient the structure of the GPI-PLC at the C-terminus; the rearranged C-terminus may then be able to bind to another factor that enables flagellar concentration . Proteomic analysis has identified peptidyl prolyl isomerases within the trypanosome flagellum so the appropriate enzymes are present [58] . There are minimally three domains in the plasma membrane in trypanosomes: the cell body , the flagellar pocket and the flagellum . The FPC is a protein ring structure located at the point at which the flagellar pocket invaginates from the cell body membrane and is necessary for the formation of a FP . BILBO1 was identified as a component of the FPC , which maintains the structure of the FP [25] . On BILBO1 knockdown by RNAi there is loss of the new FP and new FPC , and detachment of the new flagellum . The knockdown affected GPI-PLC localisation: it was no longer concentrated in the new flagellum . In order to determine if the loss of GPI-PLC concentration was due to the loss of the FPC and/or FP or flagellar attachment , a second RNAi was carried out that targeted FLA3 . RNAi knockdown of FLA3 results in the detachment of the new flagellum but importantly the FPC remains intact [47] . The mean ratio of GPI-PLC concentration between the old and new flagella in the FLA3 knockdown cells was approximately 1 , which would be expected , as the new and old flagella should contain a similar set of proteins; the detachment of the flagellum therefore does not affect the localisation of GPI-PLC . In the absence of a FPC and/or FP there is no concentration of GPI-PLC on the flagellum but it is not possible to distinguish whether flagellar concentration is a direct function of the FPC or the FP as both are lost on BILBO1 knockdown . The observations do provide evidence for the long-held view that there are selective diffusion barriers for proteins at the boundaries between the plasma membrane domains in trypanosomes . A flagellum-associated diffusion barrier has been demonstrated in other organisms by the knockdown of CEP290 , a component of the transition zone in the flagellum of Chlamydomonas reinhardtii , and Septin 2 , found at the base of mammalian primary cilia [59] , [60] . The loss of these proteins led to a redistribution of proteins between the flagellar and cell membrane . However , during BILBO1 knockdown the new FP is also lost: as GPI-PLC is trafficked through the FP , the reduction in GPI-PLC-eYFP concentration within the new flagellum could be due to a lack of GPI-PLC reaching the new flagellum . Although BILBO1 is a key component of the FPC diffusion barrier , which restricts the movement of GPI-PLC and could therefore be responsible for maintaining GPI-PLC concentration within the flagellum , there is no evidence that BILBO1 interacts directly with GPI-PLC .
Trypanosoma brucei Lister 427 bloodstream form cells were used in all experiments . All cells were grown in HMI-9 medium with 10% foetal bovine serum . All experiments were performed with logarithmically growing trypanosomes at a cell density of less than 1×106 cells/ml . The GPI-PLC −/− cell line has been described [37] and was made by replacing the entire gene at both alleles with selectable marker genes . The transgenic cell lines expressing GPI-PLC variants in the GPI-PLC −/− background were made as described [37] by returning a copy of the modified gene , including 5' and 3′ UTRs , to the endogenous locus with expression relying on endogenous transcription . The cell line GPI-PLC-eYFP/− was made by modifying the remaining GPI-PLC allele in the GPI-PLC/− cell line with a C-terminal eYFP tag as described [36] . The cell line expressing both wild type and GPI-PLC-eYFP ( GPI-PLC-eYFP/+ ) was made using the same approach except that the transgene was introduced into a GPI-PLC +/+ cell line . The RNAi experiments used the Lister 427 328 . 114 single marker cell line [38] ( a kind gift of George Cross ) . An endogenous GPI-PLC allele in the single marker cell line was tagged at the C-terminus with eYFP as described [36] . The expression of the N-terminal dTomato fluorescent protein tagged clathrin light chain was from a modified endogenous locus and relied on endogenous transcription [36] . Transgenic trypanosomes were generated using standard procedures [61] . Details of all plasmids are described in Table S1 . The sequences of all plasmids are available from the authors . Cells were harvested by centrifugation and washed with 10 ml of HMI-9 without serum and re-centrifuged . The cells were then resuspended in HMI-9 without serum at 1×108 cells/ml , 0 . 05 volumes of 10% ( v/v ) Triton X-100 was added and the mixture incubated at room temperature . Samples were removed after 0 , 2 , 5 , 10 and 20 minutes and were analysed by SDS-PAGE and Western blotting . Cells were harvested by centrifugation and washed with 10 ml of ice-cold HMI-9 without serum and re-centrifuged . The cells were resuspended in ice-cold 1 mM TLCK at 1×108 cells/ml and incubated on ice for 5 minutes then incubated at 37°C for a further 15 minutes . Samples were removed at 0 , 1 , 5 , 7 , 10 , 15 and 20 minutes and fractionated into pellet and supernatant fractions by centrifugation in a microfuge , 13 , 000 rpm for 1 minute . The samples were analysed by SDS PAGE and Western blotting . The GPI-PLC cysteine mutants were subjected to a simplified assay in which the cells were resuspended in 1 mM TLCK at 1×108 cells/ml and incubated for 20 minutes at room temperature followed by SDS PAGE and Western analysis . Hypotonic lysis was performed as above . After 5 minutes on ice the samples were vortexed for 30 seconds , a sample of the lysate was taken and the remainder separated into the pellet and supernatant fraction by centrifugation in a microfuge 13 , 000 rpm for 1 minute . The samples were then analysed by Western blotting as above . Western blots were performed using standard protocols . However a high pH transfer buffer ( 19 mM glycine , 20 mM Tris base , 0 . 08% ( w/v ) SDS and 15% ( v/v ) methanol ) was used to ensure consistent transfer of GPI-PLC as it has an unusually high pI . Detection was by either by a fluorescent secondary antibody using the Odyssey Infrared Imaging System or by enhanced chemiluminescence ( ECL ) . Recombinant full length His-tagged GPI-PLC was produced using the baculovirus system . The recombinant protein was purified on a nickel affinity column . The purified protein was run into an SDS-PAGE gel and then electro-eluted from the gel . The eluted protein was used to inoculate a rabbit ( Covalab ) . Recombinant GPI-PLC was blotted onto PVDF and used to affinity purify anti-GPI-PLC antibodies from the rabbit sera . Anti-CRD was prepared by immunising a rabbit with VSG ILTat1 . 21 and antibodies were affinity purified using immobilised VSG MITat1 . 2 . Other antibodies have been described: ISG65 [62] , BiP [63] , DHH1 [41] , PFR ( L13D6 ) [64] . Anti-GFP was obtained from Invitrogen ( A11122 ) . Cells were harvested by centrifugation washed with 10 ml of ice-cold HMI-9 without serum and recentrifuged . For live cell trypsin treatment the cells were resuspended in HMI-9 without serum to 4×107 cells/ml and trypsin was added to a final concentration of 40 µg/ml . For lysed cell trypsin treatment the cells were resuspended in HMI-9 without serum , lysed with 0 . 05 volumes of 10% ( v/v ) Triton X-100 and trypsin added to a final concentration of 100 µg/ml . The mixture was incubated at room temperature . Samples were removed at 0 , 3 , 6 , 9 , 12 , 15 and 18 minutes and analysed by SDS-PAGE and Western blotting . ECL was used to detect the GPI-PLC-eYFP signal . The GPI-PLC , ISG65 , PFR1 , PFR 2 and DHH1 signals were detected using the Odyssey Infrared Imaging System and quantification was carried out using the Odyssey software . Cells from 1 ml culture were centrifuged at 10000 rpm for 1 minute in a microfuge and washed with 1 ml of HMI-9 without serum . The cells were then centrifuged as before and resuspended in 50 µl of HMI-9 without serum . To immobilize the cells , formaldehyde was added to a final concentration of 0 . 075% ( v/v ) for 5 minutes , images were taken within the next 20 minutes . Great care was taken to check the immobilisation with formaldehyde had no visible effect on the distribution of the tagged protein . Cells were visualised using a Zeiss Axioimager M1; images were recorded using the Axiovision software ( Zeiss ) and then imported into Adobe Photoshop . Fluorescent intensity measurements were performed on individual cells using the ImageJ software ( NIH ) . The fluorescent intensity was measured from an area of the cell body or an area of the flagellum , where the flagellum had extended beyond the cell body , in cells with a single flagellum . The mean fluorescent intensity was taken from these measurements and the background fluorescence subtracted before a ratio of the fluorescent intensities was calculated . The ratios reported are the mean values ± standard error of the mean . An unpaired Student t-test was used to calculate the level of significance between the means .
|
African trypanosomes are unicellular parasites with a single flagellum that maintain a persistent infection through antigenic variation based on changes in a densely packed cell surface coat of variant surface glycoprotein ( VSG ) . The cells also contain an enzyme , GPI-PLC , able to shed the VSG from the cell surface . However , the activity is regulated and substantial shedding only occurs from dying cells . The GPI-PLC is found predominantly on the membrane of this flagellum . Here , we have investigated the relationship between this subcellular localisation and VSG shedding ability of the GPI-PLC . We found that two motifs are important: a cluster of three cysteines that are modified by the addition of fatty acids and a proline , mutation of which caused the redistribution of GPI-PLC from the flagellar to the plasma membrane . Localisation of GPI-PLC to the plasma membrane is necessary for GPI-PLC to access the VSG in dying cells . Finally , the correct localisation of the GPI-PLC was dependent on a functional flagellar pocket . These results have provided a significant and exploitable insight into the regulation of GPI-PLC and more generally into how proteins are targeted to the flagellum membrane .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"cell",
"biology",
"membranes",
"and",
"sorting",
"biology"
] |
2013
|
Determinants of GPI-PLC Localisation to the Flagellum and Access to GPI-Anchored Substrates in Trypanosomes
|
In recent years , there has been a growing interest in teichoic acids as targets for antibiotic drug design against major clinical pathogens such as Staphylococcus aureus , reflecting the disquieting increase in antibiotic resistance and the historical success of bacterial cell wall components as drug targets . It is now becoming clear that β-O-GlcNAcylation of S . aureus wall teichoic acids plays a major role in both pathogenicity and antibiotic resistance . Here we present the first structure of S . aureus TarS , the enzyme responsible for polyribitol phosphate β-O-GlcNAcylation . Using a divide and conquer strategy , we obtained crystal structures of various TarS constructs , mapping high resolution overlapping N-terminal and C-terminal structures onto a lower resolution full-length structure that resulted in a high resolution view of the entire enzyme . Using the N-terminal structure that encapsulates the catalytic domain , we furthermore captured several snapshots of TarS , including the native structure , the UDP-GlcNAc donor complex , and the UDP product complex . These structures along with structure-guided mutants allowed us to elucidate various catalytic features and identify key active site residues and catalytic loop rearrangements that provide a valuable platform for anti-MRSA drug design . We furthermore observed for the first time the presence of a trimerization domain composed of stacked carbohydrate binding modules , commonly observed in starch active enzymes , but adapted here for a poly sugar-phosphate glycosyltransferase .
Methicillin-Resistant Staphylococcus aureus ( MRSA ) is a leading cause of life-threatening nosocomial infections including pneumonia , bacteremia , and surgical wound infections [1] . Due to wide-spread β-lactam antibiotic resistance , the first-line treatment for serious MRSA infections has been vancomycin , a glycopeptide class antibiotic . However rising resistance to vancomycin has forced the use of undesirable alternatives with high cost and dose limitations due to adverse events [2] . Both β-lactam and glycopeptide antibiotics disrupt peptidoglycan cross-linking that eventually weakens the integrity of the bacterial cell wall and leads to lysis . Due to the efficacy and safety profile of β-lactam antibiotics , re-sensitization of MRSA to these drugs is a promising option that entails understanding of complex resistance mechanisms . Resistance in MRSA mainly evolves from the expression of PBP2a , a β-lactam-insensitive penicillin-binding protein that can cross-link peptidoglycan in the presence of clinically relevant concentrations of nearly all β–lactam antibiotics ( reviewed in [3] ) . Interestingly , recent reports have uncovered the role of wall teichoic acids and more specifically , their β-O-GlcNAc decorations , in mediating MRSA resistance to β-lactams [4 , 5] , opening new avenues for drug discovery efforts aimed at re-sensitization . Teichoic acids are anionic glycopolymers that compose an astonishing 60% of the dry weight of the cell wall in Gram-positive bacteria [6] . These polymers may either be attached to membranes in the form of lipoteichoic acids ( LTAs ) or transferred onto peptidoglycan as wall teichoic acids ( WTAs ) . Collectively , TAs are implicated in diverse processes such as coping with environmental stress [7 , 8] , interaction with receptors and biomaterials [9 , 10] , induction of inflammation [11–13] , phage binding [14 , 15] , immune evasion [16] , biofilm formation [17] , resistance to lysozyme [18] , and resistance to antimicrobial molecules [5 , 19–21] . This adaptability arises largely from D-alanylation and glycosylation of TA polyol hydroxyl groups , influencing the physical and interactive properties of the cell wall . In most S . aureus strains , WTAs consist of polyribitol phosphate ( polyRboP ) chains of 40–60 repeats that are attached to the peptidoglycan via a disaccharide linkage unit to C6 hydroxyls of occasional N-acetylmuramic acid residues [22] . The C4 hydroxyls of S . aureus WTAs are furthermore heavily substituted with N-acetylglucosamine ( GlcNAc ) via α- or β-O-linkages . The configuration of the glycosidic linkage varies according to strain , with some having exclusive α- or β-O-linked GlcNAc , and others displaying a mixture [23 , 24] . In S . aureus , WTA GlcNAcs serve as receptors for phage binding [15] , have long been recognized as important antigens in the host-antibody response [25–27] , and have more recently been implicated in biofilm formation [28] . Furthermore , the stereochemistry of GlcNAc glycosidic linkages appears to directly influence both the biology and pathogenicity of S . aureus and other Gram-positive bacteria on a strain-specific level . The enzymes responsible for S . aureus WTA GlcNAcylation are the α-glycosyltransferase TarM and the β-glycosyltransferase TarS . Both these enzymes reside in the cytoplasm and decorate nascent WTA chains before transport and attachment to the peptidoglycan sacculus . Of significance is the recent discovery that β-O-GlcNAcylation of S . aureus WTA is specifically responsible for methicillin resistance in MRSA , which may be due to the possible direct or indirect recruitment of the β-lactam insensitive PBP2a that mediates resistance[5 , 29] . Accordingly , the deletion of TarS has been shown to result in the re-sensitization of MRSA strains to β-lactam antibiotics [5] . TarS mediated WTA β-O-GlcNAcylation has also been implicated in the induction of anti-WTA IgG-mediated complement activation and opsonophagocytosis in clinically isolated S . aureus strains [30] . We have recently published the first structure of TarM and elucidated its catalytic mechanism [31] . In this report , we present the first structure of TarS in the presence of donor substrate UDP-GlcNAc , elucidate various features involved in catalysis , and describe a novel trimerization domain composed of tandem carbohydrate binding motifs . Due to the pivotal role of TarS in MRSA resistance , its structure is particularly valuable for rational drug design efforts in combination therapies aimed at MRSA re-sensitization .
The structures presented here are of the TarS full-length protein ( 1–573 ) , the TarS1-349 ( 1–349 ) construct consisting of the catalytic domain ( 1–319 ) and the linker ( 320–352 ) , and the TarS217-573 ( 217–573 ) construct consisting of the catalytic domain C-terminal helical bundle ( 217–319 ) , the linker and the trimerization domain ( 353–573 ) ( S1 Fig ) . Refinement statistics of the final structural models are presented in Table 1 . The full-length TarS crystals diffracted with strong anisotropy and a high resolution limit of 4 Å was applied . Due to the limited structural resolution of full-length TarS , efforts were made to obtain higher resolution structures of the individual protein domains . Limited ( thermolysin ) proteolysis of the purified full-length TarS protein resulted in crystals diffracting to ~ 2 . 3 Å resolution , with SAD phasing using an iodide derivative revealing cleavage of the N-terminal region of the catalytic domain ( 1–216 ) and the structure of the TarS217-573 as described ( Fig 1 ) . TarS1-349 , described above , was isolated by designing a structure-guided truncation mutant ( based on the TarS217-573 structure ) that resulted in a structure at 2 . 3 Å resolution with intact UDP-GlcNAc bound in the active site ( Fig 2A ) . The intact ( rather than hydrolyzed ) UDP-GlcNAc in the active site was achieved by soaking crystals in increased concentrations of substrate ( 50 mM ) before freezing . Native and UDP bound structures were also obtained to similar resolution ( S2 Fig ) . The higher resolution TarS1-349 and TarS217-573 structures were subsequently used as molecular replacement search models to solve the structure of the full-length TarS , revealing a “hanging basket” like structure ( Fig 3A ) with variation in the relative domain orientation of the three catalytic domains with respect to the trimerization domain ( Fig 3B ) . TarS furthermore displays a pronounced electrostatic sidedness , which may be related to membrane localization or substrate/partner interactions ( Fig 3C ) . In light of the data quality , refinement was closely monitored using the higher resolution structures as restraints and validated by the visualization of OMIT maps for both protein ( S3 Fig ) and ligands ( S2C Fig ) . Although TarS was co-crystallized in the presence of the UDP-GlcNAc sugar donor , only the cleaved UDP product seemed to be observed ( ligand mFo-dFc simulated annealing omit electron density shown in S2C Fig ) and the crystals were too sensitive to withstand soaking in higher concentrations of UDP-GlcNAc as for the TarS1-349 crystals . The overlap of the TarS1-349 and TarS217-573 structures with regard to the C-terminus of the catalytic domain and the linker region ( 217–349 ) further allowed us to superimpose these domains relative to each other and to the lower resolution full-length structure , providing us with essentially high-resolution views for the entire span of the TarS structure . TarS possesses a catalytic domain with a canonical GTA fold ( one of two distinct A/B folds characteristic of nucleotide-sugar dependent glycosyltransferases ) consisting of two closely associated α/β/α sandwich Rossmann motifs that abut and form a continuous central β-sheet ( Fig 2A and S1 Fig ) . According to the Carbohydrate-Active Enzymes ( CAZY ) database that classifies enzyme families based on sequence homology [32] , TarS belongs to the GT2 family that includes a large ( and evolutionarily ancient ) group of GTA fold stereochemistry-inverting enzymes acting on a variety of substrates , many of which are polysaccharides ( e . g . cellulose , chitin , hyaluronic acid , etc . ) . TarS possesses a D ( 91 ) XDD motif with aspartates coordinating a metal cation , a feature that is typical of the GTA superfamily ( Fig 4B ) [33] . The structures of native and UDP bound TarS1-349 were similar to that obtained for the lower resolution UDP bound full-length TarS ( overall main chain root mean squared deviation ( rmsd ) of 0 . 56 Å over 335 atom pairs ) , with minor differences in the linker region ( 320–349 ) . Brief soaking of the TarS1-349 crystals with UDP-GlcNAc resulted in an enzyme-donor complex with UDP-GlcNAc trapped in the active site as evidenced by the ligand mFo-dFc simulated annealing omit electron density ( S2C Fig ) . This structure displayed differences in two key loop regions that appear to be important for catalysis ( discussed below; Fig 2B and 2C ) . Two C-terminally localized regions in the TarS217-573 structure ( encapsulating the trimerization domain ) , designated here as C1 ( 353–495 ) and C2 ( 496–573 ) ( S1 Fig ) , were also observed , where a series of β-sheets of unique sequence participate in an extensive trimerization interface ( buried surface area: 6970 Å2 , predicted as stable by PISA [34] ) and connect to a linker composed of 2 anti-parallel β-strands and an α-helix that leads into the catalytic domain ( Fig 1A and S4 Fig ) . These C-terminal tandem domains assume an immunoglobulin-like fold typical for starch binding domains , and show close structural resemblance ( but low sequence identity ~ 18% ) to the N-terminal domains ( N1 and N2 ) of Anoxybacillus sp . LM18-11 pullulanase [35] , with N1 ( 10–88 ) corresponding to TarS C2 ( overall main chain rmsd of 1 . 0 Å over 25 atom pairs ) and N2 ( 110–186 ) corresponding to TarS C1 ( overall main chain rmsd of 1 . 2 Å over 36 atom pairs ) . Interestingly , the tandem domains of the two enzymes are inverted with respect to their corresponding catalytic domains , existing C-terminally in TarS and N-terminally in pullulanase . Nevertheless the order of the domains is maintained , with C1/N2 adjacent to the catalytic domain , followed by C2/N1 ( S4 Fig ) . The N2 domain is highly conserved among pullulanases and belongs to carbohydrate binding module ( CBM ) family CBM48 that typically binds pullulan and glycogen . The N1 domain is located at the highly variable N-terminus characteristic of pullulanases and is observed in complex with maltotriose and maltotetraose , classifying it to a novel CBM68 family [35] , supporting the possibility that the analogous TarS trimerization domain could participate in binding the teichoic acid glycopolymer . Furthermore , prominent basic grooves along the surface of the trimerization domain , and notably along the C1 domain leading into the active site ( Fig 1B and Fig 4C ) , suggest that both C1 and C2 domains likely participate in polyRboP binding . Another interesting feature is the close proximity ( 4 . 0 Å ) of methionines ( M521 and M532 ) from each of the 3 different monomers at the bottom surface of the trimerization interface , forming a hexameric methionine cluster that we hypothesize may be involved in promoting plasticity/adaptability of TarS during multivalent interactions with substrates or potential cell-wall partners ( Fig 1A ) . Altogether , TarS features a novel architecture with closely-associated trimerization domains connected by a linker region to three protruding catalytic domains that face away from each other ( Fig 3A ) . The physiological relevance of the trimeric architecture is supported by size exclusion chromatography-multiangle light scattering ( SECMALS ) with an elution profile that corresponds predominantly to a 200 kDa species ( theoretical monomer molecular weight is 66 kDa ) ( S5 Fig ) . Furthermore , structure-guided mutants aimed at disrupting the trimerization interface ( M521R , M532R ) ( Fig 1A ) resulted in the expression of insoluble protein aggregates . The structure of full-length TarS suggests motion between the catalytic domains and corresponding trimerization domains , where an overlay of trimer monomers reveals varying angles ( 50° vs . 64° vs . 78° as measured with the UCSF Chimera package [36] ) between the common plane of the trimerization domain and the individual planes of the catatlytic domains , with the hinge point centered on P352 ( Fig 3B and S1 Movie ) . The difference in angles is unlikely to have resulted from crystal packing artifacts , given that the catalytic domains in the full-length structure are not involved in any crystal contacts . We propose therefore that the apparent flexibility between the two domains may be related to substrate binding and/or processivity . Our analysis of the enzymatic mechanism of TarS was facilitated by the capture of various TarS1-349 structures encapsulating the catalytic domain in complex with metal and substrates . These include the native structure with bound Mn2+ ( free enzyme ) , the UDP-GlcNAc bound structure in the presence and absence of Mn2+ ( the binary Michaelis complex ) , and the UDP bound structure with Mn2+ ( product complex ) ( S2 Fig ) . The identity of the cation as Mn2+ is inferred from the observation that if Mn2+ is not specifically added during crystallization , the corresponding electron density is absent . Furthermore , inductively coupled plasma mass spectrometry ( ICP-MS ) studies revealed that TarS , purified in buffers without cation , lacked a bound metal , making it unlikely that another cation would have been carried through to crystallization . This data would further suggest that the cation does not play a structural role , but rather a functional role in the enzyme . The above structures show notable differences in two key loops , depending upon the presence or absence of the intact UDP-GlcNAc donor substrate . The first loop ( 171–178 ) , designated as the catalytic site ( CS ) loop , contains the putative base catalyst D178 and moves towards the catalytic center in the presence of the intact UDP-GlcNAc donor substrate ( Fig 2B ) . The second loop ( 205–215 ) , designated as the substrate access ( SA ) loop , is ordered only in the presence of UDP-GlcNAc and sterically occludes an otherwise open channel leading into the active site in the absence of the intact donor ( Fig 2C ) . Based on these observations it may be inferred that in the native structure , the SA loop is disordered allowing binding of UDP-GlcNAc , upon which the CS loop moves closer to the active site center and the SA loop becomes ordered , occluding the active site channel . This occlusion may serve to exclude water in order to decrease background hydrolysis , and/or to possibly guide the correct positioning of the acceptor polyRboP by providing a restricted passage for entry and facilitating binding . Both these loops contain several active site residues that are essential for catalysis as described below . Upon UDP-GlcNAc cleavage , where only UDP remains bound to the enzyme , the SA loop is once again disordered , likely allowing the release of the leaving group and subsequent binding of a new UDP-GlcNAc donor substrate . Similar adaptability is witnessed in active site proximal loops of many diverse GTA class glycosyltransferases , including as examples the GTPase glycosyltransferase TcdA [37] , the ABO ( H ) group A and B glycosyltransferases GTA and GTB [38] , and the maltosaccharide synthase glycogenin [39] , reflecting the plasticity required for binding and catalysis involving a wide range of donor/acceptor substrates . The signature DXD motif of TarS is believed to be critical for divalent cation coordination and catalysis . The two aspartates within this motif are D91 and D93 , situated adjacent to a third aspartate , D94 , whose role in catalysis was also investigated . Structural analysis revealed that Mn2+ is coordinated in a hexahedral manner , via a bivalent interaction with both side chain carboxylate oxygens of D93 , the UDP-GlcNAc/UDP diphosphates O1A and O1B , as well as two ordered water molecules . The R206 guanidinium group furthermore forms a hydrogen bond with the D93 carboxylate side chain and likely ensures its proper orientation . D94 has an indirect role in that its side chain carboxylate forms hydrogen bonds with the two Mn2+ coordinating waters . The D91 side chain carboxylate also forms a hydrogen bond with one of the waters that coordinates Mn2+ . This divalent metal and its coordinating waters , along with several active site residues , in turn contribute to binding of the UDP-GlcNAc donor substrate ( Fig 4B ) . Notably the side chain carboxylate oxygen of E177 , located in the SA loop , moves into position to form hydrogen bonds with the C4 and C6 GlcNAc hydroxyls ( Fig 4B and Fig 2B ) , the side chain imidazole of H210 and hydroxyl of S212 , located in the CS loop , form stabilizing interactions with the O1B and O2B phosphates when ordered upon UDP-GlcNAc binding , and the uracil ring of UDP-GlcNAc is stabilized by offset π-π interactions with the Y10 phenol side chain ( Fig 4B ) . Interestingly , no stabilizing interactions are observed with the N-acetyl group of GlcNAc , such that specificity may extend to similar sugars and may be of interest for inhibitor design . This observation is in agreement with the previous finding that TarS can use UDP-Glc as an alternative donor substrate , suggesting tolerance at the C2 position [5] . A key residue that appears to bridge the CS and SA loops is R75 , whose side chain guanidinium forms interactions with D91 and E177 carboxylates as well as the C4 hydroxyl of GlcNAc ( Fig 4B ) . Interestingly , TarS has also been shown to use UDP-GalNAc as a substrate , although far less efficiently than UDP-Glc or UDP-GlcNAc [5] . This suggests a lesser degree of tolerance at the C4 position , reflecting the interactions of E177 and R75 with the C4 GlcNAc hydroxyl . GTA inverting glycosyltransferases are believed to adopt a concerted SN2-like displacement mechanism for catalysis , as supported by structural studies [40–43] , hybrid quantum mechanical/molecular mechanical studies [44 , 45] , and kinetic isotope effect measurements [46 , 47] . The active site conformation of TarS appears to also adhere to an SN2-type reaction mechanism , according to which the teichoic acid acceptor molecule is activated by general base-catalysed abstraction of the polyRboP C4 hydroxyl group hydrogen in concert with nucleophilic attack at the β-face of the UDP-GlcNAc C1 anomeric centre ( Fig 4A ) . Departure of the leaving UDP is stabilized by the coordinating Mn2+ and leads to glycosyl transfer with inversion of the stereochemistry of the GlcNAc anomeric centre . UDP-GlcNAc itself is furthermore observed to assume a “tucked under” conformation ( as common in other glycosyltransferases [38 , 48–55] ) , where the GlcNAc sugar is tucked below the plane of the UDP diphosphates , allowing exposure of the scissile bond to nucleophilic attack ( Fig 4B ) . In TarS , D178 , situated 6 . 2 Å away on the β-face of the C1 anomeric carbon , is suitably positioned to act as a Brønsted base catalyst for the incoming acceptor ( Fig 4B ) , and shows good spatial agreement with the catalytic aspartate of SpsA [56] , a prototypical inverting GTA from the GT2 family ( sharing 28% identity with TarS ) , when the structures are superimposed ( overall main chain rmsd of 1 . 1 Å over 107 atom pairs ) ( S6A Fig ) . The Mn2+ that ligands D93 is also observed to closely superimpose with the equivalent D99 of the Mn2+ dependent SpsA . Indeed , a glycerol molecule trapped in the SpsA structure in proximity to the catalytic aspartate could , by analogy , provide a clue as to the positioning of the incoming polyRboP in the TarS structure ( S6B Fig ) . Trapped sulfates from our crystallization condition were also observed in the TarS1-349 structure along two basic grooves leading into the active site centre , potentially indicating where the phosphates of the acceptor polyRboP may be bound ( Fig 2A , Fig 4C ) . In support of this , we used AutoDock Vina to model a PRboPRboP molecule in the TarS active site using a highly exhaustive search protocol . The top 20 scoring poses reveal that the terminal phosphates of PRboPRboP overlap closely with the two sulfates situated in the basic grooves discussed above . A RboP C4 hydroxyl in the lowest energy pose was furthermore situated in close proximity to the catalytic D178 , and was oriented similarly to a glycerol hydroxyl superimposed from the SpsA structure ( Fig 4D ) . Brown et al . have previously shown TarS to be exclusive and highly specific for its polyRboP acceptor , having tested various alternate acceptor substrates including RboP , polyGroP-WTA , LTA , and CDP-ribitol [5] . Based on our model , it is possible that phosphate binding sites residing in basic grooves along the protein surface could determine the acceptor specificity by dictating the distance between adjacent phosphates , such that the shorter polyGroP unit chains may be incompatible with binding . Binding of multiple polyol phosphate units along the surface could also collectively increase the substrate binding affinity , allowing greater selectivity for the polymeric substrate over similar “monomer” substrates such as RboP and CDP-ribitol . To analyze the glycosyltransferase activity of TarS , polyRboP was isolated from the cell wall of S . aureus strain RN4220 and attached GlcNAcs were hydrolytically cleaved with α- and β-N-acetylglucosaminidases ( NAGLUs ) to liberate free acceptor sites . PolyRboP was then purified on a DEAE weak anion exchange column to remove NAGLUs and analyzed by ICP-MS to determine , according to the measured phosphate concentration , the molar concentration of constituent single RboP units in the WTA chain . TarS activity on this substrate was then analyzed by both direct HPLC-based and indirect fluorescence-based UDP detection methods . The HPLC method was used to test for the presence of activity , where a TSKgel DEAE-5PW weak anion exchange column was used to separate UDP-GlcNAc from released UDP upon donor hydrolysis . Using this method , the activity of wild-type TarS and its metal dependency was confirmed ( Fig 5A ) . We note that TarS displayed some promiscuity towards Mn2+ and Mg2+ , showing activity in the presence of both , with Mg2+ resulting in a higher level of UDP-GlcNAc hydrolysis . The presence of Ca2+ however did not result in observable activity ( Fig 5A ) . Several constructed active-site mutants were also tested for activity , along with a designed control ( D198A ) chosen distal from the catalytic center ( Fig 5B ) . Based on these results , mutations R75A , D91A , D93A , D94A , E177A , and H210A abolished activity as defined by UDP-GlcNAc hydrolysis , whereas mutations D178N , R206A , and S212A led to severe decreases in activity , validating the importance of these residues for catalysis , as discussed above . These mutants were also screened for thermostability , showing that , except for R75A and D94A , which were actually more stable than wild-type , the mutations did not affect overall protein stability ( Fig 5C ) . Analysis of the thermostability data also revealed that mutants E177A , D178N , H210A , and S212A , residing in flexible CS and SA loops , were stabilized in the presence of UDP-GlcNAc similar to the wild-type enzyme . This suggests that these residues may play only an ancillary role in initial substrate binding , consistent with the unfavorable positioning and disorder of respective CS and SA loops prior to UDP-GlcNAc binding . Mutants R75A , D91A , D93A , D94A , and R206A however showed no significant stabilization in the presence of UDP-GlcNAc , suggesting that these mutations are detrimental for UDP-GlcNAc binding and further emphasizing the roles of these residues in donor substrate interactions , as observed structurally ( Fig 4B ) . For kinetic measurements , continuous fluorescent monitoring of UDP release was achieved with the ADP Quest Assay kit ( Discover Rx ) , and kinetic parameters for both wild-type ( full-length ) and TarS1-349 ( lacking the trimerization domain ) constructs were determined ( Fig 5D ) . Although TarS has been previously shown unable to GlcNAcylate single RboP units [5] , we found that the reaction could be driven under high RboP concentrations . Therefore , the glycosyltransferase activity of TarS was further confirmed by mass-spectroscopic identification of the RboP-GlcNAc reaction product of UDP-GlcNAc and RboP ( S7 Fig ) . Kinetic parameters for wild-type and TarS1-349 constructs were similar , indicating that the trimerization domain appears to be dispensable for UDP-GlcNAc hydrolysis activity . Furthermore , for both constructs , the glycosyltransferase activity kcat in the presence of polyRboP was more than 10 fold greater than the hydrolysis activity in its absence . However , there were some small differences . For instance the Km ( UDP-GlcNAc; 21 ± 2 μM ) of the TarS1-349 hydrolysis reaction was somewhat lower than that of the wild-type ( 45 ± 3 μM ) . In addition , both the Km ( polyRboP; 2840 ± 140 μM ) ) and kcat ( 66 ± 2 min-1 ) of the TarS1-349 glycosyltransferase reaction were somewhat higher than those of the wild-type ( Km 1240 ± 70 μM; kcat 36 ± 1 min-1 ) ) . These differences may be associated with the trimerization domain’s increasing of polyRboP binding affinity and its possible effect on the relative orientation ( suggested structurally ) and local concentration of the linked catalytic domain that could influence the reaction . Nevertheless , the second order rate constants kcat/Km for both full-length and TarS1-349 glycosyltransferase reactions are very similar , suggesting the trimerization domain does not have a large role in catalysis , but is more likely involved in other events such as protein interactions and/or processivity . In order to examine whether TarS is a processive enzyme ( i . e . catalyzes multiple rounds of reactions before substrate dissociation [57] ) and the possible involvement of the trimerization domain in processivity , we measured association and dissociation rate constants ( kon and koff ) of wild-type and TarS1-349 using biolayer interferometry ( S8 Fig ) . Intrinsic processivity ( theoretical potential for processivity: Pintr ) was then calculated using the approximation Pintr ~ kcat /koff ( where kcat pertains to glycosyltransferase activity ) as previously described [58 , 59] ( Fig 5D ) . The results show that although TarS1-349 , lacking a trimerization domain , retains some processivity ( 133 ± 14 ) , its Pintr is reduced compared to wild-type ( 2400 ± 260 ) indicating that oligomerization does appear to contribute to TarS processivity . It is important to note that the koff measurements were nevertheless performed on a complicated system , where TarS itself is composed of multiple domains ( catalytic domain , linker , C1 , C2 ) that may take part in polyRboP acceptor binding , and that the acceptor itself is a heterogeneous polymer ranging in size from 9–11 kDa [31] . The additional C1 and C2 domains in the wild-type construct as such are likely to contribute to increased substrate avidity ( accumulated strength of multiple affinities ) compared to TarS1-349 that lacks these domains . The koff values therefore would be expected to represent a myriad of interactions , all or some of which contribute to enzyme processivity . Furthermore , the method used here to analyze enzyme processivity is simplistic and only provides probabilistic estimates that are useful for comparative purposes only . More accurate measurements would require significantly more complicated studies that would address the genuine processivity of an enzyme acting on a polymeric substrate [60] .
Gram-positive bacteria such as S . aureus invest an impressive amount of resources and energy in the synthesis of WTA , as these polymers play a pivotal role in bacterial biology and pathogenicity . In addition , the importance of accessory glycosyltransferases TarS and TarM in the decoration of S . aureus WTA polymers is becoming apparent and could provide valuable avenues for drug design . Here we have presented several structural snapshots of TarS that provide important insights into its catalytic mechanism . The structure of TarS represents a unique topology , where an inverting GTA class GT2 family [32] catalytic domain is linked to a trimerization domain composed of tandem CBMs . The catalytic domain of TarS reveals interesting structural features , such as a CS loop that positions towards the catalytic center and a SA loop that becomes ordered upon donor substrate binding . We propose that these rearrangements act to limit the level of background hydrolysis while still allowing access of the incoming UDP-GlcNAc donor and release of the UDP leaving group . The above rearrangements may also guide the correct route of entry and assist in the binding of the polyRboP acceptor substrate . Interestingly , when superimposed , the structure of UDP bound SpsA ( Bacillus subtilis ) resembles the UDP ( vs . UDP-GlcNAc ) bound TarS catalytic GTA domain with regard to the position of the CS loop and the disorder of the SA loop ( S6 Fig ) . As SpsA is an inverting GTA class GT2 family glycosyltransferase implicated in polysaccharide spore coat synthesis , and as most GT2 family members are also involved in polysaccharide synthesis or modification , the described rearrangements in the GTA domain may represent general mechanistic features of this family of enzymes acting on glycopolymer acceptor substrates . Similar loop rearrangements are observed in several other GTA families ( reviewed in [61 , 62] ) . The structure of the trimerization domain of TarS is novel in the teichoic acid pathway , in that it includes two sequence-unique tandem CBMs , of which there are currently 71 classified families [32] . These tandem CBMs were identified by their high structural homology to the N1 ( CBM68 ) and N2 ( CBM48 ) domains of Anoxybacillus sp . LM18-11 pullulanase [35] . Pullulanase is a debranching enzyme that hydrolyzes α-1 , 6 glycosidic linkages of α-glucan polysaccharides , and its N1 domain exhibits sugar binding activity , such that the homologous C2 domain of TarS may have similarly been adapted for polyRboP binding . Although the N2 domain in the pullulanase structure is not observed as a complex , there are examples of CBM48 family structures in complex with oligosaccharides , including the rice branching enzyme 1 [63] and the starch excess4 protein [64] , where the respective maltopentaose and maltoheptaose substrates interact with both the CBM48 and catalytic domains . Given the prominent electropositive charge distribution along the surface of the TarS C1 and C2 domains leading into the active site , it is likely that both CBM domains may be involved in polyRboP binding . Unlike pullulanase , the CBMs of TarS are involved in the formation of an extensive trimerization interface and interestingly , the truncation of both the TarS trimerization domain as well as the pullulanase N1 domain has only minimal effects on binding kinetics [35] . This suggests that these CBM domains are dispensable for catalytic activity and that they may assume other functions such as those involved in regulation and/or interaction . Indeed the TarS1-349 truncation mutant is capable of WTA GlcNAcylation albeit with a lower binding affinity for the polyRboP acceptor substrate ( Fig 5D ) . Despite its structural divergence with TarS , a similar observation was made with TarM , whose GlcNAcylation activity appeared to be independent of the trimeric state of the enzyme as observed with a confirmed trimerization-disrupting mutant [31] . Furthermore , the TarS1-349 truncation mutant displayed a lower level of intrinsic processivity relative to wild-type , indicating that the trimerization domain does appear to play a role in enzyme processivity ( Fig 5D ) . Another interesting observation is that although TarS and TarM are tasked with the similar undertaking of polyRboP GlcNAcylation , their trimerization domains differ dramatically in both structure as well as the extent of the trimerization interface , with TarM forming only sparse contacts . In fact , the trimerization domain of TarM belongs to pfam database [65] family domain of unidentified function ( DUF ) 1975 , present in the N-termini of various prokaryotic α-glycosyltransferases . TarM in addition shows close structural resemblance to Streptococcus pneumoniae GtfA , a GTB O-GlcNAc transferase that is involved in the glycosylation of serine rich repeat adhesion proteins [66] . The DUF1975 domain of GtfA has been crystallographically observed to interact with a similar extended β-sheet domain of its GtfB co-activator , forming a hetero-tetramer [67] . As such , the vast difference in the derivation of the TarS and TarM trimerization domains supports the notion of separate evolutionary lineages and , along with other differences , elicits curiosity regarding the respective biological roles of these distant yet seemingly redundant enzymes . An exciting study has recently reported the discovery of a novel antibody-antibiotic conjugate ( AAC ) for targeted killing of intracellular MRSA , which is strictly activated after the release of the antibiotic , rifalogue , in the proteolytic environment of the phagosome [68] . Interestingly , in designing the antibody-antibiotic conjugate , the authors found that an antibody that recognized the TarS-mediated β-O-GlcNAc WTA modification bound to all tested S . aureus strains , whereas the antibody recognizing the α-O-GlcNAc modification did not ( presumably due to the absence of the α-O-linkage in specific S . aureus isolates ) [68] . With regard to in vivo derived MRSA strains that co-produce the β- and α-O-linked GlcNAc WTA , antibodies specific for β-O-GlcNAc consistently yielded greater binding , suggesting that the TarS-mediated β-O-GlcNAc modification is either more immunogenic or abundant in MRSA pathogens . In support of this notion , it has also been found that human sera predominantly contain antibodies against β-O-GlcNAcylated WTA during infection , possibly highlighting the role of TarS in pathogenesis [69] . TarS rather than TarM has furthermore been shown to be key in MRSA β-lactam resistance [5] , where β-O-GlcNAc is proposed to act as a possible scaffold for the recruitment of the β-lactam insensitive PBP2a , based on in vitro WTA binding of PBP2a [29] . Interestingly , in contrast to TarS , the TarM gene resides chromosomally outside of the WTA gene cluster , and its phylogenic distribution in Staphylococci suggests its acquisition by an ancient horizontal gene transfer ( HGT ) event [70] . In addition , whereas nearly all S . aureus strains contain the gene for TarS , several strains appear to have lost the TarM gene during evolution [70] . The retention of TarM in the genomes of a wide variety of S . aureus strains nevertheless suggests an advantage for its bacterial hosts . A recent article has shown that whereas TarS mediated β-O-GlcNAcylation of polymers facilitates susceptibility to infection by Podoviridae “lytic” phages , the TarM mediated α-O-GlcNAcylation prevents infection [70] , thereby conferring a level of protection that likely evolved in Podoviridae-rich environments . The authors also report that S . aureus strains that express both TarS and TarM display preferential α-O-GlcNAcylation of polyRboP , leading them to suggest the possibility of TarM having a greater activity [70] . In accordance with this hypothesis , we found that TarM does appear to be a more catalytically efficient enzyme in vitro , possessing a ~10 fold greater second order rate constant kcat/Km [31] . Redundancy in WTA glycosyltransferases TarS and TarM may also be beneficial for phage-mediated HGT events that drive S . aureus evolution , where a recent study has demonstrated the binding of “helper” phages to O-GlcNAc residues of S . aureus WTA , regardless of linkage conformation [71] . This redundancy in addition appears to be relevant in other processes such as colonization , where it has recently been found that WTA O-GlcNAcylation with either an α or β linkage is pivotal to the attachment of MRSA strains to human nasal epithelial cells [72] , a process that contributes to the spread of nosocomial and community-acquired infections . Although WTA is involved in a vast number of pathological processes in S . aureus , it has been shown that its absence nevertheless results in a viable phenotype . Bacteria lacking WTA are however greatly compromised in their ability to colonize and infect , and display dysregulated cell-division [5 , 10 , 73] . The deletion of TarS , TarM , or both , resulting in non-GlcNAcylated WTA , however leads to no detectable morphological abnormalities [5] . Therefore disruption of TarS in MRSA strains ( especially those also lacking TarM ) would interfere with host colonization , effectively disarming the pathogen without creating a strong selective pressure that often results in drug resistance . Furthermore , in MRSA strains where TarS is involved in methicillin resistance , inhibition of TarS in combination therapies would allow re-sensitization of bacteria to β-lactam antibiotics [5] . Indeed the success of β-lactam and β-lactamase inhibitor combinations in overcoming resistance attests to the effectiveness of such therapeutic strategies [74] . On another front , as TarS and TarM mediated WTA O-GlcNAcylation influences phage binding , the study of these enzymes will likely prove important in the re-emerging field of phage therapy [75] . The structure and mechanism of TarS therefore provide a valuable platform for rational therapeutic design in the treatment of MRSA , which remains a leading source of drug-resistant and life-threatening infections world-wide .
The full-length open reading frame ( amino acids 1–573 ) encoding S . aureus TarS ( SAV0258 ) was cloned into the expression vector pET41b without an affinity tag . Mutagenic TarS constructs were produced with the Quick Change mutagenesis kit ( Qiagen ) . Constructs were transformed into Rosetta ( DE3 ) Escherichia coli . The TarS truncation mutant ( TarS1-349 ) was cloned into the expression vector pET41b with a C-terminal 6x His-tag . Protein expression was carried out overnight at 30°C . Cells were grown in Luria Bertani broth ( supplemented with 35 μg/mL kanamycin ) to an optical density ( 600 nm ) of 0 . 6–0 . 8 , at which point Isopropyl β-D-1-thiogalactopyranoside ( IPTG ) was added at a final concentration of 1 mM . Cells were pelleted and stored at -80°C until required . For purification of full-length TarS , cell pellets were resuspended in buffer A ( 20 mM Hepes pH 7 . 3 , 300 mM NaCl , 5% glycerol ) . A complete protease inhibitor tablet at 1x final concentration ( Roche ) and DNAse 1 at 1 μg/mL final concentration were added and cells were lysed at 12 , 000 psi using a French press ( Thermo Electron Corporation ) . Cell debris was pelleted by centrifugation at 20 , 000 x g for 30 min . The resulting supernatant was loaded onto a 5 mL Heparin HP cartridge ( GE Lifesciences ) and eluted over a 30 mL linear gradient to 100% buffer B ( 20 mM Hepes pH 7 . 3 , 2 M NaCl , 5% glycerol ) . Fractions containing the purest protein were pooled , concentrated and loaded on a Superdex 200 column ( GE Lifesciences ) equilibrated in buffer C ( 20 mM Hepes pH 7 . 5 , 500 mM NaCl , 5% glycerol ) and the fractions collected and concentrated . For purification of the TarS truncation mutant ( TarS1-349 ) , the cell lysate was produced as above ( but resuspended in buffer D: 20 mM NaPO4 pH 7 . 3 , 5% glycerol ) , was loaded onto a 1 mL HisTrap HP cartridge ( GE Lifesciences ) and eluted over a 30 mL linear gradient to 100% buffer E ( 20 mM NaPO4 pH 7 . 3 , 500 mM imidazole , 5% glycerol ) . Fractions containing the purest protein were pooled , concentrated and loaded on a Superdex 200 column ( GE Lifesciences ) equilibrated in buffer C and the fractions collected and concentrated . The protein was frozen in liquid N2 and stored at -80°C until required . The metal content of TarS was measured using an inductively coupled plasma mass spectrometer ( NexION 300D ICP-MS , PerkinElmer Life Sciences ) and the data analyzed with NexION software . A calibration standard ( CAT# IV-STOCK-4 , Inorganic Ventures ) containing metals of interest ( Mg2+ , Mn2+ , Co2+ , Ni2+ , Cu2+ , Zn2+ ) was diluted with an internal standard solution containing 10 μg/L Sc and 1% nitric acid ( CAT# IV-ICPMS-71D , Inorganic Ventures ) , and this was used to generate standard curves spanning 1 to 100 μg/L for each metal . Protein samples were appropriately diluted with internal standard solution to adjust metal concentrations within the range of the standard curve . To confirm the absence of cation , the protein sample was spiked with metals of interest and measured as a positive control . S . aureus WTA was isolated and purified according to modifications of previously established protocols [76 , 77] . S . aureus RN4220 cells were grown in a culture of 20 mL TSB overnight at 37°C and the cells collected at 2 , 000 x g for 10 min . The cells were washed once in 30 mL of buffer 1 ( 50 mM MES , pH 6 . 5 ) , resuspended in buffer 2 ( 4% SDS , 50 mM MES , pH 6 . 5 ) , and boiled in a water bath for 1 hr . The cell debris was collected at 10 , 000 x g for 10 min , resuspended in 2 mL of buffer 2 , and sedimented at 14 , 000 x g for 10 min . The pellet was washed in subsequent 1 mL volumes of buffer 2 , buffer 3 ( 2% NaCl , 50 mM MES , pH 6 . 5 ) , and buffer 1 . The pellet was resuspended in 1 mL of buffer containing proteinase K ( 20 mM TrisHCl , pH 8 . 0 , 0 . 5% SDS , 20 ug proteinase K ) and digested at 50°C for 4 hr . The sample was pelleted at 14 , 000 x g for 10 min and washed once in buffer 3 and three times with distilled H2O . The sample was then resuspended in 1 mL of 0 . 1 M NaOH and shaken at room temperature for 16 hr . The remaining insoluble cell wall debris was removed by centrifugation at 14 , 000 x g for 10 min and the supernatant containing the hydrolyzed crude WTA was neutralized with addition of HCl to a final concentration of 0 . 1 M . The sample was dialyzed against distilled H2O using a 3 MWCO membrane . For digestion of attached GlcNAc , the sample was exchanged into buffer ( 100 mM Sodium Citrate pH 4 . 5 , 250 mM NaCl ) by dialysis using a 3 MWCO membrane and incubated with 0 . 5 mg/mL α-N-acetylglucosaminidase ( R&D systems ) and 0 . 2 mg/mL β-N-acetylglucosaminidase ( New England Biolabs ) overnight at 37°C . For purification , the sample was exchanged into buffer A ( 20 mM Tris HCl pH 7 . 2 ) by dialysis using a 3 MWCO membrane , applied on a 5 mL DEAE FF cartridge ( GE Lifesciences ) and eluted over a 30 mL linear gradient to 100% buffer B ( 20 mM Tris-HCl pH7 . 2 , 1 M NaCl ) , with UV monitored at 205 nm . Peaks with the highest 205 nm readings were pooled and dialyzed against distilled H2O using a 3 MWCO membrane . The sample was frozen , lyophilized , and resuspended in distilled H2O . WTA concentration was measured according to the concentration of phosphorus detected by ICP-MS , whereby a phosphorus calibration curve spanning 1 to 100 μg/L was created using a phosphate standard solution ( Sigma ) diluted with internal standard solution ( see metal binding analysis section ) . TarS activity was studied using the ADP Quest Assay kit ( DiscoverRx , USA ) according to the manufacturer’s protocol and performed in 10 μL volume 384-well black assay plates . Various concentrations of wild-type and TarS1-349 were incubated with 1 mM UDP-GlcNAc and assay kit reagents to determine the optimal concentration of proteins for assay . Upon cleavage of UDP-GlcNAc by TarS , UDP is released resulting in a fluorescence signal monitored continuously at 530 nm excitation and 590 nm emission wavelengths using a Synergy H4 multi-mode plate reader ( BioTek , USA ) . Km and kcat values were determined by using optimal protein concentrations and varying concentrations of UDP-GlcNAc and WTA , and concentration units were obtained using UDP for the standard curve . TarS activity was further verified directly by chromatography . Here , 10 μM TarS was incubated overnight with 1 mM UDP-GlcNAc , after which the reaction mixture was filtered through a 3 KDa MWCO filtration unit to remove protein . 10 ul fractions of the filtrate were injected onto a TSKgel DEAE-5PW weak anion exchange column ( TOSOH Biosciences , USA ) and the separated UDP-GlcNAc and UDP peak areas were monitored and quantified at a UV wavelength of 254 nm using an HPLC system . TarS thermostability was measured as a function of temperature dependent aggregation by differential static light scattering ( StarGazer-2; Harbinger Biotechnology and Engineering Corporation ) according to the method of Vedadi et al . [78] . Briefly , 10 μl of 10 μM protein under selected conditions was heated from 25–85°C at a rate of 1°C/min in individual wells of a clear-bottom 384 well plate ( Nunc , Rochester , NY ) . Protein aggregation , as a measure of the intensity of scattered light , was scanned every 30 s with a CCD camera . The integrated intensities were plotted against temperature using a Boltzman regression , where the inflection point of each fitted curve was defined as the aggregation temperature , Tagg . Crystals of TarS217-573 were obtained by proteolytic cleavage of ~10 mg/ml of the full-length protein with thermolysin ( 1 μg/ml final concentration ) and immediate setup by vapor diffusion using a reservoir solution of 1 M imidazole ( pH 7 ) . Crystals belonged to space group P32 with unit cell dimensions a = b = 105 . 66 Å , c = 80 . 48 Å . For phasing , crystals were soaked in sodium iodide and a single wavelength SAD experiment was carried out . Data were processed with XDS [79] and Aimless [80] . Phasing was carried out with SHARP [81] using SHELX [82] to determine the heavy atom substructure . Model building was performed with Buccaneer [83] and refined using Refmac [84] , Phenix [85] and Coot [86] using TLS parameters in the later stages . Refinement was finished using native data to higher resolution ( see Table 1 ) . The final model with one molecule in the asymmetric unit has good stereochemistry with 97% of residues in the favoured region of the Ramachandran plot and 0% outliers . Crystals of TarS1-349 ( ~20 mg/mL ) were obtained by sitting-drop vapor diffusion in the presence ( or absence ) of 15 mM UDP-GlcNAc and 2 mM MnCl2 using a reservoir solution of 0 . 2 mM lithium sulfate , 27% w/v PEG 3350 , and 0 . 1 M Bis-Tris pH 5 . 5 . Crystals were cryo-protected with a reservoir solution containing 30% glycerol and 50 mM UDP-GlcNAc when required and were subsequently flashed cooled . Crystals belonged to space groups C2 ( native ) , P1 ( +UDP-GlcNAc:Mn , +UDP-GlcNAc ) or P21 ( +UDP ) with the same general molecular packing ( see Table 1 ) . The structures were solved with molecular replacement using Phaser [87] , initially using the overlapping region of the TarS217-573 structure as a search model and refined as above . The final models have good stereochemistry: the native structure has one monomer in the asymmetric unit with 97% of residues in the favoured region of the Ramachandran plot and 0 . 6% outliers; the UDP-GlcNAc:Mn structure has two monomers in the asymmetric unit with 97% of residues in the favoured region of the Ramachandran plot and 1 . 4% outliers; the UDP structure has one monomer in the asymmetric unit with 97% of residues in the favoured region of the Ramachandran plot and 0 . 3% outliers; and the UDP-GlcNAc structure has two monomers in the asymmetric unit with 98% of residues in the favoured region of the Ramachandran plot and 0 . 29% outliers . Full-length TarS ( ~20 mg/mL ) was crystallized by microbatch in the presence of 15 mM UDP-GlcNAc and 2 mM MnCl2 using a reservoir solution of 0 . 2 M ammonium chloride , 15% w/v PEG 6000 , 0 . 15 M Tricine ( pH 8 . 0 ) , and 0 . 4 M NDSB 195 . Crystals were cryo-protected with a reservoir solution containing 30% glycerol and 15 mM UDP-GlcNAc and were subsequently flashed cooled . Crystals belonged to space group P21 with unit cell dimensions a = 71 . 85 Å , b = 191 . 47 Å , c = 97 . 03 Å , β = 109 . 5° . Crystals diffracted with strong anisotropy ( accounting for poor dataset statistics ) and the data were cut to 4 . 0 Å . The structure was solved with molecular replacement using Phaser [87] with TarS1-349 and TarS217-573 structures as search models . The structure was refined with Phenix [85] using the high-resolution structures as restraints . The final model with three molecules in the asymmetric unit has good stereochemistry with 96% of residues in the favoured region of the Ramachandran plot and 0 . 24% outliers . Data and coordinates have been deposited to the RCSB Protein Data Bank with accession codes 5TZ8 , 5U02 , 5TZI , 5TZK , 5TZE and 5TZJ . Purified protein was applied to a Superdex 200 HR 10/30 column ( GE Healthcare ) equilibrated in 30 mM Hepes pH 7 . 3 , 500 mM NaCl , 5% glycerol buffer , using an Agilent 1100 series HPLC ( Agilent Technologies ) , coupled in-line to a Dawn Heleos II 18-angle MALS light scattering detector , and Optilab T-rEX differential refractometer monitor ( Wyatt Technology ) . Monomeric bovine serum albumin ( Sigma-Aldrich ) was used to normalize the light scattering detectors . Data were collected and analyzed with the Astra 6 software package provided by the manufacturer , Wyatt Technology . The protein molar mass was calculated , assuming a refractive index increment ( dn/dc ) value of 0 . 186 ml g−1 . Reactions in the presence of 50 μM TarS , 8 mM UDP-GlcNAc and ± 8 mM RboP proceeded overnight at room temperature in buffer consisting of 20 mM Tris pH 8 , 500 mM NaCl , and 2 mM MnCl2 . Reactions were subsequently filtered through a 3 KDa MWCO filtration unit to remove protein . In order to remove buffer and salts for LC/MS analysis , samples were diluted 50 fold in distilled H2O , applied to a HiTrap Q XL 1mL cartridge and eluted with 200 mM ammonium formate pH 8 . The eluted fraction was extensively diluted with water and repeatedly lyophilized to reduce salt concentration . The lyophilized sample was dissolved in water or water/MeOH ( 1:1 ) and dilution series were subjected to LC/MS . Spectra were recorded on a Waters ZQ2000 LC/MS attached to a Waters 2695 separation module with flow injection analysis in negative mode using electrospray ionization . Spectra were analyzed with Masslynx 4 . 0 software . Biolayer interferometry was performed using an Octet Red instrument ( FortéBio Inc . ) with streptavidin sensors ( FortéBio Inc . ) . TarS was biotinylated using EZ-Link NHS-PEG4-Biotin ( Thermo Scientific , USA ) . Biolayer interferometry was performed at 25°C in a 96-well plate ( Greiner Bio-One ) and a 200 μL well volume . After a brief equilibration of the sensors in assay buffer ( 20 mM Hepes pH 7 , 500 mM NaCl ) , full-length or TarS1-349 was loaded onto sensors for 5 minutes at 300 nM followed by the blocking of unbound streptavidin with 15 μg/mL EZ-Link Biocytin ( Thermo Scientific ) in Superblock Blocking Buffer ( Thermo Scientific ) . Next , a baseline was acquired for 3 minutes followed by the association of TarS for 5 minutes ( kon ) and dissociation for 15 minutes ( koff ) in assay buffer ( 20 mM Hepes pH 7 , 500 mM NaCl ) . Various optimal concentrations of polyRboP ( 0 . 31 mM , 0 . 62 mM , 1 . 25 mM , 2 . 5 mM and 5 mM ) were titrated with double referencing to rule out non-specific binding to sensors , and the KD was calculated based on kon and koff rates fitted to a heterogeneous ligand model using the FortéBio data analysis software .
|
Historically , β-lactam class antibiotics such as methicillin have been very successful in the treatment of bacterial infections , effectively destroying bacteria by rupturing their cell walls while posing little harm to the human organism . In recent years , however , the alarming emergence of Methicillin Resistant S . aureus or MRSA has resulted in a world-wide health crisis , calling on new strategies to combat pathogenesis and antibiotic resistance . As such , understanding the pathways and players that orchestrate resistance is important for overcoming these mechanisms and restoring our powerful β-lactam antibiotic arsenal . In this article we describe the crystal structure of TarS , an enzyme responsible for the glycosylation of wall teichoic acid polymers of the S . aureus cell wall , a process that has been shown to be specifically responsible for methicillin resistance in MRSA . TarS is therefore a promising drug target whose inhibition in combinational therapies would result in MRSA re-sensitization to β-lactam antibiotics . Here we present the first structure of TarS together with several snap-shots of its substrate/product complexes , and elucidate important catalytic features that are valuable for rational drug design efforts to combat resistance in MRSA .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"crystal",
"structure",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"pathogens",
"condensed",
"matter",
"physics",
"microbiology",
"enzymology",
"staphylococcus",
"aureus",
"petroleum",
"products",
"methicillin-resistant",
"staphylococcus",
"aureus",
"antibiotic",
"resistance",
"materials",
"science",
"pharmacology",
"crystallography",
"bacteria",
"bacterial",
"pathogens",
"antimicrobial",
"resistance",
"solid",
"state",
"physics",
"staphylococcus",
"medical",
"microbiology",
"tar",
"proteins",
"microbial",
"pathogens",
"transferases",
"petroleum",
"glycosyltransferases",
"physics",
"biochemistry",
"enzyme",
"structure",
"organic",
"materials",
"protein",
"domains",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] |
2016
|
Structure and Mechanism of Staphylococcus aureus TarS, the Wall Teichoic Acid β-glycosyltransferase Involved in Methicillin Resistance
|
Cholera burden in Africa remains unknown , often because of weak national surveillance systems . We analyzed data from the African Cholera Surveillance Network ( www . africhol . org ) . During June 2011–December 2013 , we conducted enhanced surveillance in seven zones and four outbreak sites in Togo , the Democratic Republic of Congo ( DRC ) , Guinea , Uganda , Mozambique and Cote d’Ivoire . All health facilities treating cholera cases were included . Cholera incidences were calculated using culture-confirmed cholera cases and culture-confirmed cholera cases corrected for lack of culture testing usually due to overwhelmed health systems and imperfect test sensitivity . Of 13 , 377 reported suspected cases , 34% occurred in Conakry , Guinea , 47% in Goma , DRC , and 19% in the remaining sites . From 0–40% of suspected cases were aged under five years and from 0 . 3–86% had rice water stools . Within surveillance zones , 0–37% of suspected cases had confirmed cholera compared to 27–38% during outbreaks . Annual confirmed incidence per 10 , 000 population was <0 . 5 in surveillance zones , except Goma where it was 4 . 6 . Goma and Conakry had corrected incidences of 20 . 2 and 5 . 8 respectively , while the other zones a median of 0 . 3 . During outbreaks , corrected incidence varied from 2 . 6 to 13 . 0 . Case fatality ratios ranged from 0–10% ( median , 1% ) by country . Across different African epidemiological contexts , substantial variation occurred in cholera incidence , age distribution , clinical presentation , culture confirmation , and testing frequency . These results can help guide preventive activities , including vaccine use .
Although cholera has disappeared as a public-health problem in developed countries , it remains a major concern in sub-Sahara Africa [1 , 2] . From 2007 to 2012 , at least 20 African countries reported more than 100 , 000 cases of cholera ( World Health Organization ( WHO ) weekly epidemiological records , 2007–2012 ) . However , this surveillance has weaknesses . Reporting is non-exhaustive for various reasons such as individual and community fears of stigmatization and economic loss . Reporting from district to national levels may be delayed or incomplete . According to WHO , only 3% to 5% of all cases are laboratory confirmed [3] . A variety of case definitions are used across countries , which could lead to cholera over or under-reporting . Finally , few countries have implemented case-based surveillance , with information at national level provided in the form of weekly summaries limited to cumulative case numbers and deaths [1] . Since the Haiti epidemic during 2010 , public and political attention on cholera has increased . Recently , WHO has prequalified a two-dose oral cholera vaccine ( OCV ) that is less expensive and less cumbersome to deliver than its predecessor . This , and the creation by WHO of a cholera vaccine stockpile for epidemic and potentially endemic cholera prevention , have stimulated interest in more timely , accurate , and comprehensive disease burden data from affected countries . The African Cholera Surveillance Network ( Africhol ) was launched in 2009 . Originally implemented in eight of the most affected sub-Saharan African countries , it has since expanded to three additional countries . Its primary aim is to better define cholera burden , geographic distribution , seasonal patterns , and risk groups to inform prevention strategies , including immunization . We present here incidence results and the associated case fatality ratio from eleven geographical areas located in the six Africhol countries having the strongest performing surveillance systems .
Starting in 2011 , we implemented population-based cholera surveillance in all cholera treatment facilities in a given geographic zone chosen in collaboration with ministries of health ( MoHs ) . Criteria for zone selection included: yearly occurrence of outbreaks or sporadic cholera cases; existence of dedicated diarrhea or cholera treatment facilities; and laboratory capacity for cholera confirmation by stool culture . In these zones , all health facilities providing treatment for severe cholera cases were included in surveillance . We also conducted a prospective surveillance in several outbreak sites outside of surveillance zones when these were reported to the MoH and when they had adequate laboratory facilities available . This was conducted the time of the epidemic . Patients were followed in all the cholera treatment facilities of a given surveillance area . In areas without known ongoing cholera , a suspected cholera case was defined as a patient aged two years or more that developed severe dehydration or died from acute watery diarrhoea . In areas with known cholera , a suspected case was defined as a patient aged two years or more that developed acute watery diarrhoea , with or without vomiting . A confirmed case was defined as a suspect cholera having a stool culture positive for Vibrio cholerae . Eight enhanced surveillance zones located in areas of known recent cholera occurrence were included in the analysis . Their location and starting dates were as follows: 1 ) Togo: five districts of Lome and Golfe district , Jun 2011; 2 ) Togo: Lake district in the Maritime region , Jun 2011; 3 ) Democratic Republic of Congo ( DRC ) : Goma and Karisimbi districts , Aug 2011; 4 ) Guinea: five districts of Conakry , Jul 2011; 5 ) Uganda: Manafwa , Mbale , and Butaleja districts , Dec 2011; 6 ) Mozambique: Beira city , Aug 2011; 7 ) Cote d’Ivoire: one district of Abidjan , Koumassi–Port Bouet–Vridi district ( KPBV ) , Jun 2012 . While data collection is currently ongoing , here we include only surveillance data collected through Dec 31st , 2013 . In addition to surveillance zones , we included data collected during outbreaks in Kasese district , Uganda ( Oct 2011–Dec 2012 ) ; Pemba city , Mozambique ( Jan 2013–Dec 2013 ) ; Adiake prefecture , Cote d’Ivoire ( May–Oct 2012 ) ; and three districts of Kinshasa ( Maluku , Kingabwa , and Massina districts ) , DRC ( Jul 2011–Feb 2012 ) . Within specifically defined study zones , we included all health care facilities known to treat cholera cases , including long-term facilities as well as newly established cholera treatment centers ( in Africa , these centers frequently are opened only in response to an outbreak ) . While all cholera cases were supposed to have been referred to a designated cholera treatment center , it is likely that private health centers conducted unauthorized evaluation and treatment . Included centers were the following: 1 ) Conakry , Guinea . The infectious disease and paediatric departments of Donka hospital . The additional cholera treatment center ( CTC ) in the Ratoma neighbourhood opened during the 2012 epidemic was also included; 2 ) Lome , Togo . The infectious disease and paediatric departments of the Centre Hospitalier Universitaire , Be Hospital , and other district health centres in which a temporary cholera treatment center was opened; 3 ) Lake District , Togo . The infectious disease and paediatric departments of Aneho Hospital and health centres with temporary treatment centers; 4 ) Goma-Karisimbi district , DRC: The cholera treatment centers located in the General Provincial Hospital , the Buhimba cholera treatment and the Kiziba temporary cholera treatment unit; 5 ) Maluku-Kingabwa-Massina district , Kinshasa , DRC: cholera treatment centers of Kingabwa and Malaku and the cholera treatment unit of Massina; 6 ) Abidjan , Koumassi-Port Bouet , Vridi District , Cote d’Ivoire . The infectious disease and paediatric departments of Port Bouet and Koumassi Hospitals and the temporary cholera treatment center at the Vridi Health Centre; 7 ) Adiake prefecture , Cote d’Ivoire: Adiake general hospital and temporary treatment centers; 8 ) Mbale-Manafwa-Buteleja district , Uganda: Nabiganda health center , Namatela health center and Busiu health center; 9 ) Kasese district , Uganda: Bwera hospital , Kayangi health center , Kagando hospital , Kinyamaseke health center and Kitholhu health center and other temporary treatment centers; 10 ) Beira , Mozambique: Ponta-Gea health center , Macurrungo health center , Munhava health center , Macurrungo and the central hospital of Beira; 11 ) Pemba city , Mozambique: the temporary cholera treatment center of Pemba city . In the enhanced surveillance zones and outbreak sites , the MoH teams collected data at health centers level using the same standardized data collection forms , which included sex , age , location , date of symptoms , culture results but also clinical information such as watery diarrhea , rice water stool , vomiting , dehydration . We identified all deaths among patients admitted to a cholera treatment facility . We did not include deaths occurring in the community or after treatment center discharge . In parallel , the MoH continued to register the overall number of suspected cases in their routine surveillance system using line lists with a limited number of variables ( date of onset , district , age and sex ) . We used district–level population estimates for 2011 or 2012 that corresponded to the geographic area under surveillance . The 2011 and 2012 population estimates were derived from the last census data ( Uganda , 2002; DRC , 1983; Togo , 2009; Guinea , 1996; Cote d’Ivoire , 1998; Mozambique , 2007 ) , updated each year by district health officers based on estimated national annual population growth rates . National public health laboratories in each country performed culture confirmation of suspected cases . Cholera polymerase chain reaction ( PCR ) testing was not available in any of the included countries . We aimed to collect whole stool or rectal swabs from all suspected cases . In practice , the proportion of cases with a collected stool varied according to context . During large outbreaks when laboratory capacity could become overwhelmed , local staff were advised to collect the first ten cases per day only . Samples were transported in Cary-Blair transport medium to the country national reference laboratories . There , they were enriched in alkaline peptone water and plated on thiosulfate-citrate-bile-salt-sucrose ( TCBS ) agar . Characteristic yellow colonies were sub-cultured in non-selective medium . Resulting colonies were tested for oxidase and , if positive , considered confirmed and serogrouped . External quality control was performed by the National Institute of Communicable Diseases in South Africa using PCR . We adopted the definition of rainy season from the World Bank climate portal ( sdwebx . worldbank . org/climateportal; accessed 2013 ) as follows: Uganda , Mar–Jun and Sept–Nov; Goma , DRC , Jan–May and Sept–Dec; Kinshasa , DRC , Jan–May and Oct–December; Mozambique , Oct–Mar; Cote d’Ivoire , May–Jun and Oct–Nov; Guinea ( Maritime region ) , May–Nov; Togo ( Maritime region ) , Apr–Jul and Sept–Nov . Suspected and confirmed cholera cases were summed by age group , sex , occurrence during the rainy season and clinical symptoms . We calculated the crude and corrected incidence rates for confirmed cases . Correction was done as follows: 1 ) for lack of culture testing , we extrapolated the proportion of culture positive results among suspect cases tested by culture to all notified suspect cases in each geographical area; 2 ) because culture has a sensitivity of 66% ( compared to combined results from culture , dipstick , direct fluorescent antibody , multiplex-PCR and Vibrio cholerae O1 El Tor specific lytic phage on plaque assay as gold standard ) for imperfect reported culture testing , we extrapolated the number of cultures that would have been positive if culture had a sensitivity of 100% [3] . For point 2 , we conducted a literature search and identified few studies that reported culture sensitivity relative to another gold standard , as culture itself has been the gold standard for years . Consequently the study by Alam et al . was used as an approximation , recognizing that the included data may not be definitive . For calculation of case fatality ratios ( CFR ) , we included in the denominator patients admitted to a cholera treatment center with cholera symptoms and as the numerator all deaths that were identified at the treatment center Comparisons between groups were performed using Pearson’s chi-square test . Graphs were produced with R open-access software . Statistical analyses were performed using STATA software ( version 12 . 1 , College Station , Texas 77845 USA ) . Africhol provided technical and financial resources to national MoHs to support cholera surveillance . Cholera is part of the national public health surveillance through the integrated disease surveillance and response system supported by WHO . The Africhol protocol was approved and implemented by the MoH of each country . The Togolese government further elected to submit the protocol for approval to a local Togolese institutional review board ( IRB ) . The remaining countries did not seek IRB approval as they considered that they were conducting epidemic disease surveillance and response . covered by national public health laws as an integral part of the public health mandate of the MoH and associated executing agencies .
From June 2011 to December 2013 , 13 , 377 suspect cholera case were notified: 47% ( 6343 ) occurred in surveillance zones in Goma , DRC and 34% ( 4585 ) in Conakry , Guinea ( Table 1 ) . We tested 26% ( 3536 ) of all suspected cases by culture , a figure that increased to 49% when excluding zones in Goma and Conakry , which both experienced large outbreaks in August 2012 and which respectively had testing only 7 . 4% and 0 . 5% of cases during this period ( Fig 1 and Fig 2 ) . In the surveillance zones , a median of 31% of cases were culture positive ranging from 37% in Conakry , Guinea to 0% in Beira , Mozambique ( Table 1 ) . With the exception of Adiake prefecture in Cote d’Ivoire , suspected cases were equally distributed by sex ( Table 2 ) . However , confirmed cases were more likely to be male ( Table 3 ) . The proportion of suspected cases aged under five years ranged from zero percent in surveillance zones in Abidjan , Cote d’Ivoire to 40% in Beira , Mozambique ( Table 2 ) ; for confirmed cases , the proportion aged under five years peaked at 29% in Goma , DRC . From 45–99% of suspected and 70–100% of confirmed cases occurred during the rainy season ( Tables 2 and 3 ) . The monthly distribution of cases in Goma-Karisimbi districts ( DRC ) , Mbale-Manafwa-Butaleja districts ( Uganda ) , Lome and Golfe districts ( Togo ) , Kasese district ( Uganda ) and Maluku-Kingabwa-Massina districts ( Kinshasa , DRC ) showed that cases with Vibrio cholerae identified by culture can be observed before the rainy season starts ( Figs 1 and 2 ) . The mean proportion of persons presenting with watery diarrhea at each site was 91% ( SD 7% ) and 82% ( SD 16% ) had vomiting . The percentage presenting with rice water stool varied from <1% to 86% and with dehydration from 33% to 99% ( Table 4 ) . We identified three epidemiological patterns ( Figs 1 and 2 ) . In surveillance zones in Goma ( DRC ) , confirmed cases were seen continuously throughout the surveillance period . In zones in Lome ( Togo ) , Mbale ( Uganda ) and Conakry ( Guinea ) , there were sporadic confirmed cases plus additional outbreaks at irregular intervals . Lastly , in Beira , Mozambique and Abidjan , Cote d’Ivoire , there was a history of recurrent cholera epidemics in the period leading up to Africhol implementation but as of the end of 2013 , no confirmed cases had been identified for 30 months and 17 months , respectively . Annual confirmed incidence of cholera presenting to a treatment facility per 10 , 000 population was <0 . 5 in surveillance zones , except in Goma where it was 4 . 6 . Goma and Conakry had corrected incidences of 20 . 2 and 5 . 8 respectively , while the remaining surveillance zones had a median corrected incidence of 0 . 3 . During outbreaks , the annualized confirmed incidence of cholera presenting to a treatment facility ranged from 0 . 3–3 . 3 and corrected incidence from 2 . 6 to 13 . 0 per 10 , 000 population ( Table 5 ) . The ratio of the mean annual corrected incidence of confirmed cholera to the incidence of suspected cholera varied from 0 . 1 in Abidjan to 0 . 6 in Conakry while it was of 0 . 5 ( SD 0 . 1 ) in outbreak sites . Of 5980 suspected cases identified in a treatment facility with a documented outcome , 69 died . The median CFR was 1 . 1% [IQR: 0 . 7–4 . 3] . The CFR varied from zero percent in Abidjan , Cote d’Ivoire to 10% in Lake district , Togo ( Table 6 ) . We found no statistical differences in the CFR between confirmed and non-confirmed cases . However we observed that deceased patients were less likely to have received culture testing than those alive at discharge ( 35 . 3% vs . 55 . 6% , chi-square p . value = 0 . 001 ) .
In the Africhol surveillance zones , we found an overall annual corrected incidence of confirmed cholera presenting to a treatment facility of 0 . 3 cases per 10 , 000 population , which increased to 20 cases per 10 , 000 during large epidemics . Strong spatial and temporal clustering occurred , with most cases from surveillance zones in Conakry , Guinea and Goma , DRC . Within our study many suspected cases were not cholera confirmed by culture . Furthermore the CRF measured at clinic level remained low in our surveillance sites . From the surveillance data collected in our sites , we were able to identify three epidemiological patterns of cholera: confirmed cases throughout the year such as Goma ( DRC ) ; sporadic cases plus additional outbreaks at irregular intervals such as in Lome ( Togo ) , Mbale ( Uganda ) , and Conakry ( Guinea ) ; and history of recurrent cholera epidemics but no cases during the surveillance period , such as Beira ( Mozambique ) or Abidjan ( Cote d’Ivoire ) . Whatever the location , we found that most cholera cases occurred during the rainy season . Our incidence estimates for confirmed cases showed similar fluctuations by place and time as those reported previously for suspected cases but are substantially lower than estimates modeled from WHO mortality strata [4–14] . In most national cholera surveillance systems , etiologic confirmation occurs only for the first suspected cases , before outbreak declaration . Subsequently , any person with acute watery diarrhea usually would be reported as a cholera case , even though some of these will have other etiologies . Consequently , syndromic surveillance–as reported by most previous studies–likely overestimates cholera incidence . Moreover , the proportion of culture confirmed cases varied widely by site emphasizing the utility of laboratory based studies . At the extreme , in Beira , Mozambique , where a history of large outbreaks likely led providers to have a high index of suspicion for cholera , all sampled suspected cases remained negative for V . cholera [11] . The wide variation we found may have resulted from differences in health care seeking behavior , health care access , type and extent of available health structures , health work training , and adherence to case definitions . For instance , treatment centers in Goma , DRC provided care for patients with any diarrheal disease regardless of etiology , did not charge fees , and treated persons of all ages . In other Africhol sites , cholera treatment centers offering free treatment were established only when authorities declared the outbreak . These issues also may have led to the differences in health care access behaviors and therefore to clinical presentation across sites . Other factors may lead to underestimation of incidence . For example , not all patients will present for care at a medical facility and data collection and reporting may be incomplete . However , our system was not designed to assess these issues . While our incidence rates were lower than those from early reports , CFRs for confirmed cholera cases were consistent with those for suspect cases attending health facilities [5 , 11] . The low identified CFRs emphasize the great strides some cholera endemic countries have made in identifying outbreaks rapidly and improving clinical management . They might also reflect the sensitization of populations in high-risk areas to the importance of seeking timely medical care . Our CFR estimates were limited by our inability to assess deaths in the community which contribute to potential underestimation . Lastly , both our CFRs and overall incidence rates were limited by lack of active community-based surveillance , an objective for which our work was not funded . It is likely that this problem was particularly large for deaths: for example , a study from Kenya found that most deaths occurred among persons who had not sought treatment [15] . Future geographically focused studies might address this issue . In theory , health utilization surveys and capture-recapture analysis could help with estimation of surveillance system sensitivity . However , in epidemic cholera prone settings in Africa , health care utilization surveys are seldom appropriate given the lack of human resources relative to the immediate priority of outbreak control . Capture-recapture analyses similarly are not feasible , given the fluid nature of a surveillance system in which cholera treatment centers are established and dismantled relative to cholera case counts . We identified three epidemiological patterns of cholera in our sites: those with confirmed cases throughout the year such as Goma ( DRC ) ; those with sporadic cases plus additional outbreaks at irregular intervals such as in Lome ( Togo ) , Mbale ( Uganda ) , and Conakry ( Guinea ) ; and those with a history of recurrent cholera epidemics but no cases during the surveillance period , such as Beira ( Mozambique ) or Abidjan ( Cote d’Ivoire ) . The presence of sporadic cases without ensuing outbreaks may occur from occasional introduction of infected persons into a low risk community , e . g . , a community with recent cholera and a high degree of population immunity or a community with good water and sanitation infrastructure . By contrast , sustained occurrence of confirmed cases may result from ongoing environmental source contamination from which a continuously renewed susceptible , non-immune population is infected; this may have occurred in Goma , which has experienced several waves of immigration due to regional conflicts . We found that most cholera cases occurred during the rainy season . However the presence of cases before the rain start suggests that the rainy season may play a role of outbreak amplificatory . Previous studies have found similar results [16] . Substantial precipitation can cause flooding and subsequent mixing of drinking water ( pond , well , lake , river ) with sewage in areas with poor sanitation [17] . Alternatively , the rainy season may trigger human movement , such as the seasonal migration of fishermen along the West African coast or in interior lakes [16 , 18–20] . Our study had several limitations other than those mentioned above . We report data from only eleven geographical sites located in six countries and this may not be generalizable to other African settings . Our correction of incidence based on the lack of testing was applied uniformly across the surveillance period without taking into account seasonal variations . We used a single value to correct for culture sensitivity although culture results may vary by setting based on factors such as laboratory technician skills and stool collection and transportation methods . Finally , CFRs were difficult to assess for confirmed cholera cases because of lack of testing . In the African cholera context , oral cholera vaccine may provide an important short- and medium-term prevention and control measure in addition to case management and long-term efforts to improve water , sanitation and hygiene ( WaSH Despite the utility of mass OCV campaigns have been already demonstrated in some African areas , it remains difficult to determine the best strategy to use and if a relatively circumscribed immunization campaign can prevent an epidemic on the scale of Zimbabwe or Haiti [21–25] . Short duration and geographically focal outbreaks as described in our results will make reactive OCV use challenging , as it was the case in Mozambique [11] . Even in settings with large outbreaks such as Goma or Conakry , cases may occur over a brief period in relatively small geographic areas , such as districts . Preventive immunization may be indeed more appropriate to reduce cholera in target communities , with a potential secondary benefit of reducing transmission outside the target zone… We might also learn from Neisseria meningitidis ( Nm ) meningitis in the meningitis belt [26] . As with cholera , Nm outbreaks were often highly focal , of short duration , difficult to predict , and occurred in areas with limited laboratory facilities . The strategy for years was reactive campaigns following notification of an epidemic . However , vaccine frequently arrived after the epidemic peak and thus its overall efficacy questioned . This situation changed with the introduction of a low-cost Nm serogroup A conjugate vaccine ( MenAfriVac ) through national preventive immunization programs via mass campaigns into persons 1 to 29 years of age [27] . The analogy between OCV and MenAfriVac is also based on the need for national and international commitment for an evidence-based prevention strategy , availability of low-cost vaccine produced in sufficient quantity , and the availability of adequate financial and human resources . While limited to health care facilities , our study presents some of the only prospectively obtained incidence data currently available for Africa . Our findings suggest that confirmed cholera burden is substantially lower than that reported from previous studies based on suspected cholera cases , and that incidence varies substantially over time and place . Efficient use of resources , such as vaccines , could be enhanced by better definition of cholera hot-spots , community behaviors that contribute to cholera spread , and high risk populations , particularly those likely to contribute to seasonal cholera spread . Because of the frequent occurrence of non-cholera causes of diarrhea in cholera endemic zones , development of public health strategies would benefit from reinforcement of local laboratory capacities for diagnosing Vibrio cholerae , something that also would benefit from development of better low-cost diagnostic methods . Environmental reservoirs should be identified and mitigation strategies developed . Determination of other diarrheal disease etiologies across all age groups will help determine the utility of etiology specific interventions . OCV interventions must be conducted , monitored and evaluated to better assess their cost-effectiveness and their health impact among at-risk populations in African contexts . Finally , there is a role for evaluation of low-cost water and sanitation improvements within an integrated strategy for cholera prevention and control .
|
Cholera burden in Africa remains unknown , often because of weak national surveillance systems . Reporting is non-exhaustive for various reasons , such as individual and community fears of stigmatization and economic loss . Furthermore , only 3% to 5% of all cases are laboratory confirmed . A variety of case definitions are used across countries , which could lead to cholera over or under-reporting . Our study presents the first data from prospective multi-country cholera surveillance in Africa , and the only such data based on culture confirmation and that includes a description of clinical presentation . We show how confirmed cholera cases varied over time by setting , and identified three epidemiological patterns that can guide decision-making processes . We documented that reliance on suspected cases–as is usually done in national surveillance–rather than confirmed cases can over-estimate substantially cholera incidence . Finally , our surveillance strategy of using case-based reporting and a standard comprehensive case reporting form provided more information on at-risk populations and geographical hot spots than is currently available in the literature; this is turn should facilitate development of efficient preventive strategies .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"guinea",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"tropical",
"diseases",
"geographical",
"locations",
"uganda",
"bacterial",
"diseases",
"aquatic",
"environments",
"bodies",
"of",
"water",
"neglected",
"tropical",
"diseases",
"africa",
"togo",
"infectious",
"diseases",
"cholera",
"lakes",
"epidemiology",
"marine",
"and",
"aquatic",
"sciences",
"people",
"and",
"places",
"mozambique",
"freshwater",
"environments",
"earth",
"sciences",
"disease",
"surveillance",
"côte",
"d'ivoire"
] |
2016
|
Cholera Incidence and Mortality in Sub-Saharan African Sites during Multi-country Surveillance
|
Regulated degradation of proteins by the 26S proteasome plays important roles in maintenance and signalling in eukaryotic cells . Proteins are marked for degradation by the action of E3 ligases that site-specifically modify their substrates by adding chains of ubiquitin . Innate immune signalling in plants is deeply reliant on the ubiquitin-26S proteasome system . While progress has been made in understanding substrate ubiquitination during plant immunity , how these substrates are processed upon arrival at the proteasome remains unclear . Here we show that specific members of the HECT domain-containing family of ubiquitin protein ligases ( UPL ) play important roles in proteasomal substrate processing during plant immunity . Mutations in UPL1 , UPL3 and UPL5 significantly diminished immune responses activated by the immune hormone salicylic acid ( SA ) . In depth analyses of upl3 mutants indicated that these plants were impaired in reprogramming of nearly the entire SA-induced transcriptome and failed to establish immunity against a hemi-biotrophic pathogen . UPL3 was found to physically interact with the regulatory particle of the proteasome and with other ubiquitin-26S proteasome pathway components . In agreement , we demonstrate that UPL3 enabled proteasomes to form polyubiquitin chains , thereby regulating total cellular polyubiquitination levels . Taken together , our findings suggest that proteasome-associated ubiquitin ligase activity of UPL3 promotes proteasomal processivity and is indispensable for development of plant immunity .
The ubiquitin-26S proteasome system ( UPS ) plays an essential cellular role in selective degradation of proteins that are short-lived or damaged . Degradation of proteins is mediated by an enzymatic cascade in which a small and highly conserved ubiquitin molecule is covalently attached to the substrate . Typically an ubiquitin-activating E1 enzyme forms a high-energy thioester bond to an ubiquitin adduct , which is then transferred onto the active site of an ubiquitin conjugating E2 enzyme . In partnership with an E3 ligase that recruits a specific substrate , the E2 enzyme facilitates formation of an isopeptide bond between the ε-amino group of a lysine residue within the substrate and the carboxy-terminal group of ubiquitin . Reiterations of this reaction cycle result in subsequent ubiquitin molecules being similarly attached to internal lysines of the preceding ubiquitin moiety , thereby generating a polyubiquitin chain on the substrate [1 , 2] . Lysine 48-linked chains of four or more ubiquitins show high affinity for ubiquitin receptors within the 19S regulatory cap of the proteasome [3] . Substrate degradation involves its unfolding by chaperone activity of the 19S particle , cleavage and release of the polyubiquitin chain for recycling , and subsequent threading of the unfolded substrate into the 20S subunit of the proteasome , a barrel-shaped multi-catalytic proteinase [4] . In comparison to other eukaryotes , plant genomes often encode for a disproportionally large number of genes related to the ubiquitin-26S proteasome system . Particularly E3 ligases are overrepresented , with the Arabidopsis genome , for example , encoding for over 1 , 400 different predicted E3 ligase components [5] . Accordingly , protein ubiquitination plays vital roles in numerous aspects of plant biology . Indeed , genetic analyses have shown that many developmental and environmental response pathways exhibit a high degree of dependency on components of the ubiquitin-mediated proteasomal degradation pathway [5–7] . Over the last decade it has become increasingly clear that plant immune responses are particularly dependent on ubiquitin-mediated protein degradation [8–11] . Basal resistance as well as race-specific pathogen resistance triggered by intracellular NLR ( nucleotide-binding/leucine-rich repeat ) immune receptors was compromised by mutation of UBA1 , one of two ubiquitin-activating E1 enzymes in Arabidopsis [12] . Similarly , a screen for ubiquitin conjugating E2 enzymes in tomato revealed important roles for a subset of these enzymes in both local immunity and pathogen effector-induced suppression of immune responses [13] . Furthermore , various E3 ligases of the RING and Plant U-box ( PUB ) types have been identified to play both positive and negative roles in orchestration of plant immune responses [8–11] . Whereas several PUB ligases regulate signalling by pathogen pattern recognition receptors , RING-type E3 ligases have been shown to regulate the proteins levels of NLR immune receptors . Levels of the NLR receptors SNC1 and RPS2 are regulated by the RING-type modular SCFCPR1 ( i . e . SKP1/Cullin1/F-box ) E3 ligase in which the F-box protein , CPR1 ( constitutive expressor of pathogenesis-related ( PR ) genes 1 ) , functions as the substrate adaptor that recruits these NLR receptors [14 , 15] . Failure to degrade these and other NLR receptors can lead to their excessive accumulation , which is associated with spontaneous cell death in absence of pathogen threat [16–21] , emphasising the importance of E3 ligases in cellular decisions of life and death . Ubiquitination also plays key roles in signalling by the immune hormone salicylic acid ( SA ) . Upon pathogen recognition SA accumulates in both local and systemic tissues where it induces profound changes in gene expression to prioritise immune responses over other cellular functions [22] . SA-induced transcriptional reprogramming is mediated by the transcription coactivator NPR1 ( nonexpressor of PR genes ) , a master regulator of plant immunity [23] . Mutation of NPR1 renders plants completely insensitive to SA and consequently defective in local and systemic immune responses [24–27] . Interestingly , transcription coactivator activity of NPR1 is regulated by its signal-induced degradation in the nucleus . In absence of pathogen threat , NPR1 activity is continuously restricted by proteasome-mediated clearance from the nucleus , thereby preventing untimely immune gene expression [28] . Instead of stabilising NPR1 , unexpectedly SA was found to facilitate recruitment of NPR1 to a modular multi-subunit Cullin-RING-Ligase 3 ( CRL3 ) [28 , 29] . Importantly , CRL3-mediated ubiquitination and turnover of NPR1 was necessary for the SA-induced transcriptional activation of its target genes . Taken together , these findings underline the importance of the ubiquitin-26S proteasome system in regulating diverse aspects of plant immune signalling . Upon arrival at the proteasome , ubiquitinated substrates may be extensively remodelled by various proteasome-associated ubiquitin chain modifying enzymes , including ubiquitin ligases of the HECT-type family [30 , 31] . This family of ligases utilise a conserved cysteine residue in the HECT domain that forms a covalent thioester bond with ubiquitin before it is transferred onto the substrate . The ubiquitin remodelling activities of some HECT-type ligases are thought to increase proteasome processivity [32–35] . Given the indispensable roles protein ubiquitination plays in plant immunity , we investigated if HECT-type ubiquitin ligases are involved in proteasome-mediated degradation during immune signalling . Here we report that specific HECT-type ubiquitin ligases of the Ubiquitin Protein Ligase ( UPL ) family regulate SA-mediated plant immune signalling . In particular we show that UPL3 associated with proteasomal degradation pathway components and provided the proteasome with ubiquitin ligase activity , which was necessary for large scale SA-induced transcriptional reprogramming and immunity . These data suggest that UPL3 plays a vital role in promoting immune-related proteasomal processivity .
The Arabidopsis UPL family consists of 7 members that all contain a C-terminal HECT domain that accepts ubiquitin from an E2 conjugating enzyme and then transfers it to the target substrate . N-terminal to the HECT domain , UPLs contain different interaction motifs , including ubiquitin-associated ( UBA ) , ubiquitin-like ( UBL ) and ubiquitin-interacting motifs ( UIM ) , armadillo repeats ( ARM ) , and IQ calmodulin and C-type lectin binding motifs ( Fig 1 ) [36] . As calcium and calmodulin have been implicated in plant defence[37] , we first explored if IQ calmodulin binding motif-containing UPL6 and UPL7 proteins play a role in plant immune responses . We generated upl6 and upl7 knock-out mutants ( S1 Fig ) and infected these plants with a low dosage of the bacterial leaf pathogen Pseudomonas syringae pv . maculicola ( Psm ) ES4326 . At this dosage wild-type plants were resistant to this pathogen , while the SA-insensitive npr1 mutant displayed enhanced disease susceptibility ( Fig 2A ) . Mutant upl6 and upl7 plants exhibited similar levels of resistance to Psm ES4326 as the wild type . Moreover , upl6 upl7 double mutants also effectively suppressed the growth of this pathogen , indicating that UPL6 and UPL7 do not regulate basal resistance responses . To assess if UPL6 and UPL7 regulate induced resistance responses , plants were treated with SA prior to infection with Psm ES4326 . Whereas SA induced resistance in wild-type plants , it failed to activate defences in mutant npr1 plants which remained susceptible ( Fig 2B ) . Both upl6 and upl7 single mutants as well as upl6 upl7 double mutant plants displayed normal SA-induced resistance to Psm ES4326 ( Fig 2B and 2D ) . This was accompanied by normal levels of SA-induced expression of immune marker genes in single mutants ( Fig 2C ) , while the upl6 upl7 double mutant was moderately compromised in expression of SA-responsive PR genes ( Fig 2E ) . These data suggest that UPL6 and UPL7 ubiquitin ligases play only minor roles in SA-mediated immune responses . Next we investigated if UPL ubiquitin ligases with ubiquitin-related domains were involved in orchestrating immune responses . UPL1 and UPL2 are closely related , containing both UBA and UIM signatures , whereas UPL5 harbours an ubiquitin domain ( Fig 1 ) . We selected knockout mutants for each ( S1 Fig ) and infected these plants with Psm ES4326 . At a low infection dosage all three mutants exhibited resistance responses , whereas control npr1 mutants showed the expected disease susceptible phenotype ( Figs 3A and S2A ) . In some bioassays upl5 allowed slightly lower growth of Psm ES4326 ( Fig 3A ) , but this was inconsistent between assays and did not occur at higher inoculation dosages ( Fig 3B ) . Therefore we conclude that upl1 , upl2 and upl5 display relatively normal basal resistance responses . To examine if these UPL ligases regulate induced resistance as activated by SA , plants were treated with SA prior to infection with Psm ES4326 . While SA treatment induced immunity against this pathogen in wild-type and upl2 plants , it failed to enhance resistance in both upl1 and upl5 mutants , which instead resembled the SA-insensitive npr1 mutant in this respect ( Figs 3B and S2B ) . To assess if UPL1 and UPL5 mediate SA signalling , we investigated SA-responsive immune gene expression . Treatment with SA induced strong , NPR1-dependent expression of PR-1 and several WRKY genes in wild-type plants ( Fig 3C ) . By contrast , activation of these genes was strongly reduced in both upl1 and upl5 mutant plants . Together these data indicate that UPL1 and UPL5 are positive regulators of SA-mediated gene expression and immunity . Similar to UPL1 , the related UPL3 and UPL4 ubiquitin ligases contain domains with armadillo-type folds ( Fig 1 ) . Mutant upl3 plants have previously been reported to exhibit aberrant leaf trichome morphology [36] , but otherwise show normal growth and development ( Figs 4A and S1 ) . Similarly , upl4 knockout mutants also exhibited normal growth ( Fig 4A ) . Infection with a low dosage of Psm ES4326 revealed that compared to wild type , mutant upl4 plants displayed normal disease resistance , while upl3 mutants were more susceptible to this pathogen ( Fig 4B ) . Because UPL3 and UPL4 are closely related , we generated upl3 upl4 double mutants that showed reduced growth , early senescence and produced fewer seeds compared to either parent ( Figs 4A and S3A ) . Infection of upl3 upl4 double mutants resulted in striking leaf chlorosis and enhanced levels of Psm ES4326 growth ( Fig 4B ) . These data indicate that UPL3 and UPL4 function additively in the regulation of plant growth and development , and positively modulate basal resistance . Given the pleiotropic phenotypes of the upl3 upl4 double mutant , we decided to continue with our investigation into the single mutants instead . Treatment with SA of mutant upl4 plants induced resistance to Psm ES4326 to a similar extent as in wild type ( Fig 4C ) . By contrast , upl3 mutants resembled the SA-insensitive npr1 mutant in that SA failed to induce resistance to Psm ES4326 ( Fig 4C ) . This phenotype was observed in multiple mutant upl3 alleles and constitutive expression of a transgene consisting of Yellow Fluorescent Protein fused to UPL3 ( YFP-UPL3 ) rescued SA-induced resistance in the upl3 mutant background ( S3B Fig ) . However , constitutively expressed YFP-UPL3 did not rescue basal resistance to Psm ES4326 , suggesting that dynamic UPL3 expression or 5’ and 3’ untranslated regions , which were not included in our expression construct , may also play important gene regulatory roles . The immune phenotypes observed above agreed with the SA-responsive gene expression patterns we subsequently uncovered in the mutants . While upl4 mutants showed predominantly wild type-like immune gene expression profiles in response to SA , mutant upl3 plants failed to activate several immune marker genes ( Fig 4D ) . Again this phenotype was observed in multiple mutant upl3 alleles and constitutive expression of YFP-UPL3 restored SA-responsive PR-1 gene expression in the upl3 mutant background ( S3C Fig ) . To explore if reduced SA-responsive marker gene expression was a transcriptome-wide effect , we performed an RNA Seq experiment on SA-treated wild-type and mutant upl3 plants . SA treatment resulted in differential expression of 2 , 117 genes ( ≥ 2 fold , p = 0 . 05 ) of which 1 , 177 were up- and 940 downregulated after 24 hours . Although some changes were detected between control-treated wild-type and upl3 plants , much larger differential gene expression changes became apparent after SA treatment ( Fig 5A ) . Differences in gene expression were mostly in amplitude with less dramatic activation or repression observed in upl3 mutants compared to the wild type ( Figs 5A , 5B and S4 and S1 Table ) . Indeed , of the 1 , 177 genes activated by SA in the wild type , 860 were expressed at least 1 . 5-fold lower in upl3 mutants ( Figs 5C and S5A ) . Conversely , 515 of 940 SA-repressed genes were down regulated at least 1 . 5-fold less in upl3 mutants ( Figs 5C and S5B ) . These data suggest that UPL3 acts to amplify SA-responsive gene expression changes . To identify the binding sites of potential transcription factors on which UPL3 may act , we performed promoter motif analyses on differentially expressed SA-responsive genes . Analyses of SA-induced UPL3-dependent promoters revealed they are enriched with variants of the immune-related W-box motif ( Fig 5D ) , while promoters that were suppressed by SA in a UPL3-dependent manner contained variants of the developmental E-box motif ( Fig 5E ) . The W-box motif binds WRKY transcription factors , several of which are indispensable for the full activation of SA-dependent gene expression and immunity [23 , 28] . As the W-box is pervasive in SA-responsive genes [23 , 38] and was highly enriched in UPL3 activated but not in UPL3 repressed genes ( Fig 5F ) , our findings indicate that UPL3 acts as a genome-wide amplifier of SA-responsive transcriptional reprogramming and establishment of immunity . To understand how UPL3 might function as a general transcriptional amplifier for SA-responsive genes , we performed a yeast two-hybrid screen for interactors . Because the N-terminus of UPL3 contains armadillo repeats ( Fig 1 ) that are thought to provide a large surface for protein-protein interactions [39] , we used the N-terminal 670 amino acids as bait . In addition to self-interaction , we identified six components related to the ubiquitin-26S proteasome system ( Fig 6A , S2 Table ) . These included the non-ATPase regulatory subunit RPN7 which forms part of the 19S regulatory particle , as well as the armadillo-repeat superfamily protein At3g15180 that contains a domain ( InterPro:IPR019538 ) found in proteasomal chaperones involved in assembly of the proteasome [40] . Moreover , we identified three E3 ubiquitin ligases: ( i ) the F-box protein EBF2 which is part of an SCFEBF1/2 ubiquitin ligase that targets the ethylene-responsive transcription factor EIN3 for proteasome-mediated degradation [41 , 42]; ( ii ) the U-box type E3 ligase PUB23 that has been implicated in plant immunity , interacts with and ubiquitinates the 19S proteasome regulatory particle subunit RPN6 [43 , 44]; and ( iii ) the U-box type E3 ligase PUB31 that is involved in abiotic stress tolerance [45] . Finally , UPL3 was found to interact with UBP12 , a deubiquitinase of the proteasome pathway that negatively regulates immunity [46] . In agreement with these protein-protein interactions , UPL3 was found previously to co-purify with a pathogen effector that targets proteasomes [47] , suggesting UPL3 may physically associate with proteasomes . Indeed , pull down of the proteasomal subunit S2 revealed that YFP-UPL3 co-immunoprecipitated with proteasomes largely independent of SA treatment ( Fig 6B ) . Next we considered how physical association with the proteasome allows UPL3 to function as an amplifier of the SA-responsive transcriptome . SA-responsive gene expression strongly depends on the function of the 26S proteasome [8 , 28] . Indeed , reduced activation of SA-responsive immune genes in upl3 mutants resembled the effect of pharmacological inhibition of the proteasome with the proteasomal inhibitor MG132 in SA-treated wild-type plants ( Fig 6C ) . Given the interconnection between UPL3 and multiple components of the ubiquitin-26S proteasome system , including the 19S subunit , we considered that UPL3 may regulate gene expression by altering total cellular ubiquitination levels . Therefore we treated wild-type and upl3 plants with SA and/or MG132 and pulled down ubiquitinated proteins . Figs 6D and S6 show that compared to wild type , upl3 mutants exhibited markedly reduced levels of total cellular polyubiquitination . Moreover , ubiquitination of RPN10 , a substrate of many different ubiquitin ligase types [48] , was also reduced . This remarkable phenotype suggests that UPL3 promotes polyubiquitination of either a small group of heavily ubiquitinated proteins or an extraordinary wide range of substrates . Our findings suggest that UPL3 may aid the proteasome to reinforce polyubiquitination of its substrates upon their arrival . To explore if plant proteasomes harbour E3 ligase activity immunopurified proteasomes were incubated with E1 and E2 enzymes , Flag-ubiquitin and ATP . Under these conditions proteasomes readily converted free ubiquitin into conjugates ( Fig 6E ) . To investigate if this proteasome-associated E3 ligase activity was dependent on UPL3 , we repeated the assay by comparing proteasomes from upl3 mutants with or without expression of YFP-UPL3 . Proteasomes from water-treated YFP-UPL3 ( in upl3 ) plants formed polyubiquitin conjugates and this activity was stimulated by prior treatment with SA ( Fig 6F ) . By contrast , proteasomes from both water- and SA-treated upl3 mutants exhibited markedly reduced formation of ubiquitin conjugates , demonstrating that proteasome-associated ubiquitin ligase activity was largely UPL3 dependent . Taken together our findings suggest UPL3-dependent proteasome-associated ubiquitin ligase activity is necessary for SA-responsive transcriptional reprogramming and immunity .
The ubiquitin-26S proteasome system plays indispensable roles in transcriptional regulation of plant immune genes but how substrates are processed upon arrival at the proteasome remained unclear . Here we demonstrated that members of the HECT-domain family of UPL ubiquitin ligases play an important role in SA-dependent transcriptional responses and immunity . In particular we report that proteasomes harbour UPL3-dependent ubiquitin ligase activity that was necessary for total cellular substrate polyubiquitination as well as SA-responsive transcriptional reprogramming and immunity . Our findings show that UPL1 , UPL3 , UPL4 and UPL5 function as important regulators of SA-responsive gene expression and immunity ( Figs 3 , 4 and 5 ) . Previous work has found that UPL members play roles in developmental gene expression programmes . UPL3 has been reported to regulate trichome branching by targeting for proteasomal degradation the transcription factors GLABROUS 3 ( GL3 ) and ENHANCER OF GL3 ( EGL3 ) , which control trichome development and flavonoid metabolism [36 , 49] . UPL5 was identified as an interactor of WRKY53 , a transcription factor that promotes leaf senescence [50] . In vitro and in vivo analyses indicated that UPL5 ubiquitinated WRKY53 and targeted it for degradation . Consequently , mutant upl5 plants displayed enhanced expression of a WRKY53-responsive senescence marker gene and accelerated appearance of senescing leaves [51] . Interestingly , WRKY53 is not only a regulator of developmental responses; it was also identified as a regulator of SA-dependent plant immunity . WRKY53 gene expression is SA inducible and a direct transcriptional target of the master immune coactivator NPR1 . Mutation of WRKY53 together with WRKY70 , whose expression is highly correlated with WRKY53 , resulted in susceptibility to Psm ES4326 [23] . Therefore it is plausible that UPLs also regulate the stability of WRKY transcription factors during activation of plant immunity . Indeed , transcriptomic analyses of upl3 mutants indicated that the W-box to which WRKY transcription factors bind , was highly overrepresented in SA-induced , UPL3-dependent gene promoters ( Fig 5 ) . UPLs could remove repressors such as WRKY58 from immune-responsive promoters or facilitate the turnover of WRKY activators whose transcriptional activity may require instability akin to NPR1 coactivator [8 , 23 , 28 , 52] . In this respect , it is worth noting that the broad impact of UPL3 on the SA-responsive transcriptome resembles that of WRKY18 , which functions as an auxiliary amplifier of SA-responsive gene expression [23] . Mutation of UPL3 had a remarkable impact on total cellular polyubiquitination levels , a phenotype rarely observed for E3 ubiquitin ligase mutants . So how could UPL3 have such a large effect on the cellular accumulation of so many polyubiquitin conjugates ? Yeast two-hybrid assays indicated that UPL3 may associate with the 19S regulatory particle of the proteasome ( Fig 6A , S2 Table ) and in planta YFP-UPL3 co-immunoprecipitated with proteasomes ( Fig 6B ) . We show that this interaction was responsible for proteasome-associated E3 ligase activity ( Fig 6 ) . Several proteasome-associated ubiquitin ligases have been described and consequently it has been proposed that instead of regarding substrate ubiquitination and delivery to the proteasome as separate steps , these two steps may in fact be coupled for some substrates [32] . Coupling of ubiquitination to degradation may enhance substrate affinity for proteasome receptors or prevent substrate deubiquitination . Thus , proteasome-associated ubiquitin ligases could have large substrate repertoires . Indeed , the yeast ubiquitin ligase HUL5 and its mammalian homologue KIAA10 are abundantly associated with the proteasome 19S regulatory subcomplex and show high sequence similarity to Arabidopsis UPL3 [30 , 31 , 53] . Similar to knock-out of UPL3 reported here , deletion of HUL5 led to a total cellular reduction in polyubiquitinated substrates [31] . HUL5 appears to indiscriminately amplify the degradation of substrates by elongating their ubiquitin chains , an activity that is not typical for an E3 ubiquitin ligase . Whereas most E3 enzymes have specific substrate targets , E4 enzymes are thought to extend existing ubiquitin chains without much apparent specificity [54] . Instead , their co-location with protein complexes such as the proteasome may provide substrate specificity [31 , 55] . Thus , we propose that similar to the E4 enzyme activity of HUL5 , UPL3 may also function to elongate ubiquitin chains of proteasome-bound substrates . The importance of this activity was previously demonstrated by substrate stalling and incomplete degradation by proteasomes in hul5Δ mutant yeast and in human cells by knocking down the orthologue UBE3C , indicating that ubiquitin chain elongation is necessary for processive degradation of substrates [34 , 35] . Likewise , proteasomal association of another yeast HECT-type ubiquitin ligase , UFD4 , which also shows high sequence similarity to UPL3 , including Armadillo repeats , was found to be necessary for complete substrate degradation [56] . Proteasomal stalling or incomplete degradation of immune-related transcriptional regulators could explain the immune compromised phenotypes of mutant upl3 plants . To date a number of prototypical E3 ubiquitin ligases have been found to also associate with the proteasome , albeit in lower abundance than for example HUL5 . Remarkably , core and variable subunits of the modular SCF ubiquitin ligase also bind the proteasome [32 , 57] . In Arabidopsis the major developmental SCF ligase substrate adapters UFO , COI1 and TIR1 associate with the proteasome [58] , further supporting the notion that ubiquitination of substrates and their proteasomal delivery may be directly coupled processes . Here we report that UPL3 may associate with E3 ligases , including ethylene-responsive SCFEBF2 as well as PUB23 and PUB31 U-box type E3 ligases ( Fig 6A , S2 Table ) . SCFEBF2 targets for proteasomal degradation the indispensable ethylene-responsive EIN3 transcription factor that cross-regulates SA biosynthesis and SA-responsive genes [59 , 60] , while PUB23 , together with its homologues PUB22 and PUB24 , mediates pattern recognition receptor-mediated immune signalling by targeting exocytosis regulators [43 , 61] . Thus , it is plausible that in addition to regulating proteasomal degradation of substrates from the SA signalling pathway , UPL3 may also control immunity by cooperating with E3 ligases from other immune-associated pathways . Such cooperation between E4 ligase-like activities and E3 ligases has been suggested previously . In yeast the RING-type E3 ligase Ubr1 , which targets N-end rule pathway substrates for proteasomal degradation , physically interacted with UFD4 , resulting in the formation of longer substrate-attached polyubiquitin chains [62] . This and an additional report [62] of interaction between HECT-type and other E3 ligases suggest that ubiquitin ligase pairing at the proteasome facilitates processive ubiquitination and degradation of substrates . In conclusion , our findings implicate proteasome-associated HECT-type ubiquitin ligases in the control of plant immune signalling by facilitating substrate polyubiquitination and proteasomal processivity . We reveal this unexpected E4 ligase-like activity plays important roles in the genome-wide amplification of SA-responsive gene transcription and is indispensable for establishment of immunity .
Arabidopsis thaliana wild-type Col-0 , transgenic and mutant plants were sown on soil and grown under a 16/8 hr light/dark regime . After 10–12 days seedlings were separated and transferred to larger pots and grown for an additional 2 . 5–3 weeks . Mutant upl1-1 ( SALK_063972 ) , upl2-2 ( SALK_008974 ) , upl3-2 ( SAIL_339_F05 ) [36] , upl3-4 ( SALK_035524 ) , upl4-1 ( SALK_091246 ) , upl5-1 ( SALK_116446 ) , upl6-1 ( SALK_055609 ) , upl7-1 ( SALK_119373 ) were isolated from the SALK and SAIL collections [63 , 64] and the npr1-1 mutation has been described previously [24] . Double mutants were created by crossing upl3-4 with a second UPL4 knockout mutant allele , upl4-2 ( SALK_040984 ) , while upl6 was crossed with a second UPL7 knockout mutant allele , upl7-2 ( SAIL_403_A11 ) . According to the manufacturer’s instructions the coding sequence of UPL3 ( At4g38600 ) was cloned into pCR8/GW/TOPO ( Thermo-Fisher Scientific ) and recombined with YFP-containing pEarleyGate 104 ( Earley et al . , 2006 ) using LR clonase ( Life Technologies ) to generate the 35S::YFP-UPL3 transgene . The 35S:: YFP-UPL3 vector was transferred into Agrobacterium tumefaciens strain GV3101 ( pMP90 ) using a freeze-thaw method and subsequently transformed into upl3-4 plants by floral dipping [65] . Transgenic plants were selected on soil by repeatedly spraying glufosinate ammonium . Psm ES4326 was grown overnight in liquid LB medium supplemented with 10 mM MgSO4 . Bacterial cells were collected by centrifugation , diluted to the appropriate concentrations and pressure-infiltrated into leaves . In planta bacterial growth was determined 4–5 days after infection by spreading serial dilutions of leaf extracts on LB plates supplemented with streptomycin ( 100 μg/ml ) , 10 mM MgSO4 and 50 μM cycloheximide . To test induced resistance adult plants were sprayed 24 hours prior to pathogen infiltration with water or 0 . 5 mM SA ( sodium salicylate , Sigma-Aldrich #S3007 ) until the leaves were extensively covered with fine droplets . For induction of immune genes and protein analyses , 4-week old soil-grown plants were sprayed with water or 0 . 5 mM SA until the leaves were extensively covered with fine droplets . Alternatively , 12-day-old MS-grown seedlings were submerged in 6-well plates containing 10 ml ( per well ) of water supplemented with or without 0 . 5 mM SA for 6 hours . For proteasome inhibition experiments , seedlings were submerged in solutions containing vehicle ( DMSO ) , 0 . 5 mM SA and vehicle , or 0 . 5 mM SA and 100 μM MG132 for 6 hours . RNA extractions and cDNA synthesis were performed as described [28] . Quantitative qPCR was carried out on 20-times diluted cDNA using Power SYBR Green ( Life Technologies ) and gene-specific primers on a StepOne Plus Real Time PCR system ( Life Technologies ) . For RNA Seq analyses , RNA was extracted from biological triplicate samples as described [28] and further purified using an RNeasy Mini Kit ( Qiagen ) according to the manufacturer’s instructions . qPCR was carried out to confirm appropriate induction of SA-responsive marker genes . RNA was then quantified and submitted to GATC Biotech ( Constance , Germany ) for RNA sequencing . The RNA Seq reads were aligned to the Arabidopsis thaliana TAIR10 genome using Bowtie . TopHat identified potential exon-exon splice junctions of the initial alignment . Strand NGS software in RNA Seq workflow was used to quantify transcripts . Raw counts were normalised using DESeq with baseline transformation to the median of all samples . Data were then expressed as normalised signal values ( i . e . log2[RPKM] where RPKM is read count per kilobase of exon model per million reads ) for all statistical tests and plotting . RNA-seq data have been deposited in the ArrayExpress database at EMBL-EBI ( www . ebi . ac . uk/arrayexpress ) under accession number E-MTAB-7374 . Extraction of overrepresented octamer sequences was performed as previously reported [66] on the top 281 and 292 differentially expressed UPL3-activated and UPL3-repressed gene promoters , respectively . The enriched octamers were aligned according to a conserved pentamer sequence , followed by analysis using Weblogo version 2 . 8 . 2 ( http://weblogo . berkeley . edu/ ) . Additionally , promoters were analysed for statistical over- or underrepresentation of the W-box using POBO [67] . Yeast two-hybrid screening and data analyses were performed by Hybrigenics Services ( Paris , France ) . Amino acids 1–670 of UPL3 were cloned into vector pB29 ( N-UPL3-LexA-C fusion ) and screened against a prey library derived from RNA extracted from Arabidopsis thaliana rosettes infected either with virulent P . syringae pv . tomato DC3000 or with an avirulent strain expressing AvrRpt2 . A total of 65 . 2 million interactions were analysed and 353 positive clones sequenced . Interactions were categorised by confidence scores that are based on a statistical model of the competition for bait-binding between fragments [68 , 69] . For co-immunoprecipitation experiments , tissue was pulverised in liquid nitrogen and protein extracted in 2 volumes of proteasome extraction buffer containing 125 mM Tris-HCl ( pH7 . 7 ) , 0 . 25 mM EDTA , 2 . 5 mM MgCl2 , 5% glycerol , 5 mM ATP , and protease inhibitors ( 50 μg/mL N-p-Tosyl-L-phenylalanine chloromethyl ketone ( TPCK ) , 50 μg/mL Nα-Tosyl-L-lysine chloromethyl ketone hydrochloride ( TLCK ) , 0 . 6 mM phenylmethylsulfonyl fluoride ( PMSF ) ) . Protein extracts were centrifuged ( 17 , 000 g , 20 min . at 4°C ) , supernatants filtered through 0 . 22 μM syringe filters and incubated for 2 hours at 4°C with anti-proteasome S2 antibody ( Abcam , ab98865 at ratio 1:250 ) . Next , protein A-agarose was added ( 20 μl/ml ) and incubated with gentle rocking for another hour . Agarose beads were collected by brief centrifugation and washed 5 times with extraction buffer . Bound proteins were eluted by incubation in SDS sample buffer supplemented with 50 mM dithiothreitol ( DTT ) for 5 min . at 95°C . For analyses of cellular polyubiquitin conjugate levels , twelve-day-old seedlings were placed in solutions containing vehicle ( DMSO ) , 0 . 5 mM SA and vehicle , or 0 . 5 mM SA and 100 μM MG132 for 6 hours . Tissue was then blotted dry and pulverised in liquid nitrogen . Protein was extracted in two volumes of extraction buffer , consisting of phosphate buffered saline supplemented with 1% Triton X-100 , 10 mM N-ethylmaleimide , phosphatase inhibitor cocktail 3 ( Sigma-Aldrich ) , protease inhibitor cocktail [50 μg/mL N-p-Tosyl-L-phenylalanine chloromethyl ketone ( TPCK ) , 50 μg/mL Nα-Tosyl-L-lysine chloromethyl ketone hydrochloride ( TLCK ) , 0 . 6 mM phenylmethylsulfonyl fluoride ( PMSF ) ] , and 0 . 2 mg/ml recombinant GST-tagged tandem ubiquitin binding entities ( TUBE ) [70] . Protein extracts were centrifuged ( 17 , 000 g , 20 min . at 4°C ) , supernatants filtered through 0 . 22 μM syringe filters and incubated overnight at 4°C with 50 μl/ml of packed Protino Glutathione Agarose 4B ( Machery Nagel ) . Agarose was washed 5 times with extraction buffer and bound proteins eluted by incubation in SDS sample buffer supplemented with 50 mM dithiothreitol ( DTT ) for 10 min . at 80°C . Proteasomal E3 ligase activity was assessed by extracting protein from liquid nitrogen pulverised tissue in 2 volumes of proteasome extraction buffer containing 125 mM Tris-HCl ( pH7 . 7 ) , 0 . 25 mM EDTA , 2 . 5 mM MgCl2 , 5% glycerol , 5 mM ATP , and protease inhibitors ( 50 μg/mL N-p-Tosyl-L-phenylalanine chloromethyl ketone ( TPCK ) , 50 μg/mL Nα-Tosyl-L-lysine chloromethyl ketone hydrochloride ( TLCK ) , 0 . 6 mM phenylmethylsulfonyl fluoride ( PMSF ) ) . Protein extracts were centrifuged ( 17 , 000 g , 20 min . at 4°C ) , supernatants filtered through 0 . 22 μM syringe filters and incubated overnight at 4°C with anti-proteasome S2 antibody ( Abcam , ab98865 at ratio 1:250 ) . The next day extracts were centrifuged ( 17 , 000 g , 10 min . at 4°C ) and supernatants collected . Protein A-agarose was then added ( 20 μl/ml ) and incubated with gentle rocking for one hour . Agarose beads were collected by brief centrifugation and washed 3 times with proteasome extraction buffer . Subsequently , agarose beads were incubated for 18 hours with gentle shaking at 30°C in 80 μl reaction buffer ( 125 mM Tris-HCl ( pH7 . 7 ) , 0 . 25 mM EDTA , 2 . 5 mM MgCl2 , 5 mM ATP , 1 mM DTT , 10 μM NSC632836 deubiquitinase inhibitor ) supplemented with , human recombinant E1 enzyme ( 0 . 2 or 0 . 4 μg , BioVision ) , recombinant E2 enzyme UbcH5c ( 0 . 2 μg , Ubiquigent ) , and recombinant human Flag-ubiquitin ( 10 μg , Boston Biochem ) . Agarose beads were eluted by incubation in SDS sample buffer supplemented with 50 mM dithiothreitol ( DTT ) for 10 min . at 80°C . All proteins were analysed by SDS-PAGE followed by western blotting using anti-ubiquitin ( anti-ubiquitinylated proteins clone FK2 , Merck ) , anti-RPN10 ( polyclonal antibody against Arabidopsis RPN10 , Abcam ) , anti-proteasome S2 ( polyclonal antibody against full-length Arabidopsis S2 , Abcam ) , anti-Flag ( monoclonal anti-Flag M2 antibody , Sigma ) and anti-GFP ( mixture of monoclonal antibodies from clones 7 . 1 and 13 . 1 , Roche ) antibodies .
|
Plants are continuously exposed to different disease agents , including bacteria , fungi , oomycetes and chewing or sucking insects . To protect themselves plants have evolved a sophisticated multi-layered immune system that depends on the reprogramming of large gene repertoires to prioritize the expression of immune genes over normal cellular household genes . Activity of the proteasome , a large proteolytic complex that degrades proteins , is vital to coordinate the expression of immune genes . While it is well understood that proteins marked by a chain of the small polypeptide ubiquitin can be targeted to the proteasome for degradation , it remains unclear how these proteins are processed by proteasomes . Here we identify the enzyme UPL3 that enabled plant proteasomes themselves to add further ubiquitin chains to cellular proteins destined for degradation . This is thought to be an important activity that increases the affinity of substrates for proteasomes while preventing them from stalling during degradation . Importantly , we show that this activity of UPL3 is indispensable for gene expression reprogramming and establishment of disease resistance . Thus , by enabling proteasomes to add ubiquitin marks to its substrates , UPL3 regulates key aspects of plant immunity that could be further exploited in future crop protection strategies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"enzymes",
"pathogens",
"immunology",
"enzymology",
"plant",
"physiology",
"ubiquitin",
"ligases",
"plant",
"science",
"plant",
"pathology",
"ligases",
"proteins",
"gene",
"expression",
"proteasomes",
"immune",
"response",
"biochemistry",
"plant",
"defenses",
"protein",
"complexes",
"plant",
"disease",
"resistance",
"plant",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences"
] |
2018
|
Proteasome-associated HECT-type ubiquitin ligase activity is required for plant immunity
|
The first decade of Genome Wide Association Studies ( GWAS ) has uncovered a wealth of disease-associated variants . Two important derivations will be the translation of this information into a multiscale understanding of pathogenic variants and leveraging existing data to increase the power of existing and future studies through prioritization . We explore edge prediction on heterogeneous networks—graphs with multiple node and edge types—for accomplishing both tasks . First we constructed a network with 18 node types—genes , diseases , tissues , pathophysiologies , and 14 MSigDB ( molecular signatures database ) collections—and 19 edge types from high-throughput publicly-available resources . From this network composed of 40 , 343 nodes and 1 , 608 , 168 edges , we extracted features that describe the topology between specific genes and diseases . Next , we trained a model from GWAS associations and predicted the probability of association between each protein-coding gene and each of 29 well-studied complex diseases . The model , which achieved 132-fold enrichment in precision at 10% recall , outperformed any individual domain , highlighting the benefit of integrative approaches . We identified pleiotropy , transcriptional signatures of perturbations , pathways , and protein interactions as influential mechanisms explaining pathogenesis . Our method successfully predicted the results ( with AUROC = 0 . 79 ) from a withheld multiple sclerosis ( MS ) GWAS despite starting with only 13 previously associated genes . Finally , we combined our network predictions with statistical evidence of association to propose four novel MS genes , three of which ( JAK2 , REL , RUNX3 ) validated on the masked GWAS . Furthermore , our predictions provide biological support highlighting REL as the causal gene within its gene-rich locus . Users can browse all predictions online ( http://het . io ) . Heterogeneous network edge prediction effectively prioritized genetic associations and provides a powerful new approach for data integration across multiple domains .
In the last decade , genome-wide association studies ( GWAS ) have been established as the main strategy to map genetic susceptibility in dozens of complex diseases and phenotypes . Despite the success of this approach in mapping variation in thousands of loci to hundreds of complex phenotypes [1–5] , researchers are now confronted with the challenge of maximizing the scientific contribution of existing GWAS datasets , whose undertakings represented a substantial investment of human and monetary resources from the community at large [6] . A central assumption in GWAS is that every region in the genome ( and hence every gene ) is a-priori equally likely to be associated with the phenotype in question . As a result , small effect sizes and multiple comparisons limit the pace of discovery . However , rational prioritization approaches may afford an increase in study power while avoiding the constraints and expense related to expanded sampling . One such way forward is the current trend of analyzing the combined contribution of susceptibility variants in the context of biological pathways , rather than single SNPs [7] . For example , Yaspan et al described an approach that aggregates variants of interest from a GWAS into biological pathways using genomic randomization to control for multiple testing and minimize type I error [8] . The popular software PLINK also includes an option to evaluate groups of associations at the gene level , thus enabling pathway analysis by computing enriched gene sets [9] . A less explored but potentially revealing strategy is the integration of diverse sources of data to build more accurate and comprehensive models of disease susceptibility . Several strategies have been attempted to identify the mechanisms underlying pathogenesis and use these insights to prioritize genes for genetic association analyses . Gene-set enrichment analyses identify prevalent biological functions amongst genes contained in disease-associated loci [10 , 11] . Gene network approaches search for neighborhoods of genes where disease-associated loci aggregate [12 , 13] . Jia et al . reported dmGWAS , a strategy to integrate association signals from GWAS into the human protein interaction network [14] . A similar approach was developed by our group and tested in two large studies comprising more than 15 , 000 cases [15] . Literature mining techniques aim to chronicle the relatedness of genes to identify a subset of highly-related associated genes . For example , Raychaudhuri et al . reported the Gene Relationships Among Implicated Loci ( GRAIL ) algorithm , an approach to assess relationships among genomic disease regions by text mining of PubMed abstracts [16] . Prioritization strategies generally rely on user-provided loci as the sole input and do not incorporate broader disease-specific knowledge . Typically , the proportion of genome-wide significant discoveries in a given GWAS is low , thus leaving little high-confidence signal for seed-based approaches to build from . To overcome this limitation , here we aimed at characterizing the ability of various information domains to identify pathogenic variants across the entire compendium of complex disease associations . Using this multiscale approach , we developed a framework to prioritize both existing and future GWAS analyses and highlight candidate genes for further analysis . To approach this problem , we resorted to a method that integrated diverse information domains naturally . Heterogeneous ( or multipartite ) networks are a class of networks which contain multiple types of entities ( nodes ) and relationships ( edges or links ) , and provide a data structure capable of expressing diversity in an intuitive and scalable fashion . Most existing techniques available for network analysis have been developed for homogeneous networks [17–19] and are not directly extensible to heterogenous networks . Accordingly , the early network analyses in disease biology concentrated on homogeneous networks . However in the last half-decade , the complexity of biological systems has spurred interest in heterogeneous approaches . While still a developing field , network-based biological data integration has been pursued using a variety of techniques . Approaches such as GeneMANIA , weight and then project individual data sources onto a single dimension , enabling homogeneous network algorithms to be used to characterize the resulting graphs [20–22] . Other techniques operate on multi-relational ( single node type , multiple edge type ) networks , for example by taking into account relationships among local clusters and considering the full topology of weighted gene association networks [13 , 20 , 23] . Bipartite networks contain two node types and therefore work well for predicting relationships between entities of two different types ( such as disease-gene associations or drug-disease indications ) following a ‘guilt-by-association’ paradigm [24–26] . Other approaches incorporate greater-dimension heterogeneous networks as input but conflate types and , while improving predictions compared to simpler approaches , cannot effectively identify influential network components [27 , 28] . Heterogeneous networks of arbitrary complexity have also been applied for edge prediction without a formalized feature extraction methodology , which requires manual descriptor determination for each new network design [29] . Recently , new types of edge prediction methods were reported that naturally accommodate any size heterogeneous network . These include data fusion by matrix-factorization [30–33] and metapath-based techniques [34 , 35] . This type of intermediate data fusion can treat all data sources directly ( i . e . without transforming data into “disease space” ) and has been successfully used to infer disease similarities [31] and predict gene function in slime mold and baker’s yeast [32] . A metapath-based approach was recently developed by researchers studying social sciences to predict future coauthorship [34] and provides an intuitive framework and interpretable models and results . An advantage of metapath-based approaches is that they preserve the network structure and provide the flexibility to explore a diverse set of descriptors . In this work , we extended this methodology to predict the probability that an association between a gene and disease exists .
Using publicly-available databases and standardized vocabularies , we constructed a heterogeneous network with 40 , 343 nodes and 1 , 608 , 168 edges ( Fig 1 , S2 Data ) . Databases were selected based on quality , reusability , throughput , and their aggregate ability to provide a diversified , multiscale portrayal of biology ( see Resource selection in Methods ) . The network was designed to encode entities and relationships relevant to pathogenesis . The network contained 18 node types ( metanodes ) and 19 edge types ( metaedges ) , displayed in Fig 2A . Entities represented by metanodes consisted of diseases , genes , tissues , pathophysiologies , and gene sets for 14 MSigDB collections [36 , 37] including pathways [38 , 39] , perturbation signatures , motifs [40 , 41] , and Gene Ontology ( GO ) domains [42] ( Table 1 ) . Relationships represented by metaedges consisted of gene-disease association , disease pathophysiology , disease localization , tissue-specific gene expression , protein interaction , and gene-set membership for each MSigDB collection ( Table 2 ) . Gene-disease associations were extracted from the GWAS Catalog [5] by overlapping associations into disease-specific loci . Loci were classified as low or high-confidence based on p-value and sample size of the corresponding GWAS ( see Associations in Methods and S1 Fig ) . When possible , for each loci , the most-commonly reported gene across studies was designated as primary and subsequently considered responsible for the association . Additional genes reported for the loci were considered secondary . Only high-confidence primary associations were included in the network yielding 938 associations between 99 diseases and 711 genes ( S2 Fig visualizes a subset of these associations ) . To describe the network topology connecting a specific gene and disease , we computed 24 features , each describing a different aspect of connectivity . Each feature corresponds to a type of path ( metapath [53] ) originating in a given source gene and terminating in a given target disease . The biological interpretation of a feature derives from its metapath ( S1 Table ) , and features simply quantify the prevalence of a specific metapath between any gene-disease pair . To quantify metapath prevalence , we adapted an existing method originally developed for social network analysis ( PathPredict ) [34] , and developed a new metric called degree-weighted path count ( DWPC , Fig 2D ) , which we employed in all but two features . The DWPC downweights paths through high-degree nodes when computing metapath prevalence . The strength of downweighting depends on a single parameter ( w ) , which we optimized to w = 0 . 4 and that outperformed the top metric resulting from PathPredict ( S3 Fig ) [34] . We calculated DWPC features for the 22 metapaths of length 3 or less that originated with a gene and terminated with disease . Two non-DWPC features were included to assess the pleiotropy of the source gene and the polygenicity of the target disease . Referred to as ‘path count’ features , they respectively equal the number of diseases associated with the source gene and the number of genes associated with the target disease . For all features , paths with duplicate nodes were excluded , and , if present , the association edge between the source gene and target disease was masked . Further analysis focused on the 29 diseases with at least ten associated genes ( Table 3 ) . The 698 high-confidence primary associations of these 29 diseases were considered positives—gene-disease pairs with positive experimental relationships ( as defined in the Associations section of Methods , S2 Fig ) . The remaining 551 , 823 ( i . e . unassociated ) gene-disease pairs were considered negatives . Low-confidence or secondary associations were excluded from either set . We partitioned gene-disease pairs into training ( 75% ) and testing ( 25% ) sets and created a training network with the testing associations removed . To learn the importance of each feature and model the probability of association of a given gene-disease pair , we used regularized logistic regression which is designed to prevent overfitting and accurately estimate regression coefficients when models include many features . Elastic net regression is a regression method that balances two regularization techniques: ridge ( which performs coefficient shrinkage ) and lasso ( which performs coefficient shrinkage and variable selection ) [54] . On the training set , we optimized the elastic net mixing parameter , a single parameter behind the DWPC metric , and two edge-inclusion thresholds ( S3 Fig ) . While cross-validated performance was similar across elastic net mixing parameters , ridge demonstrated the greatest consistency ( S3 Fig ) , and thus we proceeded with logistic ridge regression as the primary model for predictions . We extracted network-based features for gene-disease pairs from the training network and modeled the training set . We next evaluated performance on the 25% of gene-disease pairs ( 175 positives , 137 , 956 negatives ) withheld for testing . Our predictions achieved an area under the ROC curve ( AUROC ) of 0 . 83 ( Fig 3A ) demonstrating an excellent performance in retrieving hidden associations . Importantly , we did not observe any significant degradation of performance from training to testing ( Fig 3A ) , indicating that our disciplined regularization approach avoided overfitting and that predictions for associations included in the network were not biased by their presence in the network . Furthermore , we observed that at 10% recall ( the classification threshold where 10% of true positives were predicted as positives ) , our predictions achieved 16 . 7% precision ( the proportion of predicted positives that were correct ) . Since the prevalence of positives in our dataset was 0 . 13% , the observed precision represents a 132-fold enrichment over the expected probability under a uniform distribution of priors ( as in GWAS ) . As a next step in our analysis , we recomputed features on the complete network , which now included the previously withheld testing associations . On all positives and negatives , we fit a ridge model ( the primary model for predictions ) and a lasso model ( for comparison ) . Standardized coefficients ( Fig 4 ) indicate the effect attributed to each feature by the models . The lasso highlighted features that captured pleiotropy ( 4 features ) , pathways ( 2 ) , transcriptional signatures of perturbations ( 1 ) and protein interactions ( 1 ) . Despite the parsimony of the lasso , performance was similar between models with training AUROCs of 0 . 83 ( ridge ) and 0 . 82 ( lasso ) . However , since multiple features from a correlated group may be causal , the lasso model risks oversimplifying . Ridge regression disperses an effect across a correlated group of features , providing users greater flexibility when interpreting predictions . From the ridge model , we predicted the probability that each protein-coding gene was associated with each analyzed disease ( S1 Data ) and built a webapp to display the predictions ( http://het . io/disease-genes/browse ) . Using Markov chain randomized edge-swaps , we created 5 permuted networks . Since metaedge-specific node degree is preserved , features extracted from the permuted network retain unspecific effects . These effects include general measures a disease’s polygenicity and a gene’s pleiotropy , multifunctionality , and tissue-specificity . On the first permuted network , we partitioned associations into training and testing sets . Testing associations were masked from the network , features were computed , and a ridge model was fit on the training gene-disease pairs . Compared to the unpermuted-network model , testing performance was noticeably inferior: the AUROC declined from 0 . 83 ( Fig 3A ) to 0 . 79 and the AUPRC ( area under the precision-recall curve ) declined from 0 . 06 ( Fig 3B ) to 0 . 02 ( S4 Fig ) . We interpret the modest decline in AUROC but marked reduction in AUPRC as a direct consequence of the permutation’s particularly detrimental effect on top predictions ( S4 Fig ) . In other words , edge-specificity was crucial for top predictions , while general effects gleaned from node degree performed reasonably well when ranking the entire spectrum of protein-coding genes for association . A commonly-overlooked finding is that the discriminatory ability of gene networks largely relies on node-degree rather than the edge-specificity [55] . However , we found that for top predictions—which are the only predictions considered by many applications—edge-specificity was critical . Interestingly , predictions from the permuted-network model displayed a reduced dynamic range with none exceeding 4% , while predictions from the unpermuted-network model exceeded 75% ( S4 Fig ) . Therefore , even though they achieve reasonable AUROC , the permuted-network predictions would have little utility as prior probabilities in a bayesian analysis where dynamic range is crucial . Furthermore , the signal present in permuted-network features was greatly diminished: few features survived the lasso’s selection resulting in an average lasso AUROC of 0 . 70 versus 0 . 80 for ridge ( S5 Fig ) . Permuting the network significantly reduced the predictiveness of features based on pleiotropy ( 2 features ) , protein interactions ( 2 ) , transcriptional signatures of perturbations ( 1 ) , tissue-specificity ( 1 ) , pathways ( 3 ) , and immunologic signatures ( 1 ) ( S2 Table ) . Six of the eight features selected by the lasso and eight of the top ten ridge features ( ranked by standardized coefficients ) were negatively affected by the permutation . Since our modeling technique preferentially selected/weighted features affected by permutation , we can infer that network components where edge-specificity matters underlie a large portion of predictions . We assessed the informativeness of each feature by calculating feature-specific AUROCs . Feature-specific AUROCs universally exceeded 0 . 5 , indicating that network connectivity , regardless of type , positively discriminates associations . However , performance varied widely by feature and within feature from disease to disease ( Fig 5 ) . Top performing domains consisted of transcriptional signatures of perturbations ( AUROC = 0 . 74 ) , immunologic signatures ( 0 . 70 ) , and pleiotropy ( 0 . 68 , 0 . 67 , 0 . 64 , 0 . 63 ) . Notably , the models greatly outperformed any individual feature , highlighting the importance of an integrative approach . Features whose metapaths originate with an association ( GaD ) metaedge measure pleiotropy ( S1 Table ) . The four pleiotropic features were among the top performing features that did not rely on set-based gene categorization ( Fig 5 ) . Of the four features , GaD ( any disease ) had the highest AUROC despite its lack of disease-specificity , reflecting both the sparsity of disease-specific features and the existence of genetic overlap between seemingly disparate diseases . GaDmPmD and GaDaGaD performed best for immunologic diseases and were affected by permutation , indicating that genetic overlap was greatest between immunologic diseases . On the other hand , the performance of GaDlTlD did not decrease after permutation indicating disease colocalization was not a primary driver of genetic overlap ( S5 Fig and S2 Table ) . We also observed that the lasso regression model discarded the majority of features with a minimal performance deficit , suggesting redundancy among features . Indeed , pairwise feature correlations showed moderate collinearity among features ( S6 Fig ) . Collinearity was especially pervasive with respect to the Perturbations feature , explaining its threefold increase in standardized coefficient in the lasso versus ridge model . The disappearance of all but one other MSigDB-based feature in the lasso model indicated that Perturbations—the feature traversing chemical and genetic transcriptional signatures of perturbations—exhausted meaningful gene-set characterization . In other words , the faulty molecular processes behind pathogenesis align with and are encapsulated by the processes perturbed by chemical and genetic modifications . The Immunologic signatures feature—traversing gene-sets characterizing “cell types , states , and perturbations within the immune system”—was highly predictive and correlated with Perturbations . As expected this feature performed best for diseases with an immune pathophysiology . The one well-performing neoplastic disease ( Fig 5 ) was chronic lymphocytic leukemia , a hematologic cancer with a strong immune component [56] . Additionally , the performance of both the Perturbation and Immunologic features was affected by permutation indicating information beyond the extent of a gene’s multifunctionality was encoded . Existing network-based gene-prioritization methods , frequently rely solely on protein-protein interactions . Our results supported incorporating protein interactions as the two interactome-based features were discriminatory ( AUROCs = 0 . 65 , 0 . 56 ) and affected by permutation . However , when compared to the integrative models or other top-performing features , performance of features that relied solely on the interactome was severely limited . Pathways , another founding resource for many approaches , proved important with KEGG selected by the lasso and all three pathway resources ( AUROCs = 0 . 61 for KEGG , 0 . 60 for Reactome , 0 . 55 for BioCarta ) affected by permutation . The GeTlD feature—measuring to what extent a gene is expressed in tissues affected by the disease in question—peaked in performance around AUROC = 0 . 58 ( S3 Fig ) , was affected by permutation , and required no preexisting knowledge of associated genes . In other words , while approaches based on tissue-specificity may have limited predictive ability on their own , they are broadly applicable ( i . e . less susceptible to knowledge bias ) and provide orthogonal information that could enhance the overall performance of a model . For each type of gene set , we evaluated the effect of increased sparsity on performance by randomly subsampling gene set nodes or edges and measuring the resulting AUROC of the affected feature ( S7 Fig ) . Robustness refers to a gene set collection’s ability to withstand a high extent of masking with little performance deficit . Several of the top gene sets had this property , especially GO processes ( where supersets are common ) , which may indicate nodal redundancy . Contrastingly , the MSigDB gene set with the fewest nodes , KEGG , experienced a more immediate and linear decline in performance . Since KEGG avoids duplication and is stringently and manually curated , this finding is expected . To investigate whether the high predictivity of certain gene set collections was due only to size , we compared performance when subsampling nodes to the KEGG level ( crosses in S7 Fig ) . The two top performing collections , perturbations and immunologic signatures , which also happen to be large , continued to perform better than the majority of complete collections . While performance benefited from increasing densities , a resource’s sparsity often reflects an intrinsic property of the underlying information type . Therefore , when identifying influential mechanisms of pathogenesis , we prefered unadjusted comparisons using the complete network . The WTCCC2 multiple sclerosis ( MS ) GWAS tested 465 , 434 SNPs for 9 , 772 cases and 17 , 376 controls and identified over 50 independently associated loci [57] . Since the GWAS Catalog excludes targeted arrays ( such as ImmunoChip ) , this study remains the largest MS GWAS in the Catalog . To evaluate our method’s ability to prioritize associations identified in a future study , we masked the WTCCC2 MS study from the GWAS Catalog and created a pre-WTCCC2 network . The number of high-confidence primary MS associations was thus reduced from 50 to 13 , with the 37 novel genes identified by WTCCC2 available to evaluate performance . On the pre-WTCCC2 network , we extracted features , fit a ridge model , and predicted each gene’s probability of association with MS . Amongst all 18 , 993 potentially novel genes , the 37 WTCCC2 genes were ranked highly ( AUROC = 0 . 79 , Fig 6 ) . Finally , we designed a framework for discovering and validating novel MS genes that incorporates our network-based predictions . Meta2 . 5 is a meta-analysis of all MS GWAS prior to the WTCCC2 study [58] . We calculated genewise p-values for Meta2 . 5 using VEGAS [59] and observed a large enrichment in nominally significant ( p < 0 . 05 ) genes , suggesting multiple potential associations ( S9 Fig ) . We combined this set of experimental candidates with the top predictions from the pre-WTCCC2 network to discover genes with both strong statistical and biological evidence of association ( S12 Data ) . To ensure novelty , we excluded genes from GWAS-established MS loci and the extended MHC region . We chose a threshold ( S3 Table ) for network-based predictions that performed well in prioritizing the genes identified by WTCCC2 ( Fig 6 ) . This strategy discovered four genes , three of which—JAK2 , REL , RUNX3—achieved Bonferroni validation on VEGAS-converted WTCCC2 p-values ( Table 4 ) . The probability of the observed validation rate occurring under random prioritization is 0 . 01 ( S3 Table ) , demonstrating that incorporating our network-based predictions as a prior increased study power . JAK2 displays overexpression in MS-affected Th17 cells [60] and was implicated in an interactome-based prioritization of GWAS [15] . RUNX3 , a transcription factor influencing T lymphocyte development , has been associated with celiac disease [61] and ankylosing spondylitis [62] and was hypermethylated in systemic lupus erythematosus patients [63] . The region containing REL was uncovered in a recent MS ImmunoChip-based study with 14 , 498 cases [64] . For the gene-dense region containing REL , the ImmunoChip study reported a long non-coding RNA , LINC01185 , overlapping the lead-SNP , rs842639 . However , since greater than 80% of the genome shows evidence of transcription [65] , the probability of incidental overlap with long non-coding RNA is high . REL , however , is an essential transcription factor for lymphocyte development [66] and plays a critical role in autoimmune inflammation [67] . Hence , gene prioritization through integrative analyses offers not only to streamline loci discovery but also subsequent causal gene identification .
In this work , we developed a framework to predict the probability that each protein-coding gene is associated with each of 29 complex diseases . Our predictions draw on a diverse set of pathogenically-relevant relationships encoded in a heterogeneous network . The predictions successfully prioritized associations hidden from the network . Using MS as a representative example , we were able to combine our predictions with statistical evidence of association to increase study power and identify three novel susceptibility genes in this disease . The disease-specific performance ( measured by the AUROC ) for MS was exceeded by twelve other diseases suggesting that our predictions have broad applicability for prioritizing genetic association analyses . Prioritization can range from a genome-wide scale to a single loci where this approach can highlight the causal gene from several candidates within the same association block . For researchers focused on a specific disease , these predictions can be used to propose genes for experimental investigation . Inversely , researchers focused on a specific gene can use this resource to find suggestions for relevant complex disease phenotypes . Most previous explorations of the factors underlying pathogenicity have focused on a single domain such as tissue-specificity [68] , protein interactions [69] , pathways [7] , or disease similarity [70] . The method presented here integrates disparate data sources , learns their importance , and unifies them under a common framework enabling comparison . Therefore , we can conclude that perturbation gene sets—the core of our top-performing feature—are an underutilized resource for disease-associated gene prioritization . Not only did perturbations encompass other set-based gene categorizations , but they greatly outperformed features based on protein interactions , pathways , and tissue-specificity , which form the basis of several prominent prioritization techniques . In addition to characterizing the overall importance of each feature , our online prediction browser visually decomposes an individual prediction into its components . We observed a prominent influence of pleiotropy , consistent with previous studies that identified pervasive overlap of susceptibility loci across complex diseases [71] , especially those of autoimmune nature [72] . Since many existing prioritization techniques are agnostic to the compendia of GWAS associations , they fail to adequately leverage pleiotropy . Unlike approaches initiated from a user-provided gene list , our study only provides predictions for 29 diseases . By not relying on user-provided input , our predictions can serve as independent priors for future analyses . By predicting probabilities , we provide an extensible and interpretable assessment of association that circumvents the limitations inherent to frequentist analyses [73] . Many approaches return no assessment for the majority of genes which fall outside of their set of predicted positives . Here , we overcome this issue and provide a comprehensive and genome-wide output by returning a probability of association for each protein-coding gene . High-throughput biological data is frequently noisy and incomplete [74] . Combining orthogonal resources can help overcome these issues . Accordingly , we found that our integrative model outperformed any individual domain . While this method has shown encouraging performance , some limitations are worth noticing . For example , many biological networks preferentially cover well-studied vicinities [75] . Knowledge biases that span multiple , presumably-orthogonal resources could diminish the benefits of integration . Here , several of the literature-derived domains were removed by the lasso , suggesting redundancy . In addition , biases in network completeness can lead to high-quality predictions for well-studied vicinities and low-quality predictions for poorly-studied vicinities . The permutation analysis provided evidence of this disparity: edge-specificity was critical for top predictions yet only moderately beneficial for the remainder . Subsequently , we caution users to avoid overinterpreting predictions for poorly-characterized genes . To help place predictions in context , the online browser provides a gene’s mean prediction across all diseases and a disease’s mean prediction across all genes . However , we recognize that false negatives will continue to persist in our predictions , and users should be mindful of this limitation when interpreting results . As more systematic and unbiased resources become available [74] , high-quality predictions will emerge for more network vicinities . We reason that the desirable qualities of our predictions are the consequence of the heterogenous network edge prediction methodology . The approach is versatile ( most biological phenomena are decomposable into entities connected by relationships ) , scalable ( no theoretical limit to metagraph complexity or graph size ) , and efficient ( low marginal cost to including an additional network component ) . We have extended the previous metapath-based framework set forth by PathPredict [34] , by: 1 ) incorporating regularization allowing coefficient estimation for more features without overfitting; 2 ) designing a framework for predicting a metaedge that is included in the network; 3 ) developing an improved metric for assessing path specificity; and 4 ) implementing a degree-preserving permutation . Metapath-based heterogeneous network edge prediction provides a powerful new platform for bioinformatic discovery .
This study was approved by the UCSF institutional review board on human subjects under protocol #10–00104 . We created a general framework and open source software package for representing heterogeneous networks . Like traditional graphs , heterogeneous networks consist of nodes connected by edges , except that an additional meta layer defines type . Node type signifies the kind of entity encoded , whereas edge type signifies the kind of relationship encoded . Edge types are comprised of a source node type , target node type , kind ( to differentiate between multiple edge types connecting the same node types ) , and direction ( allowing for both directed and undirected edge types ) . The user defines these types and annotates each node and edge , upon creation , with its corresponding type . The meta layer itself can be represented as a graph consisting of node types connected by edge types . When referring to this graph of types , we use the prefix ‘meta’ . Metagraphs—called schemas in previous work [34 , 35]—consist of metanodes connected by metaedges . In a heterogeneous network , each path , a series of edges with common intermediary nodes , corresponds to a metapath representing the type of path . A path’s metapath is the series of metaedges corresponding to that path’s edges . The possible metapaths within a heterogeneous network can be enumerated by traversing the metagraph . We implemented this framework as an object-oriented data structure in python and named the resulting package hetio . Users are free to browse , use , or contribute to the software , through the online repository ( https://github . com/dhimmel/hetio ) . The included resources , and hence the metaedges and metanodes composing our network , were selected empirically based on a balance among the following properties: 1 ) quality—relevance to human pathogenesis; high accuracy and an optimal trade-off between false positives and false negatives . In some cases , quality concerns prevented the inclusion of a desired metaedge . For example , we omitted ontology-based disease similarly [76] due to an inaccurate Disease Ontology hierarchy [43] , and we omitted disease comorbidity due to high measurement error for uncommon diseases [77] . For included metaedges , we attempted to select the highest quality resource in that domain . 2 ) reusability—easily retrievable and parsable; mapped to controlled vocabularies; well documented; amenable to reproducible ( scripted ) analysis; free of prohibitive reuse stipulations . 3 ) throughput—broad domain-specific coverage generated using systematic platforms that minimize bias . While genetic interactions have previously proven informative [31] , their sparse characterization in humans was deemed unfavorable for our approach . 4 ) diversified , multiscale portrayal of biology—capturing , in aggregate , many aspects of pathophysiology across multiple levels of biological complexity . Levels of the hierarchical architecture of biological complexity include the genome , transcriptome , proteome , interactome , metabolome , cell and tissue organization , and phenome . Balancing these considerations , we integrated as many resources as possible within our computational runtime constraints . PathPredict relied on basic logistic regression to predict coauthorship status from features corresponding to nine distinct metapaths [34] . However , faced with fewer positives to train our model and a large number of features , we adopted a regularized approach , which aims to contain the overfitting tendencies inherent to regression . Regularization penalizes complexity , a trademark of overfitting . We chose the elastic net technique of regularization [54] , which is efficiently implemented for logistic regression by the R glmnet package [80] . Regularized logistic regression requires a parameter , λ , setting the strength of regularization . We optimized λ separately for each model fit . Using 10-fold cross-validation and the “one-standard-error” rule to choose the optimal λ from deviance , we adopted a conservative approach designed to prevent overfitting [80] . On the training set of gene-disease pairs , we optimized the elastic net mixing parameter ( α ) , the DWPC damping exponent ( w ) , and two edge inclusion thresholds . First , we optimized α and w on the 20 features whose metapaths did not include threshold-dependent metaedges . For each combination of α and w , we calculated average testing AUROC using 20-fold cross-validation repeated for 10 randomized partitionings . After setting α and w , we jointly optimized the two edge-inclusion thresholds using the AUROC for the GeTlD feature , whose metapath is composed from the two edges requiring thresholds ( S3 Fig ) . We adopt standardized coefficients as a measure of feature effect size . Standardized coefficients refer to the coefficients from logistic regression when all features have been transformed to z-scores . Standardization provides a common scale to assess feature effect , both within and across models [81] . Starting from the complete network , a permuted network was created by swapping edges separately for each metaedge . Edge swaps were performed by switching the target nodes for two randomly selecting edges [82] . For each metaedge , the number of attempted swaps was ten times the corresponding edge count . We adopted a Markov Chain strategy where additional rounds of permutation were initiated from the most-recently permuted network [82] . A training network was generated from the first permuted network by masking 25% of the associations for testing . Testing performance for the permuted training network model is shown in S4 Fig . When contrasting this performance with the unpermuted-network model , we employed the Condensed-ROC curve to magnify the importance of top predictions [83] . Using the exponential transformation with a magnification factor of 460—the value which maps a FPR of 0 . 01 to 0 . 99—we concentrated on the top 1% of predictions ( S4 Fig ) . A one-sided unpaired DeLong test [84] was used to assess whether feature-specific testing AUROCs from the unpermuted network exceeded those from the first permuted network ( S2 Table ) . We performed a subsampling analysis for 15 gene sets—the 14 MSigDB gene sets and tissues—to assess the effect of sparsity on feature-specific performance ( S7 Fig ) . Two without-replacement subsampling schemes were investigated: node masking and edge masking . For a specific gene set and scheme , we masked a percentage of the gene set and calculated the corresponding feature’s AUROC . We evaluated a range of percentages and performed ten subsampling repetitions for each percentage . We excluded 588 genes from the discovery phase of the multiple sclerosis analysis . First we excluded genes in the extended MHC region ( spanning from SCGN to SYNGAP1 on chromosome 6 [85] ) due to the complex pattern of linkage characterizing this region containing several highly-penetrant MS-risk alleles [57] . Second , we excluded putative MS genes: high-confidence primary genes from the GWAS Catalog and reported genes for the WTCCC2-replicated loci . We omitted genes in linkage disequilibrium with the putative genes by excluding: 1 ) consecutive sequences of nominally significant genes ( using the WTCCC2-VEGAS p-values ) that included a putative gene; and 2 ) high-confidence secondary genes from the GWAS catalog . Post exclusion , 1211 genes were nominally significant in Meta2 . 5 , four of which exceeded the network-based discovery threshold . Using a hypergeometric test for overrepresentation , we calculated the probability of randomly selecting 4 of the 1211 genes and Bonferroni validating at least 3 of the 4 on WTCCC2 ( S3 Table ) . See S1–S12 Datasets for the supporting data and S13 Data for vector figures . The website provides additional resources ( http://het . io/disease-genes/downloads/ ) as well as an interface for browsing results ( http://het . io/disease-genes/browse/ ) . Project related code is available from the github repository ( https://github . com/dhimmel/hetio ) .
|
For complex human diseases , identifying the genes harboring susceptibility variants has taken on medical importance . Disease-associated genes provide clues for elucidating disease etiology , predicting disease risk , and highlighting therapeutic targets . Here , we develop a method to predict whether a given gene and disease are associated . To capture the multitude of biological entities underlying pathogenesis , we constructed a heterogeneous network , containing multiple node and edge types . We built on a technique developed for social network analysis , which embraces disparate sources of data to make predictions from heterogeneous networks . Using the compendium of associations from genome-wide studies , we learned the influential mechanisms underlying pathogenesis . Our findings provide a novel perspective about the existence of pervasive pleiotropy across complex diseases . Furthermore , we suggest transcriptional signatures of perturbations are an underutilized resource amongst prioritization approaches . For multiple sclerosis , we demonstrated our ability to prioritize future studies and discover novel susceptibility genes . Researchers can use these predictions to increase the statistical power of their studies , to suggest the causal genes from a set of candidates , or to generate evidence-based experimental hypothesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Heterogeneous Network Edge Prediction: A Data Integration Approach to Prioritize Disease-Associated Genes
|
Drosophila harbor substantial genetic variation for antibacterial defense , and investment in immunity is thought to involve a costly trade-off with life history traits , including development , life span , and reproduction . To understand the way in which insects invest in fighting bacterial infection , we selected for survival following systemic infection with the opportunistic pathogen Pseudomonas aeruginosa in wild-caught Drosophila melanogaster over 10 generations . We then examined genome-wide changes in expression in the selected flies relative to unselected controls , both of which had been infected with the pathogen . This powerful combination of techniques allowed us to specifically identify the genetic basis of the evolved immune response . In response to selection , population-level survivorship to infection increased from 15% to 70% . The evolved capacity for defense was costly , however , as evidenced by reduced longevity and larval viability and a rapid loss of the trait once selection pressure was removed . Counter to expectation , we observed more rapid developmental rates in the selected flies . Selection-associated changes in expression of genes with dual involvement in developmental and immune pathways suggest pleiotropy as a possible mechanism for the positive correlation . We also found that both the Toll and the Imd pathways work synergistically to limit infectivity and that cellular immunity plays a more critical role in overcoming P . aeruginosa infection than previously reported . This work reveals novel pathways by which Drosophila can survive infection with a virulent pathogen that may be rare in wild populations , however , due to their cost .
It costs insects to invest in immunity . Highly immune Drosophila mate less and produce fewer offspring [1] , [2] , more immune bee colonies are less productive [3] , and crickets with heightened immunity exhibit reduced sexual displays and longevity [4] . Recently , it has been shown that resource availability can also play a role in determining the strength and direction of these trade-offs between immunity and life history traits for insects [5] . While it is clear that individual insects vary with respect to their immune performance , only in the fly are we beginning to identify the genetic basis of this phenotypic variation [6]–[8] . With an understanding of which genetic changes confer enhanced immunity we can begin to elucidate how selection drives and balances investment into immunity in general and more specifically into different aspects of the immune response . The innate immune response of insects is generally classified into cellular and humoral components [6] , [9]–[11] . Cellular aspects of defence involve both phagocytosis by hemocytes and encapsulation of pathogens with biotoxic melanin . These aspects of the immune response are constitutively expressed and broad spectrum in target [12] . The key features of the humoral reaction , in contrast , are its inducibility upon exposure to infection and its specificity of response . Selective initiation of the Toll and/or the immune deficiency ( Imd ) pathways that depend on the specific pathogen , ultimately lead to the production and secretion of different sets of antimicrobial peptides ( AMPs ) [10] , [12]–[14] . A recent study in the beetle , Tenebrio molitor , has suggested a challenge to the conventional wisdom , that the humoral response is the stronger partner of the two arms of the immune response . In the beetle , it appears that the cellular response clears the majority of infecting bacteria in the first hour after infection and that the humoral response acts secondarily to remove any persisting bacteria [15] . Here , in Drosophila melanogaster recently caught from the wild , we have artificially selected for defense against a virulent , opportunistic pathogen , Pseudomonas aeruginosa [16] , [17] . In three highly resistant lines we have examined the relationship between correlated changes in life history and patterns of immune gene transcription . In contrast to traditional approaches that tend to compare gene expression of infected with uninfected flies , our microarray experiments have paired selected lines with unselected lines both post infection . The approach has lead to the identification of transcriptional changes that explain the evolved defense response instead of the genetic basis of the induced immune response . The evolved lines exhibited an effective genetic mechanism for defense against a highly virulent pathogen characterized by an increased transcriptional investment in cellular immunity . This genetic change was costly to females in particular in terms of longevity and fecundity . Antibacterial defense also correlated with an increase in developmental rate in both males and females , which was counter to expectation . Expression changes in a handful of genes that participate both in cellular immunity and host development provided a possible mechanism for this positive correlation through the action of pleiotropy .
Three independent lines stemming from a single base population were selected for improved defense against P . aeruginosa infection over 10 consecutive generations . Three additional populations , unexposed to infection , but reared with the same population size bottlenecks served as pair matched controls . In selected lines , the proportion of flies surviving P . aeruginosa infection rose from ∼15% at G1 to ∼30% by G3 ( see Figure 1 ) . Survival then increased again to ∼70% at G5 where it remained for the duration of the selection regime . There was a significant effect of selection at both G6 ( treatment effect: F1 , 2 = 426 . 02 , P<0 . 0023 ) and G10 ( treatment effect: F1 , 2 = 117 . 44 , P<0 . 0084 ) , with selected lines showed significantly higher survivorship compared to corresponding controls . There was no sexual dimorphism in survivorship for these two generations G6 ( sex effect: F1 , 4 = 1 . 68 , P = 0 . 265 ) and G10 ( treatment effect: F1 , 4 = 0 . 47 , P = 0 . 531 ) nor was there any indication of sex-dependent evolution of survival , G6 ( sex×treatment: F1 , 4 = 0 . 32 , P = 0 . 601 ) and G10 ( sex×treatment: F1 , 4 = 0 . 04 , P = 0 . 843 ) . The mean realized heritability of the evolved survival across the three lines was 16 . 7±1 . 3% ( s . e . m ) . Unlike survivorship , the time it took for infected flies to die following infection did not change under the selection regime ( data not shown ) . After the selection experiment , all fly lines were passaged without infection for a further 5 generations ( G15 ) . In the absence of selection , survival in the selected lines returned to pre-selection baseline levels and was no different from G15 controls ( treatment effect: F1 , 2 = 0 . 5 , P = 0 . 848 ) ( Figure 1 ) . To assess the fitness cost of evolved defense in the selected flies , six life-history traits representing major aspects of host fitness were measured at G9 . Longevity was quantified by rearing virgin males and females separately and then recording their time to death in days . A general linear model demonstrated there was no sex or sex×treatment effect on longevity ( data not shown ) . While there was no effect of selection on longevity ( Figure 2B ) in males ( t2 = 1 . 70 , P = 0 . 14 ) in the absence of infection , a significant reduction ( t2 = 4 . 07 , P<0 . 01 ) in average lifespan of female flies was observed in selected flies relative to control flies ( Figure 2A ) . A general linear model demonstrated there was no sex or sex×treatment effect on body mass ( data not shown ) . The mean body mass for selected female ( 1 . 21±0 . 010 g , Figure 2C ) and male ( 0 . 71±0 . 008 g , Figure 2D ) flies were not different ( data not shown ) from their respective controls , 1 . 20±0 . 013 g and 0 . 69±0 . 007 g . Selected flies developed from egg to eclosion ( Figure 2D ) on average ∼12 hours faster ( t2 = 13 . 0 , P<0 . 01 ) than controls . Mean egg viability ( Figure 2F ) of the selection lines ( 54% egg hatch ) was lower ( t2 = 73 . 1 , P<0 . 001 ) than that of controls ( 78% ) . Number of offspring produced from a single mating between a pair of virgin flies was recorded as female productivity . The mean number of offspring produced ( Figure 2G ) in selected lines , in contrast , did not differ when compared to controls ( t2 = 3 . 3 , P = 0 . 08 ) . To assess the effect of selection on male attractiveness , a selected male and a control male were allowed to compete for a female from the base population . The mating success of male flies from selected lines did not differ compared with controls ( F1 , 1 = 0 . 68 , P = 0 . 56 ) . Both selected and control lines were infected at G10 and their RNA was extracted for transcriptional profiling experiments . This comparison specifically revealed the changes in expression due to selection for defense . This is in contrast to the traditional approach of comparing infected lines to uninfected , where the question is instead about which genes are induced after infection . A total of 414 ( 337 up , 77 down ) transcripts showed shared patterns of altered expression in all three lines after selection ( Figure 3 ) . Expression profiles of S1 and S2 were most similar to one another . Approximately , 69 immune related genes were significantly up-regulated in at least 2 of the 3 selected lines and 46 of these genes showed similar increases across all three lines ( Table S1 ) . Eighteen genes with known roles in either the cellular or humoral immune response showed parallel changes in expression in at least 2 of the 3 selected lines ( Table 1 ) . Three peptidoglycan-recognition protein ( PGRP ) genes showed up-regulation in at least two of the three selected lines ( Table 1 ) . Both PGRP-SB1 and PGRP-SD are produced in the fat body and are only induced upon infection . PGRP-SB1 codes for a bactericidal amidase [18] , while PGRP-SD , which functions as a receptor for gram-positive bacteria is involved in Toll activation [19] . PGRP-SC2 is a predicted amidase and was up-regulated in S2 and S3 [20] . Three AMP genes belonging to two families are also up-regulated in selected flies ( Table 1 ) . Drosomycin-4 and -5 , which are primarily antifungal and target gram-positive bacterium , showed increased expression in all three selected lines [21] . Diptericin B , which has previously been shown to be stimulated upon P . aeruginosa infection , showed the strongest expression changes among AMP genes [22] . Both persephone and easter which encode serine endopeptidases and that regulate the Toll signalling pathway [13] were significantly up-regulated in all selected lines ( Table 1 ) . In previous studies examining the expression profiles of infected flies in response to a range of pathogens , including Pseudomonas , the humoral response dominates in terms of numbers of responsive genes ( Table 2 ) . Here , as best seen by the ratio of the number of humoral/cellular responding genes , the nature of evolved defence has shifted toward the cellular . The cellular genes responding to selection in this study are associated with both recognition/phagocytosis and melanization/coagulation . Many of these genes ( N = 8 ) were up-regulated in all three selected lines ( Table 1 ) . The complement related , Thioester-containing proteins ( Tep ) 1 and Tep2 function as opsonins that bind to pathogen surface to promote the detection and phagocytosis of the invading microbes [23] . Tep2 has previously been shown to be required for effective phagocytosis of Gram-negative bacterium E . coli [24] . Two phagocyte specific receptor molecules Scavenger receptor class C type 1 ( SR-C1 ) and eater , which are found on hemocyte surface that bind to a broad range of pathogens [25] , [26] , are up-regulated in all selected lines . Nimc1 is another phagocytosis gene , which is structurally related to phagocytosis receptors such as eater and Draper , plays an important role in both phagocytosis and development as they are efficient in removing microbes as well as apoptotic cells [27] . Annotation of CG10345 and CG2736 suggest they have cell adhesion and scavenger receptor activities [28] . CG30427 , CG7593 and CG3891 are genes required for phagocytosis [24] , [28] and CG7593 , CG8193 [24] and Black cells [29] have monophenol monooxygenase activity and are essential for the production of melanin from tyrosine ( Table 1 ) .
The rapid response to selection by G5 , indicates that the initial population of D . melanogaster harbored substantial additive genetic variation for defense against P . aeruginosa infection . The proportion of surviving individuals in the selected population , however , did not increase above 80% despite continued selection pressure . This in combination with the rapid decrease in population survivorship after selection was removed also suggests the presence of antagonistic pleiotropy and/or physiological constraints at work . Corresponding reductions in fitness attributes in selected flies , namely female longevity and fecundity also provide evidence of a trade-off . Such negative correlations between immunity and other aspects of host fitness are predicted [30] and well-documented in the literature [1] , [2] , [4] , [31] . The consistent correlated increase in antibacterial defense and developmental rate in the selected lines was , however , surprising . An elevated investment in immune defense predicts a lengthening of the development processes caused by the depletion of essential nutrients [32] . Indeed the direction of this predicted trade-off has been confirmed in a selection experiment for sexual competitiveness in Drosophila [33] and virus resistance in moths [32] . Here the increase in developmental rate occurred without a reduction in body mass that may be attributed to a lack of competition for food under laboratory conditions . An examination of the transcriptional profiles of our selected lines revealed expression changes in a number of genes that have dual roles in both development and immunity . We , therefore , propose that pleiotropy between developmental and cellular immune processes and the multi-tasking functional role of hemocytes may underlie the shift toward faster development . The Toll signaling pathway , which is an essential component of humoral immunity , also plays a key role in dorsal-ventral pattern formation in Drosophila embryos [34] , [35] . The signal for dorsal-ventral axis formation is conveyed by serine proteases and Easter , which is the last serine protease in a cascade that modifies the transmembrane Toll receptor and leads to activation of the pathway [36] , [37] . The process of melanization requires the activation of prophenoloxidase ( PPO ) to PO . The activation of PPO and Easter are negatively regulated by a single serine protease inhibitor ( serpin27 ) [36] , [38] . Transcriptional profiling of our selected lines showed that four POs genes and Easter were up-regulated in all lines . The decrease in developmental time can thus be explained in part by the selection for PPO activation , which would consequently activate Easter and alter the timing of the dorsal-ventral axis formation in the embryo [38] . In addition to patrolling the hemolymph for invading microorganisms , the hemocytes are known to play important roles during embryonic development . Hemocytes are the prominent producer of embryonic basement membrane proteins including proteoglycan papilin and the major connective tissue collagen IV [39] , [40] , both of which are up-regulated in all selected lines . Hemocytes migrate along conserved pathways in the embryo and shape various tissues by removing apoptotic cells and depositing extracellular matrix . Hemocyte migration and number are both tightly controlled [40] . In Drosophila , the number of hemocytes is shown to influence the outcome of the infection specifically , greater numbers of circulating hemocytes confer greater immunity [41] , [42] . We found that the selected flies evolved a greater investment in cellular immunity that could translate into increases in hemocyte number and/or activity . This in turn could also alter the rate of development in selected flies . The hallmark of the humoral immune response is the production of AMPs as regulated by the Toll and Imd pathways . The signaling cascades that lead to AMP activation are well studied and it is now generally accepted that whether one or both pathways respond to infection depends on the specific pathogen [43] . Shared components that exist in both pathways also provide for some level of cross-regulation [21] , [44] , [45] . Gene knockout studies have found that flies deficient for either Toll or Imd pathways are more susceptible to P . aeruginosa infection than the wild type [46] . We compared the transcriptome of selected flies to that of controls during early infection in an attempt to identify mechanisms for limiting the initiation and the early progression of P . aeruginosa infection . Components of the Toll pathway including persephone and PGRP-SD were up-regulated in all selected lines . AMP genes from both pathways including drosomycin ( Toll ) and diptericin ( Imd ) , showed similar patterns of expression increase across all lines . Our data indicate that the Toll and Imd pathways work synergistically as part of the evolved defense against Pseudomonas aeruginosa . P . aeruginosa synthesize an extensive collection of virulence associated factors that suppress the host immune defense . Drosophila hemocytes , which are the target of several P . aeruginosa toxins , are impaired by the bacterium leading to suppression of phagocytosis [22] , [47] . We found a strong involvement of cellular immunity in selected lines that appears to have overcome this immune suppressive effect , possibly acting very early in the infection process [12] , [15] before toxins could be produced . All major aspects of cellular immunity including recognition , phagocytosis and melanization are involved in fighting the bacterium . The comprehensive list of cellular immune genes begins with opsonins and surface receptors that recognize and phagocytose bacteria . An array of lysosomal enzymes , proteases , lipases and DNases was up-regulated in selected flies that are involved in the break down of the bacterium in the phagosome ( Table S1 ) . Melanization and coagulation genes , including PO genes , which produce melanin that physically impede the growth of intruding microorganisms [14] , are up-regulated in selected flies . The conserved pattern of cellular immunity gene expression among the selected lines emphasizes the crucial role of hemocytes in suppressing P . aeruginosa . This also suggests that the synergistic activation of phagocytosis , AMP production and melanization together in selected flies is the best strategy in limiting bacterial infection [41] , [42] . The selected flies have evolved mechanisms to overcome the immune suppressive effects of P . aeruginosa that involve a substantial mobilization of cellular immunity as well as investment in the humoral response . We think we see greater evidence of a cellular component in our study as compared to previous work with Pseudomonas as well as other pathogens due to a combination of both methodology and the role of selection . First , it is important to remember that our control lines were also infected and so we are focusing only on the evolved aspects of the response . Evolution of greater investment into the cellular response may be the most effective means of pathogen control . This is in keeping with recent experimental work showing the efficacy of the cellular response over the humoral in early clearing of systemic infections [12] , [15] . It may also be that given the inducible nature of the humoral response that it is already operating at the upper limits of its functionality determined by cellular constraints instead of lack of genetic variation . In either case , the investment in both aspects of immunity has come at a cost particularly for females in terms of longevity and fecundity . Both selected males and females also exhibit accelerated development that may be due to changes in expression of shared gene sets in both processes and the multifunctional role of hemocytes . These experiments have revealed highly effective mechanisms of defense available to genetically diverse flies that are nonetheless unsustainable in the absence of continuous pathogen pressure due to their cost .
Brisbane ( BNE ) base stock was founded from 26 females D . melanogaster caught around the University of Queensland St Lucia campus in August 2006 . The flies were treated with 0 . 5% penicillin and streptomycin in the diet for one generation [48] and then passaged without antibiotic for more than 10 generations before the start of the selection experiment . A large inbred population was maintained as the base stock and reared on standard yellow corn meal medium . P . aeruginosa PA01 was cultured as in LB medium supplemented with 100 mg ml−1 ampicillin at 37 °C [49] . For infection , the concentration of an overnight bacterial culture was adjusted to an OD of 0 . 5±0 . 05 measured spectrometrically at 600 nm . The culture was then diluted 100 fold using sterile LB . This OD was determined at the start of the selection experiments to achieve a population kill rate of 80–90% . The base stock was split into 3 control and 3 selected lines . These replicate populations were used to test the reproducibility of the selection given the genetic variation present in the base population . Selected lines were infected each generation with PAO1 and the survivors allowed to produce the subsequent generation . Selection was applied for 10 generations . For each round of selection , 8 sub-replicate populations consisting of 20 flies each per gender ( 160 flies per gender per line per generation ) were infected with P . aeruginosa PA01 . Mated flies aged to 4–7 days old were anaesthetized with CO2 and infected as previously described by dipping a sterile needle in the bacterial culture and piercing the intrathoracic region of the fly [11] . Fly mortality was then monitored for each population over 48 hours . Survivors from each of the 8 sub-replicates were pooled into a single population to seed the subsequent generation . The control lines were not infected , but were exposed to the same bottleneck in population size as their paired selected lines by randomly selecting a set of individuals to found the next generation . All flies were passaged for a further 5 generations after G10 without selection . Survival in selected lines was monitored each generation . A subset of control line flies not used to found subsequent generations were also tested for survival post infection at G6 and G10 . After G10 , the lines were passaged for another 5 generations without infection followed by an additional assessment of survivorship at G15 . Realised heritability of infection survival was calculated for each of the selected lines with sexes pooled as the ratio of the cumulative selection response to the cumulative selection differential [50] . For this calculation , we modelled infection survival as a threshold character following transformation [51] . Longevity . Virgin female and male flies were kept in separate vials in populations of 20 ( 5 replicate populations per gender per line ) and moved onto fresh food weekly . Fly death was recorded at each food change . Body mass . Flies were placed in vials on the first day of eclosion and aged for a further three days . Flies were then briefly anaesthetized with CO2 and weighed individually on an electronic balance . Traits were measured for both sexes ( 40 individuals per gender per line ) . Developmental time and viability . Twelve eggs laid by a female within a 6 hour window were placed onto a vial after mating with a single male ( 40 replicates per line ) . The eggs were monitored every 6 hours . The period of time ( hours ) from the point of oviposition to the recorded time of eclosion was recorded as the development time . Viability was calculated as a percentage of eggs hatched from a possible of twelve . Female Productivity . Pairs of virgin flies were placed together in a vial and males were removed after 24 hours . The mated females ( 40 replicates per line ) were moved onto fresh vials every 5 days to lay eggs . The total number of viable adult offspring produced by each female was recorded as its productivity . Male Mating Success . A selected male and a control male each powdered with micronized dust of distinct colors were placed with a female from the base population for 90 minutes ( Variable N , 137 to 215 replicates per line ) . Female choice was scored by identifying the male that the female had chosen as a mate . Microarrays were used to screen for genes demonstrating expression changes in selected lines relative to control lines after bacterial infection in G10 . Only male flies were extracted and compared in this analysis . A dual colour reference design paired each selected and control line . Each pair was represented by technical replicates ( N = 3 ) that were then replicated with a dye swap ( total N = 6 ) . Microarrays were of the 4×44 K format ( Agilent ) each containing controls and 3 replicates of each 60 mers feature randomly distributed across the layout . The D . melanogaster genomic sequence ( Release 5 . 4 ) was obtained from Flybase [28] and was used for construction of oligonucleotides using eArray Version 5 . 0 ( Agilent Technologies Inc . , Santa Clara , CA ) . After removing probes that cross hybridised , a total of 13 , 875 transcripts which represented 12 , 041 genes were spotted onto each microarray . Pools of 20 males representing each line were snap frozen in liquid nitrogen and extracted for Total RNA using Trizol ( Invitrogen Corp . , Carlsbad , CA ) . RNA was then purified using Qiagen RNeasy kits according to manufacturer's instructions . Further sample preparations and hybridizations were then carried out by the Special Research Centre Microarray Facility at the University of Queensland . Sample quality was examined using the Agilent 2100 Bioanalyzer ( Agilent Technologies Inc . , Santa Clara , CA ) . Fluorescent cDNA was synthesized using Agilent Low RNA Input Linear Amplification Kit with Cyanine 3 or Cyanine 5-CTP . For each transcript , median signal intensity , background signal intensity , flag and saturation were extracted and analyzed using Genesping v . 7 . 0 ( Agilent Technologies Inc . , Santa Clara , CA ) . Probes that were not detected in at least one hybridization were considered uninformative and excluded from further consideration . An intensity dependent ( Lowess ) normalization ( Per Spot and Per Chip ) was used to correct for non-linear rates of dye incorporation as well as irregularities in the relative fluorescence intensity between the dyes . Hybridizations from each line were used as replicate data to test for significance of expression changes using the cross-gene error model . The Bonferroni multiple testing correction was used to reduce the occurrence of false positives . All array data have been deposited in ArrayExpress ( http://www . ebi . ac . uk/microarray-as/ae/ ) under the accession # E-MEXP-2054 . Quantitative real-time PCR ( RT–PCR ) was used to validate the expression of a subset of 6 immune genes showing increased expression across all three selected lines on the arrays ( Table 3 ) and that represented some of the major functional categories of the immune response . RNA was extracted as above and then treated with 2 µl of DNase I ( Roche ) for 30 minutes at 37°C to eliminate genomic DNA . Approximately 0 . 5 µg of total RNA was reverse transcribed using random primers and SuperScript III reverse transcriptase ( Invitrogen ) according to manufacturer's protocols . Quantitative PCR ( qPCR ) was performed on Rotor-gene 6000 ( Corbett Life Science , Sydney , NSW ) using Platinum®SYBR®Green ( Invitrogen Inc , Carlsbad , CA ) according to manufacturer's instructions . For each sample a mastermix of 2 µl RNase-free water , 5 µl of SYBR Supermix and 0 . 5 µl of each primer ( 10 µM ) was added to 2 µl of cDNA . Three replicates were run for each sample . The cycling protocol was as follows; 1 cycle UDG incubation at 50 °C for 2 minutes , 1 cycle Taq activation at 95°C for 2 minutes , 40 cycles of denaturation at 95 °C for 5 s , annealing at 60 °C for 5 s , extension at 72°C for 15 s , fluorescence acquisition 78 °C , and 1 cycle of melt curve analysis from 68–95°C in 1°C steps . The raw output data of Cycle Threshold ( CT ) was normalized by taking into consideration the differences in amplification efficiency of target and the reference genes using Q-gene software [52] . The linear normalized expression ( NE ) was analyzed using Statistica 8 . 0 ( StatSoft , Inc . ) . D . melanogaster ribosomal protein rpS17 was used as the reference gene ( Table 3 ) .
|
The fruit fly is commonly used as a model organism to understand the mechanistic nature of the immune response to bacterial pathogens . The fly is also commonly used to understand what immunity costs hosts in terms of other traits such as life span and reproductive success . Here , we examine these two questions together in flies selected for improved defense against the bacterium Pseudomonas aeruginosa . We show that selected flies develop from egg to adult more rapidly than unselected flies . It appears that the selected flies invest more heavily in a wing of the immune system that involves engulfment and walling off of invading bacteria . This investment can also explain the shift in developmental rate , as these two biological pathways are controlled by shared sets of genes . These latter two findings are counter to the conventional wisdom and reveal a costly , but effective , means for the fly to circumvent the virulence of Pseudomonas aeruginosa . This bacterium is normally deadly , as it has specific mechanisms to evade the host immune response . Our work is significant for demonstrating a pathway for flies to survive bacterial infection with Pseudomonas aeruginosa and for offering a reason why such a defense is not normally present in wild populations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology",
"microbiology/innate",
"immunity"
] |
2009
|
Effective but Costly, Evolved Mechanisms of Defense against a Virulent Opportunistic Pathogen in Drosophila melanogaster
|
Brain connectivity studies have revealed that highly connected ‘hub’ regions are particularly vulnerable to Alzheimer pathology: they show marked amyloid-β deposition at an early stage . Recently , excessive local neuronal activity has been shown to increase amyloid deposition . In this study we use a computational model to test the hypothesis that hub regions possess the highest level of activity and that hub vulnerability in Alzheimer's disease is due to this feature . Cortical brain regions were modeled as neural masses , each describing the average activity ( spike density and spectral power ) of a large number of interconnected excitatory and inhibitory neurons . The large-scale network consisted of 78 neural masses , connected according to a human DTI-based cortical topology . Spike density and spectral power were positively correlated with structural and functional node degrees , confirming the high activity of hub regions , also offering a possible explanation for high resting state Default Mode Network activity . ‘Activity dependent degeneration’ ( ADD ) was simulated by lowering synaptic strength as a function of the spike density of the main excitatory neurons , and compared to random degeneration . Resulting structural and functional network changes were assessed with graph theoretical analysis . Effects of ADD included oscillatory slowing , loss of spectral power and long-range synchronization , hub vulnerability , and disrupted functional network topology . Observed transient increases in spike density and functional connectivity match reports in Mild Cognitive Impairment ( MCI ) patients , and may not be compensatory but pathological . In conclusion , the assumption of excessive neuronal activity leading to degeneration provides a possible explanation for hub vulnerability in Alzheimer's disease , supported by the observed relation between connectivity and activity and the reproduction of several neurophysiologic hallmarks . The insight that neuronal activity might play a causal role in Alzheimer's disease can have implications for early detection and interventional strategies .
Like many other complex networks , the human brain contains parts that are better connected to the rest than others: ‘hub’ regions . Evidence is increasing that a collection of brain hub regions forms a ‘structural core’ or ‘connectivity backbone’ that facilitates cognitive processing [1] , [2] , [3] . Brain hub regions are mainly located in heteromodal association cortices ( which integrate information coming from primary cortices ) , and show a striking overlap with the Default Mode Network [4] , [5] . Furthermore , their function has been related to fundamental cognitive features such as consciousness , memory , and IQ [6]–[10] . The central role and large responsibility of hub network regions has an obvious downside: hub damage can have a dramatic impact on network integrity [11] , [12] . One of the most intriguing recent insights in this regard has emerged from network-related studies in the field of Alzheimer's disease ( AD ) : cortical hub areas turn out to be exceptionally vulnerable to amyloid deposition , hypometabolism and , eventually , atrophy [13]–[15] . This fascinating link between connectivity and susceptibility to AD pathology deserves further study: what could be causing the hub vulnerability ? The prevailing amyloid-cascade hypothesis of AD states that interstitial amyloid-beta proteins exert a toxic effect on surrounding neurons and synapses , thereby disturbing their function and eventually causing dementia [16] . However , this theory does not provide an explanation for the selective vulnerability of highly connected hub areas . Could an activity-driven mechanism , i . e . hub areas suffering most damage due to their higher connectivity and activity level have any legitimacy ? Chronic , excessive metabolic demand can lead to tissue damage in many organs , and the human brain has extraordinary energy demands . Furthermore , major AD risk factors such as age , ApoE genotype , vascular damage and female gender have all been linked to an increased burden on neuronal metabolism , activity and plasticity [17]–[19] . Recently , direct evidence was presented that excessive neuronal and/or synaptic activity leads to amyloid deposition [20] , [21] , [22] . However , whether this relation between neuronal activity and AD pathology exists in humans , and whether hub regions are indeed the most active areas of the brain has not yet been explored . We speculated that an ‘activity dependent degeneration’ scenario , in which hub regions are preferentially affected due to high neuronal activity levels , could be a plausible disease mechanism . To test this hypothesis , a model is required that incorporates both large-scale connectivity as well as ( micro-scale ) neuronal activity . The macroscopic level is needed to provide a realistic structural human brain topology , including hub regions . Topological maps are well within reach nowadays , since an increasing amount of imaging data describing the human connectome is becoming available [1] , [23] , [24] . Imposed on this structural framework , a realistic representation of network dynamics is required . For this purpose , so-called neural mass models ( NMMs ) can be employed [25]–[27] . Here , each neural mass reflects activity in a brain region by representing a large population of interconnected excitatory and inhibitory neurons , characterized by an average membrane potential and spiking density . Multiple neural masses can be coupled according to any desired structural topology ( e . g . human anatomical data ) to form a dynamic brain model , which can then be employed to investigate the relationship between connectivity and neuronal activity [28]–[30] . Structural ( anatomical ) connectivity and functional ( dynamical ) connectivity are strongly related , but not always in a straightforward way [5] , [31]–[33] . It has been shown that macroscopic models of mammalian brain networks combined with graph theoretical analysis may explain the topology of functional networks at various time scales [34]–[36] . To simulate disease , macroscopic models and graph theory have been used to predict the structural and functional consequences of various types of lesions on brain networks [11] , [12] , [30] . Similarly , the gradually progressive neuronal damage of neurodegenerative processes such as AD can be modeled using this approach , and analyzed with graph theoretical tools [14] , [37]–[39] . The novel aspect of the present study is that the degenerative damage is based on neuronal activity itself . In short , by simulating neuronal dynamics on a network that is modeled on a realistic human cortical connectivity structure we explore the relation between large-scale connectivity and neuronal activity in normal and abnormal conditions . In the present study we use this approach to a ) establish that cortical hub regions , because of their high connectivity , possess the highest intrinsic neuronal activity levels , and b ) demonstrate that ‘Activity Dependent Degeneration’ ( ADD ) , in which brain connectivity is damaged based on local neuronal activity levels , may serve as a computational model of AD that offers a potential explanation for hub vulnerability .
To assess whether the most highly connected cortical regions also showed the highest levels of neuronal activity , we plotted spike density and total power for all regions against the structural degree of nodes ( figure 1A ) . The group of 13 regions with the highest ( ‘very high’ category in the figure ) structural degree were defined as hubs; the remaining 65 regions were labeled as non-hubs . In non-hubs , spike density actually showed a weak negative relation with structural degree , but in hubs clearly higher levels were found compared to non-hubs ( p<0 . 01 ) . Furthermore , the total power of hubs was significantly higher than that of non-hubs ( p<0 . 0001 ) . Figure 1B shows the same relations , but now plotted for all regions , and for three different initial coupling strengths . When S = 1 . 5 , the correlations between structural degree and spike density ( r = 0 . 35 ) and structural degree and total power ( r = 0 . 94 ) indicate that especially the link between structural degree and total power is strong . For higher coupling strengths between the NMMs ( S = 2 . 0 ) , a strong positive correlation between structural degree and spike density was observed as well ( r = 0 . 86 ) . Thus , although coupling strength has an influence on these results , overall the positive relation between structural connectivity and neuronal activity is apparent . Since activity level might also be influenced by a nodes functional role rather than its structural connectivity status , we performed comparisons between structural and functional degree ( sum of all weighted functional connections of a node ) of all nodes for the common frequency bands ( delta 0–4 Hz , theta 4–8 Hz , lower alpha 8–10 Hz , higher alpha 10–13 Hz , beta 13–30 Hz , gamma 30–45 Hz ) . Results of this analysis and of direct comparisons between functional degree and neuronal activity are reported in Text S1 section 1 . In most bands , clear positive correlations were found , demonstrating that functional hub regions generally have high neuronal activity levels as well . Table 1 shows all 78 regions ranked by structural degree , with their functional degree , total power and spike density levels .
Our first aim was to find out whether the level of activity in a region is related to its degree of structural connectivity . An expected positive correlation was indeed found in repeated experiments across all degrees of connectivity ( see figures 1 , 3 , and 4 ) : structural hub regions possess the highest average power and spike densities . As can be judged from figure 1 , an exception is the relation between structural connectivity and spike density for low values of NMM coupling ( S ) . This result indicates that the relation between connectivity and activity might be more complex than we expected . Nevertheless , similar analysis performed using functional connectivity results ( see figure S1 ) led to clear positive correlations in the large majority of cases . It should further be noted that there is no unique definition of hub status , and in this experiment ( and the rest of the study ) we adhered to the pragmatic choice of taking a selection of nodes ( n = 13 ) with the highest structural degree . However , since connectivity and activity are clearly positively related in regions with higher structural degrees , we do not believe that a different hub definition would have led to a different interpretation . Still , although high neuronal activity in hub regions was a solid finding that might have been expected intuitively , it should ultimately be verified in experimental data . As can be judged from table 1 , many Default Mode Network ( DMN ) -related regions possess a high degree of connectivity and activity . The well-documented high resting-state activity level of the DMN is therefore in line with our findings [5]; however , instead of being attributed to ongoing cognitive processing or mental phenomena like introspection , high resting-state activity in the DMN might actually be ( partially ) explained by the underlying degree of structural and functional connectivity Based on the findings in our first experiment , we expected that ADD would probably preferentially target hub regions , since they possessed the highest level of activity . Analyses of both structural and functional connectivity changes due to ADD seem to be in agreement with this expectation ( see figures 2–5 ) . Furthermore , total ( or absolute ) power decreases rapidly , largely accounted for by weakening of hub regions , and the initial correlation between degree and power is lost ( figure 3 ) . Thus , large-scale brain connectivity loses its efficient ‘hub’ topology in ADD , like in AD . Surprisingly , the steady loss of power is accompanied by an initial rise of spike density on average ( see figure 4 ) , before a final oscillatory slowing sets in . This effect is stronger in hubs; spike density rises more quickly , reaches its peak rate sooner , and seems to slow down more rapidly . One explanation for the increase in spike density observed in our results is neuronal disinhibition . In fundamental neuroscience disinhibition is a well-known phenomenon and it is widely accepted that inhibitory interneurons have a large influence on oscillatory behavior [40] . Besides damaging excitatory connections , ADD impairs connectivity to and from inhibitory neurons within the neural masses , and the resulting loss of inhibition seems to be a dominant influence on spike density in the first stage . This then leads to a vicious spiral of increasing activity , more activity-dependent damage , etc . until the weakening network can no longer support an increase in spike density ( the inter-mass excitatory coupling weakens substantially , which leads to breakdown of the system , see also figure 6 ) . The eventual spike density decrease due to ADD resembles the oscillatory slowing known from AD neurophysiologic literature [41] , [42] . Several authors have argued for a prominent role of neuronal disinhibition in AD pathophysiology: for example , Gleichmann et al . propose a process they call ‘homeostatic disinhibition’ , which is based on a different underlying mechanism but might explain the higher prevalence of epilepsy that is seen in AD , reduced gamma band activity , and , interestingly , the increase in neuronal activity as measured by fMRI [43] . Schmitt argues that AD is accompanied by a loss of inhibition that leads to alterations in calcium homeostasis and excitotoxicity , respectively [44] . Olney et al . hypothesize that a disinhibition syndrome caused by hypoactive NMDA receptors triggers excitotoxic activity and widespread neurodegeneration [45] . Palop & Mucke suggest that amyloid itself causes dysfunction of inhibitory interneurons causing an increase in neuronal activity [46] , [47] , possibly also accounting for the higher prevalence of epileptic activity in AD [48] . Kapogiannis & Mattson review reports that in aging excitatory imbalance is due to a decrease in GABA-ergic signaling , and that this mechanism is exacerbated in AD [19] . An early but transient rise was also found in functional connectivity results ( see figure 5 ) , and interestingly , this is in line with experimental data of Mild Cognitive Impairment ( MCI ) patients , where increased functional connectivity levels are often interpreted as a compensatory mechanism [49]–[52] . However , this increase of functional connectivity has not been directly related to cognitive improvement , and according to our model , it might well be a part of the degeneration process itself . Finally , the ADD induced changes in functional network topology , such as the weakening of small-world structure and modularity ( see figure 5 ) , are in line with recent findings in resting-state EEG and MEG studies in AD [14] , [39] , [53]–[55] . In recent years , brain disconnectivity and disturbed network topology has been observed in an increasing number of disorders ( for example schizophrenia , multiple sclerosis , brain tumor , autism , epilepsy ) [56]–[59] . It is conceivable that different disease mechanisms and types of network damage ( for example extensive non-hub network damage ) could lead to a similar situation of hub overload and decay . Computational models like the one described here could be employed to investigate various underlying pathologies and to examine the differences between them . Several recent studies support the notion that node properties such as degree and centrality may play a crucial role in the pathophysiology of degenerative brain disease [60]–[62] . The results of this study suggest that hub regions are vulnerable due to their intrinsically high activity level . The assumption of activity dependent degeneration leads to hub vulnerability along with many neurophysiologic features of AD ( i . e . as found in quantitative EEG and MEG literature ) . A recently conducted large fMRI study demonstrated that highly connected cortical regions like the precuneus are even stronger hubs in females than in males: could this perhaps explain the higher levels of early amyloid deposition ánd the higher prevalence of AD in women [63] , [64] ? The computational model used in this study offers a possible mechanism by which excessive neuronal activity in hubs might lead to the observed macro-scale disruption of brain connectivity and dynamics in AD . In addition to the presumed role of disinhibition mentioned in the previous paragraph , a prominent role of excessive neuronal activity in AD pathogenesis has been suggested before: several studies have demonstrated a direct link between neuronal activity and the development of amyloid plaques in transgenic mice [20] , [21] , [22] . Regions that are most active during resting-state show the most outspoken AD-related pathology [4] , [5] , [13] . Excessive hippocampal activity is related to cortical thinning in non-demented elderly persons , is present in MCI patients , and is related to neurodegeneration in AD [49] , [65] , [66] . Finally , known risk factors for AD such as genetic profile , age , vascular damage , or common comorbidities like sleep disorders and epilepsy , all predispose to excessive activity and a subsequent burden on metabolism and plasticity [17] , [18] , [66]–[68] . On the other hand , protective factors like high level of education and sustained cognitive activity might relieve the burden on hub regions due to frequent activation of task-related circuits , and accompanying DMN deactivation . Summarizing , vulnerability of cortical hub regions due to their high activity levels may be aggravated or alleviated by the presence of one or more predisposing or protective factors , respectively ( see figure 7 ) . This line of reasoning implies that changes in brain activity and connectivity are already involved in the very early stages of AD pathology . In this regard , it is interesting to note that an increasing number of studies show that changes in activity and functional connectivity can be detected before cognitive complaints arise or pathological levels of amyloid are detected with PET and CSF analysis [18] , [69]–[73] . Although activity dependent degeneration is quite different from amyloid-induced damage , they need not be mutually exclusive: chronic , excessive activity might lead to amyloid deposition , which in turn causes aberrant activity and neuronal damage: a pathological cycle with different stages ( see also figure 6 ) . Relatively small increases of extracellular amyloid-beta can increase neuronal activity , especially in neurons with low activity , whereas higher levels cause synaptic depression [74] , [75] . Palop and Mucke emphasize the role of inhibitory interneuron dysfunction , leading to hypersynchronization [47] . In conclusion , although these studies provide compelling evidence for a prominent role of neuronal activity , our predictions that hub regions might form the weakest links in AD pathogenesis should be tested in further studies . Several recent studies use similar computational modeling approaches to examine AD related neurophysiological phenomena: Bhattacharya et al . focus on thalamo-cortico-thalamic circuitry and its relation with alpha band power in AD [38] . By varying the synaptic strengths in the thalamic module of the model they find that especially the connectivity of synaptic inhibitory neurons in the thalamus has a large influence on alpha power and frequency . Pons et al . use a neural mass model and human EEG data to investigate the influence of structural pathways on functional connectivity in the aging brain and pre-clinical stages of AD [37] . Findings in line with our present results are the higher functional connectivity values in MCI and the relation between structural and functional connectivity . An increase in functional connectivity and network randomness during a memory task was found by Buldú et al . in a MEG study of MCI patients [76] . Interestingly , the authors also provide a network degeneration model which might explain these observations . The combination of neural mass modeling and graph theory was used in a recent study from our group [36] . This study explores the manifestation of modularity in developing networks and investigates the effect of more acute lesions on network dynamics . The gradual recovery of functional network characteristics that was observed after lesions raises the question whether and to what extent similar mechanisms play a role in neurodegenerative damage; this should be subject of further study . To describe functional network modularity , the same algorithm and heuristic was used as in the present study . The computational models used in these studies provide a framework to address different questions and hypotheses concerning brain disease , e . g . different functional lesions . A novel aspect of the approach in the present study is that a single hypothesis ( ADD ) is proposed as main pathophysiological mechanism of AD . Comparison to a ‘random degeneration’ ( RD ) model provides further support for the ADD hypothesis , but does not rule out the possibility that other plausible degenerative models exist . Various methodological choices might have affected our results , and should be taken into account when interpreting them . First , although the DTI-derived connectivity matrix that served as the basis of our model is in our opinion a solid overall large-scale representation of human cortical connectivity , it was based on data of healthy young adults [24] . Since AD mainly affects the aging population , and since it has been shown that structural connectivity is altered during aging [77] , results might have been different if structural connectivity data of older subjects had been implemented . However , the major hub regions seem largely independent of age , justifying our approach that mainly focuses on hub versus non-hub differences . Furthermore , we now know that AD affects many people below the age of 65 , and that AD pathology is presumably already present for decades before initial symptoms appear . In a similar way we expect that individual variability in structural connectivity will not have had a major influence on our present approach , since major hub regions appear to be consistent across studies [3] , [64] . Although the computational model used here could be refined in many ways , e . g . by implementing a larger number of regions , assigning different weights to structural connections , using DSI-derived data , correcting for spatial scale and/or DTI biases , or by using more elaborate and detailed graph analysis , we believe that this would not have affected our main outcome dramatically , since comparing characteristics of hub and non-hub cortical regions does not necessarily require a high level of detail . By keeping the model and hypotheses as simple as possible , it might be easier to discover or test underlying basic principles and mechanisms of degeneration . The main motivation to use an NMM network of this size was the observation that topographical maps and atlases of the human cerebral cortex of this order of magnitude are quite common in macroscopic structural and functional connectivity studies ( for an overview , please refer to [39] , [56]–[58] . Also , since EEG and MEG studies have comparable network sizes ( 21–300 sensors ) , this is a fairly realistic spatial resolution for NMM-generated dynamics . Two relevant references are recent computational modeling studies by Deco et al . [27] and Pons et al . [37] . Varying the structural coupling strength S in our neural mass model can lead to different results , and therefore we have reported its influence on our outcomes . Similarly , the arbitrary ‘loss’-rate parameter of the ADD function will affect the speed of the degeneration process . However , since we were mainly concerned with a topological ‘hub versus non-hub’ comparison , the absolute rate of degeneration was of minor importance for this study . Moreover , loss-rates were equally applied to all connections; network distribution was not selectively influenced by these parameters . Observations from this study that could be explored further include ADD-induced changes in structural network topology , the relation between spike density and anatomical region , and the lower alpha peak frequency in hub regions ( see Text S1 section 4 ) . Predictions from our model , especially the close link between local neuronal activity and large-scale connectivity should be verified in longitudinal clinical studies , preferably of normal aging as well as patients with subjective memory complaints ( SMC ) , Mild cognitive impairment ( MCI ) and AD . To assess structural and functional connectivity as well as large-scale neuronal activity , a combination of DTI and MEG might be the most appropriate method . Source space analysis of MEG data may help to develop topologically accurate neural mass models . On a fundamental level , the relation between neuronal connectivity , activity and pathology should be further explored in animal models . Interestingly , the relation between regional activity and large-scale functional connectivity has recently also addressed with respect to schizophrenia [78] , [79] . In both studies it is argued that more knowledge of this relation is essential for understanding mechanisms of altered functional connectivity , and this is very much in line with the main message of this study . Different disease conditions may have specific causes or patterns in which this relationship is harmed , but at the same time universal principles may apply that can help us gain more insight in a range of neuropsychiatric disorders . In this study we used a neural mass model with DTI-based human topology to demonstrate that brain hub regions possess the highest levels of neuronal activity . Moreover , ‘Activity dependent degeneration’ ( ADD ) , a damage model motivated by this observation , generates many AD-related neurophysiologic features such as oscillatory slowing , disruption of functional network topology and hub vulnerability . Early-stage , transient rises of firing rate and functional connectivity in ADD matches observations in pre-clinical AD patients , suggesting that this chain of events is not compensatory , but pathological . Overall , the results of this study favor a central role of neuronal activity and connectivity in the development of Alzheimer's disease .
We used a model of interconnected neural masses , where each neural mass represents a large population of connected excitatory and inhibitory neurons generating an EEG or MEG like signal . The model was recently employed in two other graph theoretical studies [30] , [36] . The basic unit of the model is a neural mass model ( NMM ) of the alpha rhythm [26] , [80] , [81] . This model considers the average activity in relatively large groups of interacting excitatory and inhibitory neurons . Spatial effects ( i . e . distance ) are ignored in this model; brain topology is introduced later by coupling several NMMs together . The average membrane potential and spike density of the excitatory neurons of each of the NMMs separately were the multichannel output related to neuronal activity that was subject to further analysis . All neural mass model parameters and functions are summarized and explained in Text S1 , section 3 ( see also figure S4 and table S1 ) . A diffusion tensor imaging ( DTI ) based study by Gong et al . published in 2009 that focused on large-scale structural connectivity of the human cortex resulted in a connectivity matrix of 78 cortical regions [24] , [82] . The connectivity matrix was implemented in our model software , and used as topological framework for the 78 coupled NMMs . Coupling between two NMMs , if present , was always reciprocal , and excitatory . Note that at the start of the simulation , the coupling strength between all NMM pairs ( S ) was identical , and the only difference between the cortical regions ( or NMMs ) was their degree of connectivity to other neural masses ( cortical regions ) . Since the coupling strength S was an arbitrarily chosen parameter , repeated analyses were performed with different values of this variable ( see for example figure 3 ) . For the present study the model was extended to be able to deal with activity dependent evolution of connection strength between multiple coupled NMMs . Activity dependent degeneration ( ADD ) was realized by lowering the ‘synaptic’ coupling strength as a function of the spike density of the main excitatory neurons . For each neural mass the spike density of the main excitatory population is stored in a memory buffer that contains the firing rates of the last 20 steps in the model . Step size depends on the sample frequency . At each new iteration , the highest spike density value of the last 20 sample steps is determined and designated as maxAct . From maxAct a loss is determined according to the following relation: ( 1 ) Since maxAct is non-negative , loss will be a number between 0 and 1 . Next , the coupling values C1 ( connections between main excitatory population and inhibitory population ) , C2 ( connections between inhibitory population and main excitatory population ) , Pt ( thalamic input to main excitatory population ) and S ( structural coupling strength between neural masses ) are all multiplied by loss to obtain their new lower values . To assess the specificity of ADD , results were compared with a random degeneration ( RD ) model in which the maxAct variable was discarded , so damage was equally applied to all regions , regardless of their level of activity . The effects of ADD and RD were measured by changes in total power ( local average membrane potential ) and spike density , and these two measures were used as representations of neuronal activity in further analyses . Note that the time scale of the data generated by the model is equal to normal brain activity and EEG/MEG data , but that the ADD and RD procedures have a more abstract time scale . The exact duration of the degenerative procedures was not considered relevant to our present focus on the relation between connectivity and activity , but could be considered to reflect a length that is representative of a neurodegenerative process , spanning years to decades ( see figures 3–5 ) . The computational model was programmed in Java and implemented in the in-house developed program BrainWave ( v0 . 9 . 04 ) , written by C . J . Stam ( latest version available for download at http://home . kpn . nl/stam7883/brainwave . html ) . Since spectral analysis is a common neurophysiological procedure that provides clinically relevant information in neurodegenerative dementia , we included this in our experiments . Fast Fourier transformation of the EEG-like oscillatory output signal was used to calculate for all separate regions the total power ( absolute broadband power , 0–70 Hz ) as well as the absolute power in the commonly used frequency bands delta ( 0 . 5–4 Hz ) , theta ( 4–8 Hz ) , lower alpha ( 8–10 Hz ) , higher alpha ( 10–13 Hz ) , beta ( 13–30 Hz ) and gamma ( 30–45 Hz ) . To quantify large-scale synchronization as a measure of interaction between different cortical areas , we used the Synchronization likelihood ( SL ) , which is sensitive to both linear and non-linear coupling [83] , [84] . SL was calculated for all frequency bands , and the matrix containing all pairwise SL values served as the basis for all further graph theoretical analyses of functional network characteristics . Graph theoretical properties of the structural DTI network that were relevant for our hub study such as node degree , betweenness centrality , and local path length were published in the original article by Gong et al [24] . One new measure we introduced was the ‘normalized node strength’ , which is the ratio of the structural degree of a node after activity dependent damage over its original degree . This measure was used to track structural connectivity loss and to compare the loss of degree in hubs and non-hubs . For functional network analysis , connectivity matrices were subjected to topographical analysis . The functional degree of a node is defined as the sum of all its link weights [85] . Averaging the functional degree over all nodes gives the overall functional degree of a network . To match the functional network to the given structural network ( minimizing effects of graph size and density ) , we constructed a binary , unweighted matrix that was obtained after using a threshold that resulted in a network with an average degree of 8 , close to that of the structural network ( which was 8 . 1 ) . All graph theoretical measures used in this study are summarized in table 2 , for more exact definitions please refer to [14] , [85] . For functional modularity analysis , we used Newman's modularity metric combined with a simulated annealing process ( previously described in [55] , [86] ) . For the baseline , pre-ADD analysis in experiment 1 and 2 , the data-generating procedure using the model was repeated twenty times to obtain a representative amount of data . On each run the subsequent spectral , functional connectivity and graph theoretical analysis was performed , and then all results of these twenty runs were averaged prior to further statistical analysis . Regional results were visualized using 6 bins ascending in structural degree , each containing 13 regions . All 13 regions in the bin with the highest mean degree were classified as hubs . Standard deviations of bins are displayed as error bars . For bivariate correlations Pearson's test was used .
|
An intriguing recent observation is that deposition of the amyloid-β protein , one of the hallmarks of Alzheimer's disease , mainly occurs in brain regions that are highly connected to other regions . To test the hypothesis that these ‘hub’ regions are more vulnerable due to a higher neuronal activity level , we examined the relation between brain connectivity and activity in a computational model of the human brain . Furthermore , we simulated progressive damage to brain regions based on their level of activity , and investigated its effect on the structure and dynamics of the remaining brain network . We show that brain hub regions are indeed the most active ones , and that by damaging networks according to regional activity levels , we can reproduce not only hub vulnerability but a range of phenomena encountered in actual neurophysiological data of Alzheimer patients as well: loss and slowing of brain activity in Alzheimer , loss of synchronization between areas , and similar changes in functional network organization . The results of this study suggest that excessive , connectivity dependent neuronal activity plays a role in the development of Alzheimer , and that the further investigation of factors regulating regional brain activity might help detect , elucidate and counter the disease mechanism .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"neural",
"networks",
"neuroscience",
"mathematics",
"computational",
"neuroscience",
"alzheimer",
"disease",
"biology",
"dementia",
"nonlinear",
"dynamics",
"central",
"nervous",
"system",
"neurological",
"disorders",
"neurology",
"neurophysiology"
] |
2012
|
Activity Dependent Degeneration Explains Hub Vulnerability in Alzheimer's Disease
|
Brain metastases are prevalent in various types of cancer and are often terminal , given the low efficacy of available therapies . Therefore , preventing them is of utmost clinical relevance , and prophylactic treatments are perhaps the most efficient strategy . Here , we show that systemic prophylactic administration of a toll-like receptor ( TLR ) 9 agonist , CpG-C , is effective against brain metastases . Acute and chronic systemic administration of CpG-C reduced tumor cell seeding and growth in the brain in three tumor models in mice , including metastasis of human and mouse lung cancer , and spontaneous melanoma-derived brain metastasis . Studying mechanisms underlying the therapeutic effects of CpG-C , we found that in the brain , unlike in the periphery , natural killer ( NK ) cells and monocytes are not involved in controlling metastasis . Next , we demonstrated that the systemically administered CpG-C is taken up by endothelial cells , astrocytes , and microglia , without affecting blood-brain barrier ( BBB ) integrity and tumor brain extravasation . In vitro assays pointed to microglia , but not astrocytes , as mediators of CpG- C effects through increased tumor killing and phagocytosis , mediated by direct microglia-tumor contact . In vivo , CpG-C–activated microglia displayed elevated mRNA expression levels of apoptosis-inducing and phagocytosis-related genes . Intravital imaging showed that CpG-C–activated microglia cells contact , kill , and phagocytize tumor cells in the early stages of tumor brain invasion more than nonactivated microglia . Blocking in vivo activation of microglia with minocycline , and depletion of microglia with a colony-stimulating factor 1 inhibitor , indicated that microglia mediate the antitumor effects of CpG-C . Overall , the results suggest prophylactic CpG-C treatment as a new intervention against brain metastasis , through an essential activation of microglia .
Ten to twenty percent of cancer patients develop brain metastases , commonly as the final stage of cancer progression , with lung and melanoma cancers having the highest incidence ( 40%–50% and 30%–50% , respectively ) [1] . Therapies include surgery and radiation; however , both treatments result in only a modest survival advantage and are associated with cognitive impairments [2] . Chemotherapy is often inefficient due to impermeability of the blood-brain barrier ( BBB ) [1] , and as it often induces astrocyte-derived tumor-protecting responses [3] . Overall , the efficacy of currently available treatments for brain metastasis is extremely limited , making it a deadly disease with a short survival period [2] . Thus , prophylactic approaches against the establishment of brain metastasis , or early elimination of brain micrometastases , could prove key in treating cancer [2 , 4 , 5] , and even more so given ongoing progression in early cancer detection and prevention of peripheral metastases . In recent years , immune modulation using toll-like receptor ( TLR ) agonists has been given much attention as a therapeutic approach against primary tumors and metastasis [6] . Specifically , the TLR9 agonists CpG-oligodeoxynucleotides ( ODNs ) are being explored in a wide range of tumor types , both as single agents and as adjuvants [7 , 8] , and are being tested in several clinical trials [9] . In various animal models , CpG-ODN treatment was shown to reduce mammary lung metastases by eliciting antitumor natural killer ( NK ) activity [9] , and even results in rapid debulking of large tumors by macrophage stimulation [10] . Employed prophylactically , CpG-ODNs were shown to markedly improve resistance to experimental and spontaneous peripheral metastasis of mammary [11] , colon [12] , and melanoma [13] tumors . Given the low success rate of treatments against established brain metastases [1] , prophylactic treatment against metastatic brain disease may be key to improve survival rates [4] . Such treatment should be given chronically between primary tumor diagnosis and until several days/weeks following tumor removal . This time frame includes the short perioperative period , which was shown to constitute a high-risk period for initiation or accelerated progression of metastasis [14] . Prophylactic treatment should be especially advantageous in patients with primary tumors that have high potential of developing brain metastases , such as lung , melanoma , and breast cancers [15] . In fact , the concept of prophylactic treatment against brain metastasis is not unprecedented and is routinely practiced in the clinic . Small-cell lung cancer ( SCLC ) patients without detectable brain metastases often undergo prophylactic whole brain radiation therapy , thereby reducing occurrence of brain metastases and improving survival [16 , 17] . However , to implement a prophylactic approach against brain metastases in a wider range of patients , a less toxic [18] treatment is required . TLR9 stimulation using CpG-ODNs is particularly well suited to meet this need , as it has negligible toxicity in humans [19–21] , and has already promising preclinical outcomes in other organs [10–13]; therefore , it should also be considered a potential prophylactic approach against the establishment of brain metastases . In the brain , TLR9 is expressed on neurons , astrocytes , microglia , and endothelial cells [22 , 23] . Recent studies suggest that TLR9 signaling plays a key role in cerebral ischemia [24] , cerebral malaria [25] , Alzheimer’s [26 , 27] , and seizures [28] , pointing to its key role in healthy brain function and neuro-immune modulation . Notably , intracerebral [29 , 30] and retro-orbital [31] administrations of CpG-ODNs were shown to hinder growth of glioma [31] and intracranially injected melanoma cells [29 , 30] . Importantly , CpG-ODN yielded promising initial outcomes with minimal toxicity in a few Phase I/II clinical trials of recurrent [20 , 21] and de novo [19] glioblastoma , when injected into tumor-excised lesions . However , as a prophylactic measure against potential brain metastasis , CpG-ODNs would need to be administered systemically , provided they can cross the BBB . Here , we assessed the efficacy of a systemic administration of CpG-C as a prophylactic treatment for brain metastasis using three preclinical mouse models , including experimental metastasis of syngeneic and of human lung carcinoma , and spontaneous metastasis of syngeneic melanoma . Importantly , the inoculation methodologies and imaging approaches implemented here preserve intact both the neuro-immune niche and brain hemodynamics intact: crucial factors affecting metastatic early stages [32] . We demonstrate that acute and chronic prophylactic treatments result in reduced brain metastasis in both sexes and across ages . Notably , we found that NK cells and monocytes/macrophages do not take part in the initial steps of the metastatic process in the brain , nor do they mediate the effects of CpG-C , as opposed to their role in peripheral organs . We establish that peripherally administered CpG-C crosses into the brain parenchyma without affecting BBB permeability , and that cerebral endothelial cells , astrocytes , and microglia take it up . We found that CpG-C–activated primary microglial cells and the N9 microglial cell line ( but not primary astrocytes ) eradicate tumor cells in vitro through direct contact , by increasing microglia cytotoxicity and phagocytic potential . Importantly , we demonstrate in vivo that following systemic CpG-C treatment microglia cells contact , kill , and phagocytize tumor cells during the early stages of invasion into the brain . Blocking microglia activation or depleting them abolished the beneficial effects of CpG-C . Taken together , our results point to CpG-C as an important potential prophylactic treatment against brain metastasis through direct activation of microglia .
Cultures were subjected to 100 nM/L of CpG-C or non-CpG ODN ( control ) for 24 hours , and media was harvested for conditioned-media experiments ( Fig 5B , 5D and 5G ) . Cultures were washed twice and supplied with fresh media . For contact cocultures experiments ( Fig 5A , 5C , 5E and 5H ) , D122 cells were plated on top of the microglia cultures for six hours , following which cell-lysis ( primary cultures; Fig 5A and 5C ) , bioluminescence imaging ( N9 cultures; Fig 5E ) , or FACS analysis ( Fig 5H ) assays were conducted . For no-contact cocultures ( Fig 5I ) , D122 cells were plated on 13-mm coverslips and placed on top of 2-mm-thick custom-made polydimethylsiloxane 11-mm rings over the microglial cultures ( sharing the same media for six hours ) . For conditioned-media experiments ( Fig 5F ) , D122 cultures were washed and supplied with fresh or conditioned media harvested from microglia or astrocyte cultures , for six hours . Standard cytotoxicity assay was conducted as previously described [36] . Briefly , we used two concentrations of 125IUDR-labeled D122 cells in 12-well plates ( 1 × 104 and 2 × 104 cells/well ) . The radioactive signal from the media was measured using a gamma counter ( 2470 , PearkinElmer ) . Specific killing was calculated as follows: [samplerelease−spontaneousreleasemaximalrelease−spontaneousrelease]×100 N9 cells were plated in 24-well plates ( 40 × 103 cells/well ) and treated with CpG-C or non-CpG ODN for 24 hours . We used two concentrations of Luc2-labeled D122 cells in 24-well plates ( 1 . 6 × 104 and 3 . 2 × 104 cells/well ) . D-luciferin ( 30 mg/mL , 10 μL ) was mixed in each well and the bioluminescence signal was immediately measured for two minutes using Photon Imager and analyzed with M3 Vision ( Biospace Lab ) . Lysis and bioluminescence assessments were repeated in at least three separate experiments , each one conducted in quadruplicates or more . Cocultures of N9 and D122 cells were stained for annexin V ( 88-8005-72 , eBioscience ) , as per manufacturer’s instructions . We quantified the percent of annexin V–positive ( apoptotic ) cells from all mCherry-positive ( D122 ) cells using FACScan ( Becton Dickinson ) . N9 cells were plated in 96-well plates ( 30 × 103 cells/well ) and treated with CpG-C or non-CpG ODN for 24 hours . Cultures were washed twice and plated with pHrodo Red Zymosan Bioparticles ( Thermo Fisher Scientific ) conjugate for phagocytosis , according to the manufacturer’s instructions . These particles become fluorescent only after phagocytized into the lysosomes . Fluorescence was measured with Synergy HT ( BioTek ) microplate reader at 545/585 ( Ex/Em ) every 30–60 minutes thereafter ( up to six hours ) . The maximum difference between experimental groups was used for statistical analysis . N9 cells were plated in 96-well plates ( 30 × 103 cells/well ) and treated with CpG-C or non-CpG ODN as above . Plates were washed and fresh media was added . Confluent cultures were scratched ( 700 μm ) using the IncuCyte Zoom system ( Essen BioScience ) , washed , and imaged once every 2 hours for 28 hours . All studies were approved by the Tel Aviv University ( protocols numbers 04-16-039 , 10-19-002 , 10-13-002 , and 10-13-015 ) and Columbia University ( protocol number AAAX3452 ) corresponding ethics committees for animal use and welfare and were in full compliance with IACUC guidelines . Male and female mice were used . Animals were housed under standard vivarium conditions ( 22 ± 1°C , 12-hour light/dark cycle , with ad libitum food and water ) . Anesthesia was first induced by 5% isoflurane and then maintained on 1 . 5%–2% throughout the procedures . When anesthetized , core body temperature of animals was maintained at 37°C . Animals placed with a cranial window received a single injection of carprofen ( 5 mg/kg; intraperitoneal [i . p . ] ) following the surgical procedure . Animals losing 10% of body weight were euthanized . For euthanasia of animals , we used excess CO2 or sodium pentobarbital ( 200 mg/kg; i . p . ) . Tumor cells were injected using the assisted external carotid artery inoculation ( aECAi; Fig 1A ) , as previously described [32] . Briefly , mice were anesthetized and the external carotid artery ( ECA ) uncovered . A 6–0 silk-suture ligature was loosely placed around the ECA proximal to the bifurcation from the common carotid artery ( between the superior thyroid artery and the bifurcation ) . A second ligature was tied on the ECA distal to the bifurcation . A NANOFIL-100 ( WPI ) syringe with a 34G beveled needle was mounted to a micromanipulator ( M33 , Sutter ) . The needle was inserted slowly into the lumen of the ECA and advanced to the point of bifurcation . The first ligature was tied around the needle , and 1 × 105 cells in PBS ( 100 μL ) were slowly infused into the internal carotid artery . The needle was then removed , the ligature quickly tied , and the skin sutured . Total time for the complete procedure is 10 minutes . For a spontaneous brain metastasis model ( Fig 1G ) , we used the Ret-melanoma model we have recently established and validated [33] . Briefly , mice were anesthetized by isoflurane , and a total of 5 × 105 ( 50 μL ) Ret-mCherry–sorted ( RMS ) cells in a 1:1 suspension of PBS with growth factor–reduced Matrigel ( 356231 , BD Biosciences ) were injected subdermally to the right dorsal side , rostral to the flank , with a 29G insulin syringe ( BD Biosciences ) . Tumors were measured four times weekly by calipers . Tumor volumes were calculated using the formula X2 × Y × 0 . 5 ( X = smaller diameter , Y = larger diameter ) . We aimed to remove the tumor at a size of approximately 500 mm3 . Therefore , and based on our experience , once tumors reached a size of about 125 mm3 , mice were given two injections of CpG-C or PBS ( control group ) every other day . One day later ( i . e . , three days following the first CpG-C treatment and one day following the second CpG-C treatment ) , tumor sizes were verified ( meeting our expectations , with no differences between treatment groups ) and immediately removed . The last tumor removal was carried out six days after the first removal . For tumor excision , mice were anesthetized with isoflurane , and an incision , medial to the tumor , was made in the skin . Tumors were detached from inner skin with clean margins to prevent recurrence . Tumor-associated connective tissue and blood vessels were detached , and the incision was sutured . Primary tumors were sectioned and measured with no difference identified at excision time ( S2 Fig ) . Mice were weighed weekly and monitored for relapse . Nine weeks after last tumor excision , mice were deeply anesthetized with isoflurane and transcardially perfused with cold PBS . Brains and lungs were harvested , macroscopically examined for abnormal lesions , and flash-frozen in liquid nitrogen . RNA was isolated using EZ-RNA II kit ( 20-410-100 , Biological Industries ) according to the manufacturer’s instructions . Whole organs were homogenized in denaturation solution A in M tubes ( 130-096-335 , Milteny Biotec ) by gentleMACS dissociator ( Milteny Biotec ) . Reverse transcription was performed with qScript ( 95047–100 , Quanta Biosciences ) . RT-qPCR analyses were conducted using PerfeCTa SYBR Green FastMix , ROX ( 95073-012-4 , Quanta Biosciences ) with primers for Hprt ( F sequence: GCGATGATGAACCAGGTTATGA; R sequence: ATCTCGAGCAAGTCTTTCAGTCCT ) and mCherry ( F sequence: GAACGGCCACGAGTTCGAGA; R sequence: CTTGGAGCCGTACATGAACTGAGG ) . In all analyses , expression results were normalized to Hprt . RQ ( 2−ΔΔCt ) was calculated . Of the 50 animals initially injected with tumor cells , two animals did not develop primary tumors and were withdrawn from the experiment; of the remaining 48 animals , 28 animals were treated with CpG-C and 20 with PBS ( control ) . Twenty-two animals ( 45% of control and 46% of CpG-C treated ) died during the period between tumor excision and the day of tissue collection , leaving 15 CpG-C–treated animals and 11 control animals . In three CpG-C animals and two control animals , we did not detect mCherry RNA in the brains . The herein development of primary tumor and metastases is expected based on our previous studies in this tumor model [33] . Tumor burden was compared in animals bearing brain micrometastasis . CpG-C , CpG-C-FITC , and CpG-C-TAMRA ( ODN 2395: 5′-TCGTCGTTTTCGGCGCGCGCCG-3′ ) with a phosphorothioate backbone and non-CpG ODN ( ODN 2137: 5′-TGCTGCTTTTGTGCTTTTGTGCTT -3′ ) , endotoxin-free , were purchased from Sigma-Aldrich . Two different controls were used: PBS and non-CpG ODN , which lacks C-G motifs ( counterbalanced within experiments with no differences in results ) . Both CpG-C variants and non-CpG ODN were diluted in PBS and administered intraperitoneally ( 100 μL ) at a dose of 0 . 4 or 1 . 2 mg/kg ( S3 Fig ) , or 4mg/kg ( all in vivo experiments ) . No differences were found between PBS- and non-CpG ODN–treated animals , and therefore they were combined in the statistical analyses ( S7 Fig ) . For depletion of NK cells ( Fig 2A and 2B ) , anti-NK1 . 1 monoclonal antibodies ( mAbs ) were intraperitoneally administered ( 4 mg/kg ) 24 hours before tumor cell injection . mAb ( 12E7 ) against human cluster of differentiation ( CD ) 99 served as control . Antibodies [38] were kindly provided by Prof . Ofer Mandelboim ( The Hebrew University of Jerusalem , Israel ) . To verify depletion of NK cells , blood was collected from animals during tissue collection , and prepared for staining with NK1 . 1 FITC ( eBioscience ) and NKp46 PE ( BioGems ) [39] . FACS analysis indicated >90% depletion ( S4 Fig ) . For monocytes/macrophages depletion ( Fig 2C and 2D ) , we administered clodronate liposomes ( ClodronateLiposomes . org ) intravenously ( 200 μL ) 24 hours before tumor cell injection . PBS liposomes served as controls . To verify depletion of monocytes , but not microglia , blood was sampled , animals were perfused , and brains were collected . Samples were prepared for staining with F4/80 FITC and CD11b APC ( BioGems ) . FACS analysis indicated >85% depletion of monocytes/macrophages , without affecting microglia viability ( S4 Fig ) . To block microglia activation and transition into an inflammatory state [40] , minocycline hydrochloride ( Sigma-Aldrich ) was administered intraperitoneally at a dose of 40 mg/kg ( 200 μL ) at 48 , 32 , and 24 hours before tumor cell injection . For depletion of microglia cells , mice were administered the dietary inhibitor of colony stimulating factor-1 receptor ( CSF1R ) , PLX5622 ( 1 , 200 mg/kg chow; provided by Plexxikon and formulated in AIN-76A standard chow by Research Diets ) for 18 days , resulting in near-complete elimination of microglia cells [41] . AIN-76A standard chow served as control ( Research Diets ) . C57BL/6J and athymic nude mice from the bioluminescence experiments were perfused with PBS supplemented with 30 U heparin ( Sigma ) and 4% paraformaldehyde ( PFA ) ( EMS ) . Brains were harvested , fixed overnight in 4% PFA , and placed in 30% sucrose overnight . Thirty-micrometer sections ( Leica SM 2000 microtome ) were counterstained with DAPI ( MP Biomedicals ) . Images of the sections were obtained using a fluorescent microscope ( Olympus ix81; Fig 1B ) . To visualize CpG-C uptake in the parenchyma , TAMRA-labeled CpG-C was injected to CX3CR1GFP/+ mice . Twenty-four hours later , animals were perfused and brains fixed and sectioned . Astrocytes were stained using a primary anti-GFAP antibody ( 1:800; Invitrogen ) and endothelial cells with anti-CD31 ( PCAM-1 ) antibody ( 1:500; Santa Cruz ) . A secondary Alexa 647 antibody ( 1:600; Invitrogen ) was used , coupled with DAPI ( 1:1 , 000; ENCO ) staining . Images of the sections were obtained using a Leica SP8 confocal microscope at 0 . 5-μm intervals using a ×63 ( NA– 1 . 4 ) oil immersion objective . Preparation of tissue for ImageStream ( MK II; Amnis ) FACS analysis—mice were perfused with PBS supplemented with 30 U heparin ( Sigma-Aldrich ) and brains removed . Brains were mechanically minced , suspended in a solution containing collagenase ( 0 . 1% w/v; Worthington ) and dispase ( 0 . 2% w/v; Roche ) for 20 minutes and then in DNAse ( Sigma-Aldrich ) for 20 minutes , and suspensions were passed through a 70-μm filter . Fatty tissue was removed using Percoll ( Sigma-Aldrich ) , and cells were resuspended in PBS supplemented with 1% EDTA ( Sigma-Aldrich ) , 0 . 01% NaN3 ( Sigma-Aldrich ) , and 1% FBS ( Biological Industries ) . In each experiment , 1 × 104 events were collected and analyzed using Amnis IDEAS software ( Version 6 . 2 ) . Analysis gates were manually corrected based on images of the events . Internalization was quantified automatically using the software’s internalization wizard ( Figs 6F–6H and 7B; S9 Fig ) . For quantification of CpG-C infiltration into the brain and its internalization by endothelial cells , astrocytes , and microglia ( Fig 3 ) , mice were injected with FITC-labeled CpG-C 24 hours before perfusion , and single cell suspensions were stained using anti-CD31 ( PCAM-1 ) PE-Cy7 ( eBioscience ) , Anti-GLAST ( ACSA-1 ) -PE ( MACS ) , and anti-CD11b APC ( BioGems ) . To avoid GLAST staining of Bergmann glia , cerebellums were removed before preparation of the samples in this experiment . In each population , we quantified the percent of cells with internalized FITC . Cultures of N9 cells grown on coverslips were treated with CpG-C-TAMRA for 24 hours and washed three times . LysoTracker Blue DND-22 ( 50 nM , Thermo Fisher Scientific ) was applied for 30 minutes at 37°C , and coverslips were washed and mounted on slides . For staining of CpG-C uptake in vivo , a single cell suspension was prepared from CX3CR1GFP/+ mice treated with CpG-C-TAMRA as described in the ImageStream FACS analysis protocol herein . LysoTracker Blue DND-22 ( 50 nM , Thermo Fisher Scientific ) was then mixed into the cell suspension for 30 minutes at 37°C , and cells were mounted on a cover glass and imaged with a Leica SP8 confocal microscope using a ×63 ( NA– 1 . 4 ) oil immersion objective ( Fig 5B ) . Similarly , for the cells extracted from the brains of animals , we imaged only green fluorescent protein ( GFP ) –positive cells ( i . e . , microglia ) . Tg eGFP-claudin-5 [37] mice were treated with CpG-C or PBS , and 23 hours later were injected with 1% biocytin-tetramethylrhodamine ( TMR ) ( i . v . , Life Technologies ) . One hour later , animals were perfused with PBS and 4% PFA . Brains and livers were harvested , fixed for six hours in 4% PFA at 4°C , and placed in 30% sucrose overnight at 4°C . Tissues were embedded in O . C . T ( Sakura ) , sectioned ( 12 μm ) using a Leica cryostat , and stained for eGFP ( 1:1 , 000; Life Technologies ) and IgG ( 1:1 , 000; Invitrogen ) . Z-stacks of the sections were obtained with a Zeiss LSM700 confocal microscope using a water immersion ×40 objective ( NA– 1 . 2 ) , and maximum projections were created using Fiji ( version 1 . 0 ) . At least five images were used for quantification for each anatomical region . Biocytin-TMR and IgG intensity was quantified using Fiji software and normalized on fluorescence intensity in the liver ( Fig 4A and 4B and S5 Fig ) . For quantification of gaps in tight junctions ( Fig 4C and S5 Fig ) , we quantified the percentage of junctional strands showing at least one gap ( defined as a discontinuity in eGFP-Caudin5 signal >0 . 4 μm ) over the total number of junctional strands [37] . To test whether CpG-C affects immune cell infiltration into the brain , sections of PBS and CpG-C–treated mice were stained for CD68 ( 1:1 , 000; Abcam ) and CD4 ( 1:200; Abcam ) . To assure we do not analyze immune cells arrested in the vessels , we costained slices with laminin ( 1:1 , 000; Sigma ) to detect vessel walls . As a positive control we used spinal cord sections of experimental autoimmune encephalomyelitis ( EAE ) mice ( refer to [42] for experimental procedure; Fig 4D ) . To follow progression of tumor growth in vivo , we used an IVIS SpectrumCT ( PerkinElmer ) for the syngeneic model and Photon Imager ( Biospace Lab ) for the xenograft model . Briefly , mice were anesthetized and injected with D122-mCherry-Luc2 ( C57BL/6J ) or PC14-PE6-mCherry-Luc2 ( athymic nude ) cells . Imaging sessions were conducted on days 1 , 4 , 7 , 14 and 21 following tumor cell administration ( in the xenograft model , also on day 25 ) . After the last in vivo imaging session in the syngeneic model , mice were killed , and brain and extracranial head tissue were rapidly imaged , separately . Notably , tissue from one control animal was lost in the final process . Each imaging session was preformed between 10 and 20 minutes following D-Luciferin sodium salt injection ( 30 mg/mL , 100 μL , i . p . ; Regis Technologies ) , as this time frame exhibited maximal and steady intensity . Analysis was done using Living Image software ( version 4 . 3 . 1 ) for the IVIS images data , and M3 vision for the Photon Imager data . To quantify fluorescence in brains of athymic nude mice injected with PC14-PE6-mCherry-Luc2 , animals were decapitated and brains were extracted immediately following the last imaging session . We used a Maestro spectral fluorescence imaging system ( Cambridge Research and Instrumentation ) and quantified the fluorescent signal using Maestro version 2 . 2 software . Regions of interest ( ROIs ) were drawn on each fluorescent signal to quantify the area of the fluorescent signal . Mice were injected with 1 × 105 125IUDR-labeled D122-LLC cells using the aECAi approach [32] and euthanized 24 hours later . Animals were transcardially perfused with 20 mL PBS supplemented with 30 U heparin ( Sigma-Aldrich ) . Brain and lungs were collected , and radioactivity was measured using a gamma counter ( 2470 , PearkinElmer; Figs 1F–1H , 2 , 7A and 7C ) . For two-photon microscopy measurements , CX3CR1GFP/+ and WT mice were implanted with a polished and reinforced thin-skull ( PoRTS ) window over the somatosensory cortex ( window center at bregma minus 2 mm and 3 mm lateral to the midline ) , as previously described [43] . Importantly , this craniotomy does not elicit an inflammatory response [43] . Mice were then habituated to the imaging apparatus for 7 days to reduce procedural stress . CX3CR1GFP/+ animals were injected with 1 × 105 tdTomato-labeled D122 cells . Before imaging , mice were injected with Alexa Fluor 633 hydrazide ( 2 . 5% w/v , i . v . ; Invitrogen ) for visualization of arteries [40] . Imaging sessions were initiated 2–4 hours after tumor cell inoculation , and at days 1 , 2 , 4 and 7 , returning to the exact same location each session . Imaging was conducted at depths of 50–200 μm with a custom-modified two-photon laser-scanning microscope based on a Sutter MOM ( Sutter ) controlled through the ScanImage software ( Vidrio Technologies ) . A Chameleon Ultra II ( Coherent ) provided the 80-MHz , 140-fs pulsed light used for imaging and laser photodamage . For quantification of microglia-tumor cells interaction , 150-μm stacks were obtained and max projected every 10 μm . The number of contacts and internalization events in each stack were manually quantified blindly at 4 hours following tumor cell injection , and at days 1 , 4 , and 7 ( Fig 6D and 6E ) . For imaging CpG-C uptake by microglia in vivo , baseline imaging of cortices of CX3CR1GFP/+ mice was performed at 890 nm , CpG-C-TAMRA was injected ( 4 mg/kg; 100 μL; i . p . ) , and 24 hours later mice were imaged again at the exact same locations ( S6 Fig ) . For longitudinal BBB assessment ( Fig 4E and 4F ) , WT mice implanted with a PoRTS window were treated with four PBS or CpG-C injections every other day ( similarly to the spontaneous melanoma brain metastasis experiment ) . BBB leakage dynamics of a low molecular weight dye ( sodium fluorescein; NaF; 376 Da; Sigma-Aldrich ) and of a higher molecular weight dye ( Texas Red; 70 kDa; Invitrogen ) was imaged simultaneously at 940 nm . The imaging session took place at baseline ( before treatment ) , one day following the first CpG-C/control treatment , and one day following the last treatment . To this end , 10 minutes following dye injection , 100-μm stacks were taken every 10 minutes for a total of 90 minutes . For quantification of dye leakage , max-projections of each session were aligned using Fiji software ( 2 . 0 ) plugin linear stack alignment with SIFT . Eight vessels ( four capillaries 5 μm and smaller , and four vessels 20–50 μm in diameter ) were blindly selected manually , and average intensity was measured inside the vessel and adjacent to it ( in the parenchyma ) . The ratio over time between the amount of dye inside and outside the vessels was computed ( Fig 4F ) . For display purposes only , image contrast was automatically adjusted using the Fiji autoadjust display function , while measurements were taken directly from pixel values . In order to assess microglia reactivity , focal laser-induced thermal damage insults were performed as previously described [44] ( S8 Fig ) . Briefly , CX3CR1GFP/+ mice underwent craniotomy , and three weeks later microglia , were imaged at 890 nm . A baseline stack ( 0–30-μm depth ) was imaged and a small ( 15–20-μm ) localized injury was achieved by focusing a two-photon laser beam ( 780 nm; 150 mW at the sample; about 1 μm in size ) at 15-μm depth for two seconds . Stacks were imaged every two minutes for 40 minutes . Using Fiji software ( 1 . 0 ) , maximum z-projections were turned into binary images . A 60-μm circle was drawn around the ablation area , and for each time point , the number of white pixels were counted inside the small circle ( x ( t ) ) . For the baseline image , another 120-μm circle was drawn , and the white pixels in the ring between the two circles were counted ( y ( 0 ) ) . Response was calculated as follows: x ( t ) −x ( 0 ) /y ( 0 ) . In two independent experiments ( Fig 8 ) , male and female mature ( 4–6 months ) CX3CR1GFP/+ mice were treated with CpG-C or non-CpG ODN/PBS . Twenty-four hours later , mice were perfused and brains were harvested and processed into a single cell suspension as described above . GFP-positive cells ( microglia ) were sorted ( FACSAria IIU , BD Biosciences ) , and RNA was extracted with TRIzol ( Invitrogen ) . cDNA was prepared and used for quantitative PCR , and the results were normalized to Gapdh . All primers and probes were purchased from Applied Biosystems , Cd36 ( Mm00432403_m1 ) , Cd47 ( Mm00495006_m1 ) , Cd68 ( Mm03047343_m1 ) , Fas ligand ( Fasl ) ( Mm00438864_m1 ) , Gapdh ( Mm99999915_g1 ) , interferon gamma ( Inf-γ ) ( Mm01168134_m1 ) , interleukin ( Il ) 1-β ( Mm00434228_m1 ) , Il-6 ( Mm00446190_m1 ) , Macrophage receptor with collagenous structure ( Marco ) ( Mm00440250_m1 ) , nitric oxide synthase 2 ( Nos2 ) ( Mm00440502_m1 ) , transmembrane protein 119 ( Tmem119 ) ( Mm00525305_m1 ) , tumor necrosis factor ( Tnf ) ( Mm00443258_m1 ) , tumor necrosis factor superfamily member 10 ( Tnfsf10 ) ( Mm01283606_m1 ) , and triggering receptor expressed on myeloid cells 2 ( Trem2 ) ( Mm04209424_g1 ) . One control sample was removed as an outlier from statistical analysis of Tnf and Inf-γ ( 25 SEMs and 50 SEMs , respectively ) . Three-dimensional volumetric reconstruction of single cells ( Fig 3A , S6 Fig ) or fields of view ( Fig 6C ) were performed in a semi-automatic way using Amira software ( Thermo Fisher Scientific ) . Auto-thresholding mode was initially used to detect the brightest object , which , depending on the experiment and the spectral channel under analysis , represented either cell soma or aggregates of labeled CpG-C . Cell morphology was partially reconstructed by manual labeling after thresholding . Prism ( version 7 . 0c ) and Python ( version 3 . 6 . 3 ) were used for statistical analysis . When appropriate , the Kolmogorov–Smirnov normality test was used to determine normal distribution of the data , and the F-test or Brown-Forsythe test for determining homogeneity of variance . For normally distributed data with equal variance , we used one-way ANOVA ( Fig 5A–5G and 7B , S7 Fig ) , two-way ANOVA ( Figs 1C . i , 1E . i , 3B and 6E and S2 , S4 , S8 Figs ) , two-tailed unpaired Student t test ( Figs 1D , 1E . ii , 1G , 4A–4C , 5H , 5I and 6H , S1 and S2 Figs ) , or one-tailed unpaired Student t test ( Fig 8 ) to compare experimental groups . For normally distributed data with unequal variance , we used Mann–Whitney U test ( Figs 1C . ii , 1F , 6F and 6G , S1 and S3 Figs ) or Kruskal–Wallis ( S3 Fig ) to compare experimental groups . For non-normally distributed data , we used two-way permutations ( Figs 2 , 7A and 7C ) to compare experimental groups . For post hoc analysis , multiple comparisons were corrected using Dunn test , Tukey , or Bonferroni , according to the primary analysis and the software’s recommendation . For quantification of primary tumor growth dynamics ( S2 Fig ) and for longitudinal BBB leakage ( Fig 4E and 4F ) , we applied a least squares fit model of an exponential growth curve or one-phase exponential decay curve , respectively , and compared fits of the treated and control groups . The p-values smaller than 5% were considered significant . In all experiments , measurements were taken from distinct samples ( different animals for in vivo experiments and different wells for in vitro experiments ) .
To study the prophylactic efficacy of the TLR9 agonist CpG-C in reducing brain metastasis , we first employed two models of non-small-cell lung carcinoma , given the clinical prevalence of brain metastases in this type of cancer [15 , 45] . To this end , we used the highly metastatic D122 variant of the syngeneic LLC in C57BL/6 mice [32] and the human xenograft PC14-PE6 cells in athymic nude mice [46] . For exclusive injection of tumor cells to the cerebral circulation , we used a novel approach that we have recently developed and validated—the aECAi ( Fig 1A and 1B ) [32]—which results in improved targeting of tumor cells to the brain and avoids cerebral blood flow perturbations . A single prophylactic systemic injection of CpG-C was given 24 hours before tumor cell injection . Brain tumor growth was monitored thereafter using in vivo bioluminescence imaging . Animals pretreated with CpG-C displayed reduced cerebral tumor growth in both the syngeneic ( p = 0 . 0011; Fig 1C ) and the xenograft ( p < 0 . 0001; Fig 1E ) models , exhibiting a statistically significant difference starting on days 14 and 4 , respectively . At end point , signal intensity , which is indicative to the tumor burden , was 77-fold lower in the CpG-C–treated mice in the syngeneic D122 model ( n = 6; p < 0 . 0001 ) and 82-fold lower in the xenograft model ( n = 7; p < 0 . 0001 ) , compared with their matching control groups ( n = 7 in both models ) . To assure that the differences between groups originated from tumor growth within the brains , rather than from extracranial growth [32] , we harvested the brains and measured the tumor signal in both models . In the syngeneic model , brains from CpG-C–treated animals had a 48-fold lower bioluminescent signal compared with control animals ( p = 0 . 0040; Fig 1D ) . Similarly , in the xenograft model the mean area of fluorescence signal , indicative of brain tumor burden , was significantly smaller in CpG-C–treated animals ( p = 0 . 0373; Fig 1F ) . To test the efficacy of CpG-C treatment in a context that better resembles the clinical setting , we used a murine model of spontaneous brain metastasis that we have recently established [33] . In this model , mCherry-expressing Ret melanoma cells are injected orthotopically , resulting in growth of a primary tumor in the flank . During the perioperative period—three days before and after primary tumor excision—animals were treated with CpG-C ( n = 15 ) or vehicle ( n = 11 ) , with no measurable impact on primary tumor growth ( p = 0 . 6066 for tumor growth dynamics and p = 0 . 9260 for tumor size at time of excision; S2 Fig ) . Approximately nine weeks after excision of the primary tumor , brain and lung metastatic burden ( i . e . , mCherry expression ) were quantified . CpG-C treatment significantly reduced the overall metastatic burden in the brain ( n = 9 and 12 for control and CpG-C , respectively; p = 0 . 0345; Fig 1G ) . Notably , in the lungs ( as in the primary tumor ) , CpG-C treatment had no effect ( p = 0 . 7858; S2 Fig ) , suggesting that the beneficial effects of CpG-C in the brain were not secondary to generic or peripheral effects on tumor burden . These results provide direct evidence that systemic prophylactic CpG-C treatment during the perioperative period can reduce metastatic growth in the brain . In the subsequent experiments , we aimed to pinpoint mechanisms underlying the beneficial effects of CpG-C . We focused on the first 24 hours of tumor colonization in the brain , for the following reasons: ( i ) a single administration of CpG-C , which we herein found effective , is known to exert immune activation within hours and for up to 72 hours [11]; ( ii ) bioluminescence imaging indicated a nonsignificant trend for beneficial effects a day following tumor inoculation ( S1 Fig ) ; and ( iii ) tumor cells successfully proliferate to macrometastases only if they extravasate into the brain parenchyma within the first three days [46] . Therefore , to maximize our ability to focus on the first days following CpG-C administration , we administered syngeneic D122 tumor cells in C57BL/6J mice , employing the aECAi approach [32] , known for its high temporal inoculation efficiency . Importantly , as we suspected that bioluminescence was not sensitive enough , we assessed brain tumor seeding by measuring radioactive signals of isotope-labeled tumor cells within an entire excised organ—an approach that allows maximal signal-to-noise sensitivity . A single prophylactic injection of CpG-C resulted in reduced brain tumor retention similarly in males and females ( S3 Fig ) and in young , juvenile , and old mice ( S3 Fig ) . While CpG-C was effective in reducing brain tumor retention already at a dose of 1 . 2 mg/kg ( p = 0 . 0455 ) , its efficacy increased at 4 mg/kg ( p = 0 . 0003; S3 Fig ) —a dose we previously showed as beneficial in reducing peripheral metastases [39] . In the clinical setting , a prophylactic treatment should rely on a chronic schedule , and therefore , we tested whether a regime of five injections of CpG-C given every other day has similar effects as a single injection , and does not result in tolerance to the effects of the agent . Indeed , CpG-C treatment resulted in reduced tumor retention ( p = 0 . 0001; S3 Fig ) following both the acute ( n = 6; p = 0 . 0298 ) and the chronic ( n = 6; p = 0 . 0013 ) treatments , compared with control animals ( n = 6 ) . Notably , single and multiple CpG-C injections were well tolerated , as indicated by a lack of weight loss compared with control animals ( n = 6; p = 0 . 2593; S3 Fig ) , in line with previous reports [11] . These data suggest that CpG-C is efficient both as an acute and as a chronic prophylactic treatment for brain metastasis in both sexes and across ages , affecting early stages of tumor cell seeding . It has previously been shown that CpG-ODNs have beneficial effects in the periphery , reducing seeding of tumor cells and their subsequent growth . These antitumor effects were found to be mediated by NK cells [12 , 47] and macrophages [10] . To study in vivo whether these leukocytes also take part in the metastatic process in the brain and mediate the effects of CpG-C , we depleted NK cells and monocytes/macrophages using anti-NK1 . 1 and clodronate liposomes , respectively ( Fig 2 ) . In the lungs , NK depletion resulted in a 5-fold increase in tumor retention ( p = 0 . 0001 ) and partially blocked the beneficial effects of CpG-C ( p = 0 . 0038; Fig 2A ) evident in naïve animals ( p = 0 . 0019 ) ( in line with previous results [48] ) . In contrast , in the brains of the same animals , NK depletion did not affect tumor retention ( p = 0 . 3935 ) , nor did it mediate the beneficial effects of CpG-C ( p = 0 . 0811; Fig 2B ) , evident in both naïve ( p < 0 . 0001 ) and NK-depleted animals ( p = 0 . 0056 ) . Similarly , depletion of monocytes increased lung ( p = 0 . 0401; Fig 2C ) but not brain tumor retention ( p = 0 . 3081; Fig 2D ) , while the effects of CpG-C were not mediated by monocytes in the lungs ( p = 0 . 0003 ) or in the brain ( p = 0 . 0001 ) . These findings demonstrate that NK cells and monocytes/macrophages play a key role in the metastatic process in the lungs , but not in the brain , nor do they mediate the beneficial effects of CpG-C in the brain . As peripheral innate immune cells do not seem to mediate the effects of CpG-C , we turned to evaluate the role of central nervous system cells that express TLR9 [23 , 49] . We focused on cells that are known to play key roles in the metastatic process , including endothelial cells , astrocytes , and microglia [50] . First , to evaluate whether CpG-C can cross the BBB and affect cerebral components , we systemically administered mice with TAMRA- or FITC-conjugated CpG-C . Twenty-four hours later , we analyzed CpG-C uptake by brain endothelia , astrocytes , and microglia in histological sections ( Fig 3A ) and using ImageStream FACS analysis ( see methods; Fig 3B ) . Approximately 74% of endothelial cells , 58% of astrocytes , and 62% of microglia cells internalized CpG-C ( n = 4 mice; Fig 3B ) . This internalization is expected , as TLR9 ligands are internalized into the cell to bind with the endosomal receptors [51] . Indeed , lysosomal staining of microglia extracted from CpG-C-TAMRA–treated animals indicated that CpG-C is internalized into the lysosomes ( S6 Fig ) . For malignant cells to infiltrate into the brain parenchyma , they must cross the BBB . Endothelial cells connected by tight junction act as the first physical barrier , preventing uncontrolled infiltration of blood-borne cells . As endothelial cells uptake CpG-C ( Fig 3A and 3B ) , we sought to test whether it had an effect on BBB permeability and integrity . To this end , we measured biocytin-TMR and IgG infiltration and continuity of claudin-5 ( tight junctions ) in animals expressing GFP under the claudin-5 promotor . CpG-C did not affect biocytin-TMR or IgG infiltration in the brain vasculature and choroid plexus ( ≥5 images were averaged in five anatomical regions in three mice—n = 12; Fig 4A and 4B S5 Fig ) , nor continuity of claudin-5 in the brain ( Fig 4C , S5 Fig ) . Furthermore , no infiltration of immune cells ( i . e . , CD4+ or CD68+ ) into the brain or choroid plexus was evident following CpG-C treatment ( Fig 4D ) . Thus , these results strongly suggest that the effects of CpG-C on tumor seeding in the brain are not mediated by perturbations to the BBB or the choroid plexus permeability . Astrocytes [52] and microglia [53] have key roles in innate and adaptive immunity , and combined with their significant uptake of CpG-C ( Fig 3A and 3B and S6 Fig ) , they were our primary candidates for mediating the effects of this agent . Therefore , we investigated their in vitro capacity to induce tumor cell lysis and the impact of prestimulation with CpG-C . Primary astrocytic cultures were treated with CpG-C or non-CpG ODN , and tested for their ability to induce tumor cell lysis by contact or by secretion of apoptosis-inducing factors . The cultured astrocytes did not induce tumor cell death , with or without CpG-C treatment , in both contact and secretion conditions ( Fig 5A and 5B ) . In contrast , primary microglial cells induced cytotoxicity in D122 tumor cells , and CpG-C treatment markedly increased this lysis when tumor cells were in contact ( Fig 5C ) , while their conditioned media alone had no effect ( Fig 5D ) . We further extended this testing in the N9 immortalized microglia cell line . Similar to the effects observed in the primary microglia culture , N9 cells reduced tumor cell viability when in contact ( Fig 5E ) , but failed to do so in a paracrine setting ( Fig 5F and 5G ) . To study whether non-CpG ODN impacted the tumoricidal activity of N9 cells , we repeated the contact coculture experiment with an additional group of PBS-treated N9 cultures ( S7 Fig ) . We found PBS and non-CpG ODN treatments to have a similar effect ( p = 0 . 7745 and p = 0 . 1420 for 16 × 103 and 32 × 103 D122 cells/well ) , while CpG-C significantly reduced tumor cells’ viability ( for 16 × 103 , p = 0 . 0017 and p = 0 . 0062 compared with PBS and non-CpG ODN , respectively , and for 32 × 103 , p = 0 . 0003 and p = 0 . 0477 compared with PBS and non-CpG ODN , respectively ) . Next , we studied the mechanisms by which microglia cells eradicate D122 tumor cells . We found that N9 cells treated with CpG-C induced apoptosis in tumor cells , as indicated by increased annexin V staining ( Fig 5H ) . Additionally , CpG-C treatment resulted in a 3-fold elevation in phagocytosis capacity ( Fig 5I ) , in line with previous reports [27] . Notably , it appears that the effect of CpG-C on microglia activity is not a general activation , as we found no effects of the agent in a scratch migration assay [54] ( n = 9; p = 0 . 6732 for wound confluency , and p = 0 . 6039 for wound width; and also in vivo as described below , S8 Fig ) . Taken together , these findings indicate that contact between microglia and tumor cells is essential for the effects induced by CpG-C . A combination of elevated microglial cytotoxicity and enhanced phagocytic capacity underline these effects . We found that CpG-C affects brain tumor retention as early as 24 hours post–tumor cell inoculation . Interactions between microglia and tumor cells at early stages of tumor cell extravasation have been reported elsewhere [55] . However , the significance of these interactions with respect to microglial tumoricidal characteristics at this time point is yet unknown . To this end , we first established that microglia indeed phagocytize tumor cells at this early time point . Longitudinal intravital imaging revealed that microglia cells interact with tumor cells and initiate phagocytic processes as early as a few hours after tumor cell inoculation ( Fig 6A–6C ) . To assess the effects of CpG-C on this phagocytic capacity , CX3CR1GFP/+ mice were injected with CpG-C or non-CpG ODN , and 24 hours later injected with either tdTomato-labeled or mCherry-labeled D122 tumor cells for two-photon or ImageStream FACS analysis , respectively . The numbers of microglia-tumor cell contacts and microglia internalization of mCherry particles ( originated from tumor cells ) were quantified four hours following tumor cells’ inoculation and at days 1 , 4 , and 7 thereafter ( Fig 6D and 6E ) . As early as four hours following tumor cells’ inoculation , there were more contacts between microglia and tumor cells in CpG-C–treated animals ( p = 0 . 0128 ) , with no differences at later times . Moreover , the number of internalization events in CpG-C–treated animals was higher four hours ( p = 0 . 0372 ) and one day ( p = 0 . 0041 ) following tumor cell inoculation . No differences were evident at days 4 and 7 , probably due to the dismantling process of the tumor cells , evident as early as two days following tumor cell inoculation ( Fig 6A ) . Using ImageStream FACS analysis 24 hours after tumor inoculation , we found first that CpG-C did not affect the total number of microglia in the brain ( n = 5; p = 0 . 4201; Fig 6F ) nor the total number of infiltrating tumor cells ( p = 0 . 3455; Fig 6G ) , in accordance with our above findings regarding the lack of CpG-C impact on BBB permeability . However , CpG-C increased phagocytosis of tumor cells by microglia ( p = 0 . 0055; Fig 6H ) . These results alone do not specify whether CpG-C increases the killing of tumor cells by microglia or whether it merely increases endocytosis of tumor debris by microglia . To distinguish between these alternatives , we turned to a set of experiments in which microglia activation was impaired or microglia were depleted from the brain , and quantified the ability of CpG-C to reduce the total amount of live tumor cells by assessing the radioactive signaling that originated from radiolabeled tumor cells . Employing this approach , animals were treated with minocycline , an inhibitor of microglial activation [40 , 56] ( Fig 7A ) , which resulted in a significantly increased brain tumor retention ( p = 0 . 0118 ) , without affecting the total number of infiltrating tumor cells ( see below ) . Importantly , CpG-C treatment reduced tumor retention in naïve mice ( p < 0 . 0001 ) , but not in minocycline-treated animals ( p = 0 . 1863 ) . Moreover , the effects of CpG-C were completely blocked by minocycline treatment ( p < 0 . 0001 ) , indicating the mediating role of microglia in the beneficial effects of CpG-C . To further validate these significant results , animals were treated with CpG-C , or with minocycline and CpG-C , and mCherry ( tumor cells ) uptake by microglia was quantified using ImageStream FACS analysis and compared with saline-treated animals ( Fig 7B ) . In line with the radioactive-based quantification , CpG-C increased tumor cell phagocytosis ( i . e . , events in which the mCherry signal could be identified inside GFP-positive segmented objects; p = 0 . 0100 ) , and this effect was blocked by minocycline ( p = 0 . 0493 ) . Notably , infiltration capacity of tumor cells was not affected by minocycline treatment , as indicated by total area of mCherry ( i . e . , all detection events combined ) in the brain ( p = 0 . 8994 ) . Depletion of all microglia ( activated and nonactivated ) with PLX5622 [41] , a CSF1R inhibitor , blocked the beneficial effects of CpG-C ( p = 0 . 0068 ) , again indicating the mediating role of microglia . Microglia depletion alone did not affect tumor retention in brains of naïve animals ( p = 0 . 7490; Fig 7C ) . Given our in vitro and in vivo results , we predicted that CpG-C administration would result in elevated expression of apoptosis- and phagocytosis-related factors by microglia cells . We therefore preformed transcriptional analysis of microglia cells isolated from CpG-C–treated or control animals ( Fig 8A ) . We revealed a robust impact of the agent on the induction of mRNA encoding of apoptosis-inducing , phagocytosis-related , and inflammatory factors , while not affecting the inflammation-independent microglial marker Tmem119 [57] ( p = 0 . 7258; Fig 8A ) . Specifically , mRNA expression of the key apoptosis-inducing ligands , Tnfsf10 and Fasl , increased by 3–4-fold in microglia from CpG-C–treated animals ( p = 0 . 0252 and p = 0 . 0324 , respectively; Fig 8B ) . In addition , CpG-C treatment resulted in increased expression of receptors related to phagocytosis [58] , including CD47 ( p = 0 . 0186 ) and Trem2 ( p = 0 . 0199 ) , while Cd36 and Cd68 mRNA expression levels did not change ( p = 0 . 7080 and p = 0 . 9874 , respectively; Fig 8C ) . Marco , another important phagocytosis receptor [59] , was not detected in microglia of control animals , yet it was highly expressed in CpG-C–treated animals ( p = 0 . 0108; Fig 8C ) . While mRNA of the inflammatory cytokines Il-6 and Il1-β was not affected by CpG-C treatment ( p = 0 . 9690 and p = 0 . 6772 , respectively ) , Tnf and Inf-γ , which are known to synergistically induce apoptosis in tumor cells [54] , were increased by approximately 2- and 7-fold , respectively ( p = 0 . 0163 and p = 0 . 0374 , respectively; Fig 8D ) . mRNA of Nos2 , an inflammation-associated enzyme with tumoricidal properties at high concentrations [60] , was not detected in control animals , while abundantly expressed in CpG-C–treated animals ( p = 0 . 0203; Fig 8D ) . Irrespectively , and in line with our in vitro results , CpG-C did not affect microglia reaction to a non–tumor-related stimulus in vivo ( i . e . , laser-induced photodamage; p = 0 . 7474; S8 Fig ) . Overall , these in vivo findings strengthen the notion that prophylactic treatment with CpG-C is beneficial in reducing brain metastasis by triggering nonactivated microglia cells to adopt tumoricidal characteristics .
Brain metastasis is a detrimental manifestation of cancer progression with limited treatments , and a better understanding of this process is expected to improve therapeutic interventions . Here , employing three tumor models , we report that prophylactic systemic treatment with CpG-C , a TLR9 agonist , exerts beneficial effects through reducing tumor cell seeding and growth in the brain . Notably , NK cells and monocytes did not mediate antimetastatic processes in the brain , nor the beneficial effects of CpG-C , in contrast to their important role in the periphery ( shown also here in the lungs ) . Instead , we identify microglia as key mediators of these beneficial effects in the initial steps of metastatic brain colonization . Moreover , we show that activation of microglia is essential for its antimetastatic function . Thus , CpG-C stimulates microglia to adopt antitumor characteristics , inducing tumor apoptosis and phagocytosis , thereby reducing the formation of brain metastases ( Fig 9 ) . Systemic treatment against brain metastasis has been proposed as a first therapeutic choice [2 , 4 , 61] , but no effective clinical routine is yet available . A previous study indicated that systemic administration of a CpG-ODN can result in altered cerebral mRNA expression profile [62] , suggesting that the agent could have reached this organ . Furthermore , CpG-ODN was shown to stimulate BV2 microglia cells in vitro [63] , and intracranial injection of CpG-ODN resulted in activation of microglia cells in vivo [64] . However , there was no direct in vivo evidence demonstrating that such an agent could enter the brain parenchyma if administered systemically and elicit a beneficial effect , fundamental requirements for a prophylactic treatment in cancer patients . Notably , direct intracranial injection of tumor cells or CPG-ODN ( or any other agent ) alter the neuro-immune environment by eliciting an inflammatory response [65]; thus , interpreting the role of immune cells in these settings is less straightforward . We overcome these technical hurdles and show here , for the first time , that following systemic administration ( i . e . , intraperitoneally ) , CpG-C was abundantly taken up by TLR9-expressing cells across the brain without affecting BBB integrity or infiltration of immune cells into the brain ( Fig 4 , S5 Fig ) , and dramatically reduced brain colonization by circulating tumor cells ( Figs 1 , 2 , 6 and 7 , S2 and S7 Figs ) . These findings pave the road for exploiting this compound in the clinic , as it could be easily administered systemically to serve as a prophylactic agent for patients with high risk of developing brain metastases . An even more urgent clinical scenario in which this treatment could prove life saving is the perioperative period—days to weeks before and after tumor excision—which is now acknowledged as a critical therapeutic window for reducing postoperative metastatic disease [14 , 66] . Indeed , various short perioperative interventions were reported to markedly impact short- and long-term cancer outcomes [14 , 67–69] . As brain metastases are common in cancer patients and are associated with poor prognosis [1] , reducing their postoperative occurrence is key in improving survival [4] . Here , we show that in a spontaneous brain metastasis model of melanoma , a short perioperative treatment with CpG-C , spanning three days prior to and following primary tumor excision , results in reduced brain tumor burden ( Fig 1G ) . Importantly , CpG-C was shown to have negligible toxicity in humans [19–21] . While we did not directly test whether systemic CpG-C administration has any deleterious effects on neuronal activity , it has been shown by others that when administered directly into the brain ( resulting in higher local concentrations ) , CpG-ODNs do not cause neurotoxicity in animals [27] , nor result in significant or permanent neurological deficits in humans [19–21] . Therefore , while traditional chemo- and radiation therapies cannot be used during the perioperative period ( due to their deleterious effects on tissue healing and immune competence ) , the use of CpG-C could be a promising prophylactic approach during this critical time frame [14] . In preclinical trials , acute and chronic systemic CpG-ODNs ( including CpG-C ) were shown to reduce primary tumor growth and metastases in peripheral organs [10–12 , 70] . Importantly , CpG-ODNs are evaluated as standalone antitumor agents as well as vaccine adjuvants in several clinical trials of different cancers , and systemic administration is considered well tolerated , with negligible toxicity [71 , 72] . Given the low toxicity of CpG-C and its wide-range antitumor effects , extended use beyond the perioperative period can also be considered . Additionally , TLR9 stimulation of microglia cells has also been shown to be beneficial in various neurological pathologies , including Alzheimer [26] and seizure-induced aberrant neurogenesis [28] , although systemic treatment has not been studied for these conditions . As such , systemic CpG-C treatment could be considered as a therapeutic intervention for cancer and non–cancer-related pathologies . It is well established that innate immune cells play a key role in preventing and eradicating metastases in the periphery [73–75] . Indeed , we herein show that depletion of NK cells and monocytes results in elevated tumor seeding in the lungs ( Fig 2A and 2C ) . However , in the brains of the same animals , we made the novel observation that NK and monocyte depletions have no effect , and that they do not mediate the beneficial effects of CpG-C ( Fig 2B and 2D ) . While mature NK cells are abundant in the capillaries of the lungs and liver [76] , only limited numbers of immature NK cells are found in cerebral capillaries [77] . Also , while patrolling monocytes [21] and pulmonary-resident macrophages are the first line of defense in the lungs [78] , monocytes infiltrate the brain parenchyma only under pathological conditions in which the BBB is compromised [79 , 80] , a condition that does not characterize the early stages of tumor cell infiltration [81] . Notably , systemic CpG-C administration did not affect infiltration of T cells ( i . e . , CD4+ ) or monocytes ( i . e . , CD68+ ) into the brain ( Fig 4D ) , or the number of GFP+ cells ( i . e . , monocytes/microglia ) evident in the brain 24 hours following administration of tumor cells ( Fig 6F ) . These differences between the periphery and the brain underscore the importance of studying brain-specific mechanisms that regulate the metastatic process , to allow tailoring of relevant therapies . In the brain , microglia are the primary immune effector cells [53] . Close interactions between macrophages/microglia cells and established metastases have been reported in human brain samples [82 , 83] . In mice , it has been shown that heterogeneous microglia cells , activated and nonactivated , accumulate proximal to invading tumor cells [55] and infiltrate established metastases generated by intracranial injection [84] . However , the role of microglia in regulating brain tumor progression , especially during the initial steps of tumor colonization , remains unclear [83 , 85–88] . Here , we show that during this time frame ( as early as the first 24 hours following administration of tumor cells ) , CpG-C activates microglia to eradicate micrometastases . Notably , established tumors can modulate activation of microglia , recruiting them to support tumor progression , whereas enabling microglia activation has an opposite effect [87–90] . Therefore , while not directly tested , CpG-C may have also had a secondary effect on development of macrometatases . Our in vitro results indicate that both primary cultured and N9 microglia cells exert low tumoricidal activity ( Fig 5 ) , in line with previous findings [91] . Here , however , we clearly show that activation of microglia with CpG-C markedly increases this cytotoxic activity , mediated through direct physical contact with tumor cells and not in a paracrine fashion . While it has been argued that microglia cells promote initial steps of colonization of breast tumor cells in vitro and in acute slices [83] , we found through in vivo two-photon imaging that microglia contact and phagocytize tumor cells immediately after their infiltration into the brain ( Fig 6A–6C ) , and they do so more abundantly following systemic administration of CpG-C ( Fig 6D and 6E ) . Accordingly , CpG-C increased mRNA expression of apoptosis-inducing and phagocytosis-related genes in microglia ( Fig 8 ) , without affecting microglia density ( Fig 6F ) . Furthermore , by blocking microglia activation with minocycline ( Fig 7A and 7B ) and by depleting them with CSF1R inhibitor ( Fig 7C ) , we show that microglia mediate the beneficial in vivo antimetastatic effects of CpG-C . While we cannot rule out the possibility that minocycline affected the tumoricidal activity of macrophages [92] , given monocyte depletion did not affect tumor burden or mediate the effects of CpG-C ( Fig 2D ) , it is unlikely that the effects of minocycline were mediated by macrophages . Intriguingly , the complete depletion of microglia cells did not affect brain tumor burden in the first 24 hours ( Fig 7C ) . This could be explained by the fact that all the different subsets of microglia are depleted , both those that could have a tumor supportive and tumoricidal role [88] . The metastatic process involves several steps , including arrest in the brain vasculature; infiltration through the BBB , meninges , and blood-cerebrospinal fluid barriers [93]; and colonization of the brain parenchyma [94] . Although CpG-C could have affected all of these steps in different magnitudes , as endothelial cells and astrocytes also uptake the adjuvant ( Fig 3 ) , we clearly show that the pool of metastatic cells infiltrating the brain was not altered ( Fig 6G ) , leading to the conclusion that , even if not directly measured , arrest and infiltration were not significantly affected by CpG-C . Support for this argument comes also from our findings that the permeability of key brain-immune interfaces was not altered ( Fig 4 and S5 Fig ) . Nevertheless , this conclusion does not overrule the potential changes in the infiltration rate of circulating metastatic cells through the brain immune interfaces , a topic for future research . Overall , we demonstrate that shifting the balance from nonactivated to activated microglia , as with the systemic CpG-C treatment presented herein , results in the killing of invading tumor cells and prevents establishment of brain metastases . Such an approach could lay the foundation for a novel clinical perioperative therapy .
|
Brain metastases are prevalent and often terminal . Thus , reducing their occurrence could markedly improve cancer outcome . We show that systemic prophylactic and perioperative administration of a TLR9 agonist , CpG-C , reduced metastatic growth in experimental and spontaneous brain metastasis models , employing mouse and human tumors . CpG-C was taken up in the brain without affecting blood-brain barrier integrity and tumor extravasation . In vitro assays , imaging flow cytometry , and intravital imaging pointed to microglia as mediators of CpG-C effects through contact-dependent tumor killing and phagocytosis , corresponding with in vivo mRNA profile . In vivo depletion studies proved that microglia , but not NK cells or monocytes , mediated the beneficial effects of CpG-C , also hindered by blocking microglial activation . In toto , perioperative treatment with CpG-C should be considered clinically relevant .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2019
|
Prophylactic TLR9 stimulation reduces brain metastasis through microglia activation
|
West Nile Virus ( WNV ) is now endemic throughout North America , with annual recurrence dependent upon successful overwintering when cold temperatures drive mosquito vectors into inactivity and halt transmission . To investigate whether avian hosts may serve as an overwintering mechanism , groups of eight to ten House Sparrows were experimentally infected with a WN02 genotype of WNV and then held until necropsy at 3 , 5 , 7 , 9 , 12 , 15 , or 18 weeks post-infection ( pi ) when they were assessed for the presence of persistent infection . Blood was collected from all remaining birds every two weeks pi , and sera tested for WNV RNA and WNV neutralizing antibodies . West Nile virus RNA was present in the sera of some birds up to 7 weeks pi and all birds retained neutralizing antibodies throughout the experiment . The detection of persistently infected birds decreased with time , from 100% ( n = 13 ) positive at 3 weeks post-infection ( pi ) to 12 . 5% ( n = 8 ) at 18 weeks pi . Infectious virus was isolated from the spleens of birds necropsied at 3 , 5 , 7 and 12 weeks pi . The current study confirmed previous reports of infectious WNV persistence in avian hosts , and further characterized the temporal nature of these infections . Although these persistent infections supported the hypothesis that infected birds may serve as an overwintering mechanism , mosquito-infectious recrudescent viremias have yet to be demonstrated thereby providing proof of principle .
West Nile virus ( WNV; Flaviviridae: Flavivirus ) is a relatively new arbovirus in the Western Hemisphere that has been detected each year [1] since its introduction into New York , NY in 1999 . West Nile virus is an enveloped , single-stranded RNA virus [2] that is primarily transmitted in an enzootic cycle involving Culex mosquitoes and passerine birds . Humans and horses are infected tangentially and generally do not contribute to the transmission cycle . The success of the WNV invasion can be attributed , in part , to the presence of competent mosquito vectors and avian hosts [3]–[5] , and to the virus' ability to survive temperate winters that drive mosquito vectors into inactivity and halt the transmission cycle . The mechanisms allowing WNV to overwinter likely rely on persistent infection of either mosquito vectors or avian hosts . Previous studies have reported the winter collection of WNV-infected Culex mosquitoes [6]–[9] . Vertical transmission of WNV in mosquitoes , although demonstrated infrequently [10]–[13] , was most likely the mechanism by which these overwintering mosquitoes became infected . Alternatively , persistent WNV infections have been described in vertebrates , including mice ( Mus musculus ) [14] and Golden Hamsters ( Mesocricetus auratus ) [15] . However , because these rodents are not a natural host for WNV , direct implications for WNV overwintering cannot be inferred . Persistent WNV infections also have been reported in several avian species . Early work described the recrudescence of WNV persistent infections in Blue-gray Pigeons ( Columba cf . livia ) , with virus isolated from blood at 16 , 93 and 100 days post-infection ( pi ) [16] . Reisen et al . [13] demonstrated that WNV RNA could be demonstrated in spleen , kidney and lung tissues in several avian species up to 6 weeks pi . In addition , infectious virus was isolated from four of six of RNA positive House Finches ( Carpodacus mexicanus ) after passage in C6/36 mosquito cells . Nemeth et al . [17] isolated infectious virus from the mouth of an experimentally infected House Sparrow ( Passer domesticus ) 44 days pi and found RNA in kidney and/or spleen tissues at 65 days pi . Our previous work [18] revealed that WNV RNA persisted in the spleen and/or kidney tissues of some experimentally infected birds up to 6 months pi and in some naturally-infected birds greater than 4 months pi . Because attempts at isolating infectious virus at these time periods failed [18] , it is unknown whether this detection of WNV RNA could be attributed to infectious virus , with the potential to recrudesce and restart transmission . The purpose of the current work was to characterize temporal changes in persistent WNV infections using the House Sparrow as an avian model . This was accomplished by conducting a time course experiment in which House Sparrows were experimentally infected with WNV , then evaluated for viral persistence at multiple time points up to 18 weeks pi . Spleen , kidney , skin , and brain tissues taken at necropsy were tested for both WNV RNA and infectious virus . In addition , birds were monitored throughout the experiment for recrudescence by screening sera for WNV RNA and for immune status by testing for neutralizing antibodies .
This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The University of California Davis campus is approved for animal studies by the US National Institutes of Health under Animal Welfare Assurance number A3433 . The collection , housing , transport , infection and euthanasia of birds were conducted under approved University of California , Davis , Institutional Animal Care and Use Committee protocols 12876 and 12880 . Birds were collected by grain-baited traps and mist nets under USGS Master Station Banding Permit 22763 and State of California Scientific Collecting Permit 801281-01 and taken for experimentation under Federal Permit MB082812 . BSL3 laboratory facilities were approved under Biological Use Authorization 0554 and 0873 by the University of California , Davis , Environmental Health and Safety Institutional Biosafety Committee and USDA Permit 47901 . African Green Monkey kidney ( Vero ) cells were maintained in Dulbecco's Modified Eagle medium ( DMEM; Life Technologies: Gibco , Carlsbad , CA ) supplemented with 10% fetal bovine serum ( FBS ) , 500 U/mL penicillin and 0 . 5 mg/mL streptomycin , and maintained at 37°C and 5% C02 . The WNV strain used was isolated originally from a dead Yellow-billed Magpie ( Pica nuttalli ) collected in 2004 in Sacramento , CA ( WNV CA04 , GenBank accession number DQ080059 ) , and had been passaged three times on Vero cells prior to experimentation . Viral titers were assessed by Vero cell plaque assay [19] . House Sparrows were selected as an avian model because they were considered an important maintenance host [3] , [4] for WNV , and frequently have been reported positive for WNV antibodies in seroprevalence studies [20] , [21] . In addition , House Sparrows were abundant , easily collected , adapted well to laboratory conditions , and previously were shown to develop persistent WNV infections [13] , [17] , [18] . Birds were collected in Kern and Yolo Counties , CA , banded with a uniquely numbered aluminum leg band , screened for antibodies attributable to previous WNV infection , and given a minimum of two weeks for cage-adaption . During the cage-adaption period , a 14-day course of chlorotetracycline ( Fort Dodge; Overland Park , KS ) was administered in drinking water at 0 . 2 mg/mL . Until adequate numbers were collected , birds were held at the Arbovirus Field Station in Bakersfield , CA or the University of California , Davis , in mosquito-proof , outdoor flight cages equipped with perches , water and food dispensers , and bird baths . Prior to WNV infection , birds were transported to the California Animal Health and Food Safety Laboratory ( CAHFS ) animal containment facility in Davis , CA where they were held in wire cages [dimensions: 0 . 9 m ( W ) ×0 . 6 m ( L ) ×0 . 6 m ( H ) ] within Horsfall-Bauer containment units , each fitted with a HEPA-filtered negative air system . Similar to outdoor aviaries , cages were furnished with perches , water baths , and food and water dispensers . Diet included a mixture of formulated High Energy Breeder Diet ( Roudybush™; Woodland , CA ) and mixed bird seed including: canary grass seed , oat chips , golden German millet , canola rape seed , flax seed , Nyjer seed and hemp seed ( The Seed Factory; Ceres , CA ) . Sparrows were assigned to cages and study groups , with seven or less individuals per cage . Birds were needle-inoculated with 103 plaque forming units ( pfu ) of WNV , suspended in 0 . 05 mL viral transport diluent ( DMEM , 500 U/mL penicillin , 0 . 5 mg/mL streptomycin and 20% FBS ) , subcutaneously over the upper right pectoral muscle . Negative controls were sham-inoculated with transport diluent alone . Blood was collected in 0 . 1 mL volumes by jugular venipuncture and diluted in 0 . 45 mL of transport diluent , for a 1∶10 serum dilution . Clinical signs of WNV infection were rarely observed , but included puffing of feathers , lethargy , and failure to react when disturbed . Birds were euthanized prior to necropsy by CO2 inhalation . Groups of 8 to 12 experimentally infected House Sparrows were euthanized at 3 , 5 , 7 , 9 , 12 , 15 and 18 weeks post-infection . Negative controls were sacrificed concurrently with the last time point . Blood samples were collected at three or four days pi to measure the viremia response and then biweekly throughout the experiment ( Figure 1 ) . After blood collection all samples were centrifuged ( 9000× g; 5 min ) , sera collected , aliquoted , and either heat-inactivated and stored at −20°C for serology , or stored at −80°C for RNA extraction and detection by qRT-PCR . At the end of each holding period , a final blood sample was collected , House Sparrows were euthanized by CO2 inhalation , and tissues were collected to assess WNV infection . At necropsy , spleen , kidney , brain including a portion of the cerebellum , and skin from over the pectoral muscle at the inoculation site were collected . To prevent cross-contamination tools were cleaned between birds using 1-Stroke Eviron Germicidal Detergent ( Steris; Mentor , OH ) . After collection , each tissue was placed into cryovials and kept cold on wet ice until processing; tissues were not allowed to freeze . Co-culture medium ( DMEM , 500 U/mL penicillin , 0 . 5 mg/mL streptomycin , 0 . 05 mg/mL gentamicin sulfate , 2 µg/mL amphotericin B solution , and 10% FBS ) was added to each sample to make a 2 . 5% tissue homogenate . Two 5 mm glass beads were added to each tissue sample , whereas two 5 mm copper ball-bearings were added to skin samples . Tissues were homogenized by mixer mill ( MM300 , Retsch; Haan , Germany ) at a frequency of 24 cycles/second for four min . All sera were tested for antibodies by plaque reduction neutralization test [22] , using a 90% end point ( PRNT90 ) . Serum samples were heat-inactivated at 56°C for 30 min , and serially diluted 2-fold starting at 1∶10 . Diluted sera were mixed 1∶1 with virus diluent containing approximately 100 pfu of WNV ( CA04 ) . Assays were conducted using confluent Vero cells grown in 6-well plates , with 0 . 1 mL of virus/sera mixture added to the Vero cell monolayer and allowed to incubate for one hour at 37°C , 5% CO2 . A double overlay system was used; the first overlay ( nutrient medium , 1% agarose and 2 . 3 mg/mL sodium bicarbonate ) was applied after the incubation period . The second overlay ( nutrient medium , 1% agarose , 0 . 1 mg/mL neutral red and 2 . 3 mg/mL sodium bicarbonate ) was applied 48 h after the first; plates were read 72 h after inoculation . The highest dilution at which >90% of >75 pfu were neutralized was considered the PRNT90 end point titer . Sera collected on days three and four were tested for infectious virus by plaque assay [19] using Vero cell cultures in 6-well plates . Sera were serially diluted 10-fold and allowed to incubate for 60 min at 37°C with 5% C02 . Overlay schedule was as explained above . Plaques were enumerated at 72 h to obtain viral titers . Because sera were diluted 1∶10 at collection , the detection limit was 100 pfu/mL . Total RNA was extracted from tissue homogenates the day of necropsy; sera were frozen at −80°C and RNA extracted within four months of collection . RNA was extracted from a 50 µL sample of tissue homogenate or sera using a MagMAX™ 96 and viral isolation kits according to manufacturer protocols ( Life Technologies: Applied Biosystems; Carlsbad , CA , USA ) . Care was taken to avoid cross-contamination and numerous negative control wells containing only viral transport medium were included on each extraction plate . Confirmation samples from virus isolation attempts were processed on a different day . Contamination was not observed throughout in tissue homogenates , sera from sham-inoculated sparrows , or in negative control wells containing viral transport medium alone . RNA samples were analyzed for the presence and quantity of WNV RNA by TaqMan one-step quantitative reverse transcriptase-polymerase chain reaction ( qRT-PCR ) utilizing an ABI7900 platform ( Life Technologies; Applied Biosystems; Carlsbad , CA , USA ) . Samples were evaluated with two primer/probe sets in separate reactions . The first set , WN1 , was specific for the envelope region of the viral genome ( WN1 ) [23]: ( forward ) 5′- TCA GCG ATC TCT CCA CCA AAG -3′ , ( reverse ) 5′- GGG TCA GCA CGT TTG TCA TTG -3′ , and ( probe ) 6FAM-TGC CCG ACC ATG GGA GAA GCT –TAMRA . Confirmation was attempted with a second primer/probe set ( WN2 ) specific for NS1 region of the viral genome [24]: ( forward ) 5′-GGC AGT TCT GGG TGA AGT CAA -3′ , ( reverse ) 5′-CTC CGA TTG TGA TTG CCT CGT -3′ , and ( probe ) 6FAM-TGT ACG TGG CCT GAG ACG CAT ACC TTG T-TAMRA . All samples with a threshold cycle ( Ct ) score <40 were considered WN1 and/or WN2-positive . Sera with WN1 Ct scores >30 that failed to confirm with WN2 were re-extracted and qRT-PCR was repeated using the WN1 primers/probe . All sample plates contained a standard curve generated from cultured virus of known titer ( plaque forming units/mL ) and negative water controls . Tissue homogenates were evaluated for infectious virus using a co-culture method adapted from Appler et al . [14] and others [15] , [25] . In brief , Vero cells were grown in co-culture medium , as described above , to 75% confluence in 6-well plates; 100 µL of tissue homogenate was added to each well . Plates were observed daily for cytopathic effect ( CPE; rolling of cells and/or sloughing of the monolayer ) ; when CPE was observed the cell culture supernatant was tested for WNV antigen by VecTest ( Medical Analysis Systems Inc . , Camarillo , CA ) . If WNV antigen was detected , the supernatant was collected and no further passages were attempted . After seven days , 1 . 0 mL of supernatant was transferred onto fresh Vero cells . Each homogenate was passaged three times , unless WNV antigen was detected . After three passages RNA was extracted from the CPE negative samples and tested for WNV RNA by qRT-PCR . Statistical analyses were performed using GraphPad Prism version 5 . 04 for Windows ( GraphPad Software; La Jolla , CA ) . Student's t tests compared mean viremia titers ( log10 pfu/mL ) between birds bled on either three or four days pi , and between birds that survived or succumbed to WNV infection . Student's t test was also used to compare mean WN1 qRT-PCR Ct scores between samples that were WN2 primer/probe confirmed and unconfirmed . To test whether WNV persistence as indicated by recovery of RNA at necropsy led to greater antibody titers , loge transformed PRNT90 antibody titers were compared by a 2-way general linear model ANOVA with persistence status and time after infection as main effects .
Overall , 85 House Sparrows were infected experimentally with WNV , and 6 were sham-inoculated and held as negative controls . Over the course of the experiment , two birds died after blood sampling ( one of which was a negative control ) and two died approximately three weeks post-infection of unknown causes . In total , 13 birds succumbed during acute WNV infection between days two and twelve pi , with the majority ( 54% ) succumbing on the sixth day . To decrease stress birds were bled only once during the acute infection period . Based on our previous studies and the literature , blood was collected at four days pi to measure the magnitude of peak viremia . Unexpectedly , 11 of 70 experimentally infected birds had sera that were negative for infectious virus by plaque assay at this time , but all of these sera were positive for WNV RNA by qRT-PCR . In addition , all developed a WNV-neutralizing antibody response . Therefore , blood was collected from the remaining birds ( n = 13 ) at three dpi , at which time , all had detectable viremias by plaque-assay . Viremia titers were transformed to log10 plaque forming units ( pfu ) /mL and compared among the birds that survived infection . The mean viremia ( ± standard deviation ) of birds that survived acute infection and were bled on day three ( 4 . 2±1 . 1 log10 pfu/mL , n = 13 ) was significantly greater ( t = 3 . 0 , df = 58 , P<0 . 005 two-tailed ) than the mean viremia of surviving birds bled on day four ( 3 . 3±0 . 9 log10 pfu/mL , n = 47 ) . Sample sizes for birds that succumbed to WNV infection were low . However , in contrast to surviving birds , there was no significant difference ( t = 0 . 98 , df = 8 , P>0 . 05 two-tailed ) in mean viremia titers between birds that succumbed to WNV infection and were bled at three ( 8 . 5±0 . 07 log10 pfu/mL , n = 2 ) or four dpi ( 7 . 2±1 . 8 log10 pfu/mL , n = 8 ) . Birds that succumbed to infection had significantly higher viremias on both days three ( t = 5 . 4 , df = 13 , P<0 . 001 two-tailed ) and four pi ( t = 9 . 4 , df = 53 , P<0 . 001 two-tailed ) than those that survived . By 14 dpi all experimentally infected birds were PRNT90 positive for WNV antibodies , and as previously reported [26] , the humoral response was robust . Figure 2 shows the inverse of the geometric mean titers at 2 to 18 weeks pi . Neutralizing antibody titers peaked between five and nine weeks pi , but all experimentally infected birds maintained titers ≥1∶20 throughout the experiment . None of the sham-inoculated birds developed a WNV-specific antibody response ( PRNT90<1∶20 ) . When necropsied at 5–18 weeks pi , the mean ( ± standard deviation ) reciprocal of the loge PRNT90 titer for birds with persistent RNA in one or more organs ( 6 . 53±0 . 98 , back transformed mean = 645 . 5 , n = 21 ) was not significantly greater ( 2-way ANOVA , F = 1 . 95 , df = 1 , 44 , P = 0 . 17 ) than the mean for negative birds ( 5 . 79±1 . 35 , back transformed mean = 358 . 9 , n = 35 ) . Similar results were obtained when titers from birds necropsied at weeks 5–9 pi were analyzed separately ( data not shown , P>0 . 05 ) . Sparrows were evaluated for persistent WNV at 3 , 5 , 7 , 9 , 12 , 15 , and 18 weeks pi . West Nile virus RNA was detected in the tissues of at least one bird at all time points , except for birds necropsied at 15 weeks pi . The number of birds in each group that were positive for WNV RNA decreased with time pi: 100% ( n = 13 ) were positive at 3 weeks pi , 75% ( n = 8 ) at 5 weeks pi , 50% ( n = 10 ) at 7 and 9 weeks pi , 40% at 12 weeks pi , none ( n = 10 ) at 15 weeks pi , and 12 . 5% ( n = 8 ) at 18 weeks pi . In persistently infected birds , spleen and kidney tissues were most commonly RNA-positive; skin at the inoculation site and brain tissues were infrequently positive ( Table 1 ) . All birds were screened for WNV RNA by qRT-PCR using the WN1 primer/probe set . Tissue homogenate samples that were positive for WNV RNA ( Ct score <40 ) were retested using the WN2 primer/probe set , and 25% ( n = 4 ) of brain , 33% ( n = 9 ) of skin , 75% ( n = 20 ) of kidney , and 78% ( n = 27 ) of spleen samples were again qRT-PCR positive ( Ct<40 ) . The mean ( ± standard deviation ) WN1 Ct score for tissue samples that confirmed by WN2 ( 27 . 6±2 . 8 , n = 40 ) was significantly less ( t = 6 . 3 , df = 58 , p = <0 . 001 one-tailed ) than that of samples that failed to confirm ( 33 . 0±3 . 6 , n = 20 ) . These results are similar to previous findings [18] , where the reduced sensitivity of WN2 primer/probe set limited confirmation of some samples with Ct scores >30 . Figure 3 compares the sensitivity of WN1 and WN2 qRT-PCR assay results to WNV plaque assay titers for viral standards prepared in Vero cell culture . Virus isolation was successful , but an infrequent occurrence , with WNV generally isolated from the spleen and on two occasions the kidney ( Table 1 ) . Amongst the birds tested for persistence at 3 weeks pi , WNV was isolated from the spleen and kidney of two birds , and the spleen alone from a third . In addition , virus was isolated from the spleen of one bird at 5 weeks pi , two birds at 7 weeks pi , and a final bird at 12 weeks pi . In nearly all cases CPE was observed during the second passage on Vero cells , except for the two spleen samples collected 3 weeks pi , where CPE was detected on the first passage . Despite the addition of amphotericin B to the co-culture medium , skin homogenate cultures were often lost to fungal contamination . This contamination was not seen in other tissues and was attributed to contamination introduced from the skin itself , and not contamination acquired in the laboratory . To prevent contaminating other cultures , skin samples were cultured separately and discarded when/if contamination was noticed . Future attempts may require additional anti-fungal additives to facilitate successful co-culture of skin homogenates . Biweekly throughout the experiment and at termination , blood samples were collected and sera tested for WNV RNA . Interestingly , ten sparrows had post-acute ( ≥2 weeks pi ) serum samples positive for WNV RNA by qRT-PCR using the WN1 primer/probe set ( Table 2 ) . Among these birds , half were positive on more than one sample date , and two were positive up to seven weeks pi . All birds with sera positive for WNV RNA , aside from one ( bird #74 ) that died unexpectedly at three weeks pi , were assessed for persistent infection . Seven of nine birds with WNV RNA positive sera also had tissues that were WNV RNA positive ( Table 2 ) . The two birds with negative tissues were necropsied at 15 weeks pi , but had positive sera at two ( bird # 550 ) and two and six ( bird #513 ) weeks pi . Presumably , these birds cleared their persistent infections prior to assessment at 15 weeks pi . Based on estimates from our qRT-PCR standards ( Figure 3 ) , Ct scores obtained from WNV RNA positive sera corresponded to very low viral titers of approximately 10–100 pfu/mL . As reported above , 11 experimentally infected birds failed to show a viremia response by plaque assay at 4 dpi , although sera from these 11 birds were positive for WNV RNA by qRT-PCR using WN1 . Viral RNA in 10 of these 11 birds was confirmed by qRT-PCR using WN2 . The WN1 Ct scores from positive samples ranged from 28 . 7 to 32 . 9; the sample that failed to confirm had a WN1 Ct score of 35 . 1 . In contrast , of the 18 post-acute serum samples that were qRT-PCR positive with WN1 , only one sample ( bird 74 , two weeks pi ) was confirmed by WN2 , and this sample had a WN1 Ct score of 27 . 1; the remaining 17 positive sera had a mean ( ± standard deviation ) WN1 Ct score of 35 . 6±1 . 02 . The un-confirmed samples were re-extracted and qRT-PCR repeated using WN1; 35% of these samples were confirmed by re-extraction . Repeated freeze-thaw of samples that contained very low copy numbers of target RNA may have reduced the number of samples that re-confirmed by this method .
WNV-infected House Sparrows , and other bird species , develop persistent WNV infections . In House Sparrows these persistent infections were increasingly undetectable with time post-infection and existed concurrently with elevated serum antibody titers . In the current study , WNV RNA was detected in the kidney of one bird at 18 weeks pi and infectious virus was isolated from the spleen of another bird at 12 weeks pi . The isolation of infectious virus indicated the persistence of intact WNV and not just residual RNA from the acute infection . Experimental infection was conducted with an isolate of the North American or WN02 genotype of WNV circulating in California during 2004 , which differed from previous studies of WNV persistence that utilized NY99 isolates [13] , [15] , [17] or infectious clones [14] . However , our previous study [18] reported that the infecting WNV strain did not significantly alter the proportion of birds that developed persistent infection . Both this study and research published previously by Nemeth et al . [17] , in which birds were infected with NY99 , reported similar proportions of House Sparrows positive for persistent WNV ( WNV RNA from any tissue ) at approximately 9 weeks pi . Here , 5 of 10 House Sparrows were positive for persistence , whereas Nemeth et al . [17] reported 2 of 14 positive . Although sample sizes were low , there was no significant difference in these two proportions ( Fisher's exact; P>0 . 05 ) . The mean 4 dpi viremia of experimentally infected birds in the current study was 3 . 5 log10 pfu/mL of sera , and was much lower than previous WNV experimental infection studies in House Sparrows that reported 5 . 0 [4] , 9 . 9 [27] and 10 . 3 log10 pfu/mL [3] at 4 dpi . This decrease in amplitude of the 4 dpi viremia is interesting and has important implications for the transmission of WNV , because decreases in viremia titer are well correlated with decreases in host competence or the proportion of mosquitoes infected through blood-feeding [28] . Although , it is not specifically known why the House Sparrows in our study developed a lower 4 dpi viremia , nearly eight years of WNV selection pressure on House Sparrows from the collection areas may have led to selection for birds with innate resistance . In addition , birds in our study may have been under less sampling stress , because they were only handled and bled once during the acute infection period . To improve overall colony health , sparrows were treated with chlorotetracycline prior to WNV infection . Although this antibiotic may have improved the health status of the birds , chlorotetraclcline has been demonstrated to be ineffective against WNV [29] and should not have altered the course of infection . Regardless of the cause of the decreased viremic response , this outcome did not appear to impact persistence . Ten experimentally infected birds had sera that were positive for WNV RNA from 2 to 7 weeks pi and after the acute infection period . It was not clear whether sera were intermittently positive , due to recrudescence , or if some birds simply maintained low-grade persistent viremias . Three birds ( 520 , 524 , and 548 ) appeared to have developed persistent viremias , and two birds ( 508 and 513 ) alternated positive/negative/positive . The five remaining birds were RNA-positive on only one occasion . Quantities of RNA detected corresponded to extremely low viral titers of 10–100 pfu/mL of serum . Blood was drawn in 0 . 1 mL volumes that were diluted to create a 1∶10 serum dilution . Because of the small collection volume and dilution , it is possible that some samples containing 10–100 pfu/mL may have been missed through sampling . Confirmation of these low-grade viremias was vexing; however , several birds were positive on more than one occasion , negative controls were always negative , and all but two birds tested 15 weeks post-infection had one or more tissues positive for WNV RNA at necropsy . Despite the fact that some birds developed either low-grade persistent or recrudescent viremias , these viremias would not likely result in mosquito infection . We previously demonstrated that avian antibodies protect mosquitoes from WNV infection [28] , and all birds in the current study retained robust antibody titers sufficient to bind virus at these low viremia levels . Experimentally infected House Sparrows developed and maintained robust neutralizing antibody titers post-infection . Previous studies reported this response in both House Sparrows [26] and mice [14] . It is becoming apparent that WNV infections are not always cleared after acute infection . These lingering or perhaps recrudescent infections may provide continual stimulation to the host immune system , thus serving to amplify and maintain the humoral response . However , similar to our previous study [18] , antibody titers at necropsy of persistently infected birds in the current study were not significantly higher than the titers of birds negative for persistence . These analyses may have been confounded by the loss of detectable RNA over time , so that some of the birds , such as those that had post-acute serum positive for WNV RNA ( birds 513 and 550 ) but tissues negative at necropsy , may have cleared persistent infections prior to sampling . At this time , events or conditions that may initiate a recrudescence event and thus restart transmission in nature have not been elucidated . Optimizing the limit of virus detection is critical for characterization of WNV persistence and therefore some birds were reported as qRT-PCR positive , even if they failed to confirm via WN2 or re-extraction and WN1 testing . Although a confirmatory primer/probe set can be useful for avoiding a type I error , results may be confounded if the primer/probe used is less sensitive than the screening probe . In contrast , a type II error would certainly result if all samples that failed to confirm were disregarded . To balance these two types of error , extreme care was used to avoid cross-contamination , numerous negative controls were included on RNA extraction and qRT-PCR plates , and samples with high Ct scores were re-extracted and qRT-PCR repeated . Our study confirmed previous reports of WNV persistence in avian hosts [13] , [16]–[18] , and further characterized the dynamic nature of these infections utilizing a House Sparrow model . Persistent infections appeared to resolve steadily with increasing time post-infection . Although , virus isolation was not a common occurrence , recovery of infectious virus at 12 weeks pi indicated that persisting RNA may be attributed to infectious , intact virus , but was likely moderated by the host immune system . West Nile virus RNA was detected in the sera of persistently infected birds up to 7 weeks pi , but it was not clear if these low-grade post-acute viremias were persistent or recrudescent . All experimentally infected birds maintained neutralizing antibody titers throughout the experiment . When present in sufficient quantity , host neutralizing antibodies , consumed within a bloodmeal , protected mosquito vectors from WNV infection [28] . Therefore , it currently appears that healthy birds that maintain neutralizing antibody titers are unlikely to develop a mosquito-infectious recrudescent viremia . However , evidence indicates that persistent infections maintain virulence . Therefore , factors that compromise a persistently infected bird's immune system possibly could enable mosquito-infectious recrudescent viremias . In some cases , WNV infections in avian hosts persisted sufficiently long to serve as a potential overwintering mechanism for the virus . However , for proof-of-principle persistent WNV infections must reinitiate a transmission event either through a mosquito-infectious viremia or bird to bird transmission . Future work is required to determine which factors enable recrudescence and whether this occurs in nature .
|
House Sparrows experimentally infected with West Nile virus [WNV] were necropsied at multiple time points from 3 to 18 weeks post infection ( pi ) . The percent of birds with tissues positive for WNV RNA decreased from 100% at 3 wks to 13% at 18 wks pi; infectious virus was recovered from some birds by tissue co-cultivation and Vero cell passage from 3 to 12 wks pi , even though positive birds retained neutralizing antibody . WNV RNA also was detected in sera at 2 to 7 wks pi . Collectively , these data indicated that House Sparrows frequently developed persistent infections and could serve as an overwintering mechanism for WNV . However , recrudescent viremias suitable to infect mosquitoes have yet to be demonstrated and would seem to require host Immunosuppression .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"west",
"nile",
"fever",
"viral",
"diseases"
] |
2012
|
Dynamics of West Nile Virus Persistence in House Sparrows (Passer domesticus)
|
Schistosomiasis remains a significant health burden in many areas of the world . Morbidity control , focused on limiting infection intensity through periodic delivery of anti-schistosomal medicines , is the thrust of current World Health Organization guidelines ( 2006 ) for reduction of Schistosoma-related disease . A new appreciation of the lifetime impact of repeated Schistosoma infection has directed attention toward strategies for greater suppression of parasite infection per se , with the goal of transmission interruption . Variations in drug schedules involving increased population coverage and/or treatment frequency are now undergoing field trials . However , their relative effectiveness in long-term infection suppression is presently unknown . Our study used available field data to calibrate advanced network models of village-level Schistosoma transmission to project outcomes of six different community- or school age-based programs , as compared to the impact of current 2006 W . H . O . recommended control strategies . We then scored the number of years each of 10 typical villages would remain below 10% infection prevalence ( a practicable level associated with minimal prevalence of disease ) . All strategies that included four annual treatments effectively reduced community prevalence to less than 10% , while programs having yearly gaps ( ‘holidays’ ) failed to reach this objective in half of the communities . Effective post-program suppression of infection prevalence persisted in half of the 10 villages for 7–10 years , whereas in five high-risk villages , program effects on prevalence lasted zero to four years only . At typical levels of treatment adherence ( 60 to 70% ) , current WHO recommendations will likely not achieve effective suppression of Schistosoma prevalence unless implemented for ≥6 years . Following more aggressive 4 year annual intervention , some communities may be able to continue without further intervention for 8–10 years , while in higher-risk communities , annual treatment may prove necessary until eco-social factors fostering transmission are removed . Effective ongoing surveillance and locally targeted annual intervention must then become their mainstays of control .
Schistosomiasis is an environmentally transmitted parasitic disease that results in increased morbidity and mortality among millions of people living in tropical and subtropical regions [1]–[3] . Different control initiatives have had success in reducing either the prevalence or the mean intensity of Schistosoma infections in a number of individual countries [4]–[9] although , for other settings , the potential effectiveness of such strategies remains mostly unknown . With combined approaches of snail control , chemotherapy , health education , and hygienic improvement , China has made substantial progress against S . japonicum , which has been successfully eliminated in several provinces [10]–[13] . In Brazil , mass chemotherapy programs have achieved at least a 50% to 70% decrease in S . mansoni prevalence , with a marked reduction in infection-related morbidity and hospitalization in many states [14] . However , the effects of those initiatives have been uneven because of the regional- and country-specific differences in local transmission , in available resources , and in overall program commitment [12] , [15] . In the meantime , schistosomiasis remains a major public health problem in sub-Saharan Africa [3] , [8] . Possible measures for controlling African schistosomiasis include chemotherapy ( primarily praziquantel therapy ) , snail control ( by mollusciciding or habitat modification ) , provision of safe water alternatives , and/or education with behavior modification to limit exposure [4] . Among these , chemotherapy-based mass treatment remains the most widely used at present [8] . ‘Morbidity control’ , focused on limiting infection intensity through periodic delivery of anti-schistosomal medicines , is the thrust of the World Health Organization ( WHO ) schistosomiasis control guidelines [16] . However , a new appreciation of the lifetime impact of repeated Schistosoma infections [17] , [18] has directed attention toward strategies for greater suppression of parasite infection per se , with the ultimate goal of transmission interruption as a more effective elimination of disease risk . However , experience in schistosomiasis control programs suggests that current broad-based drug delivery campaigns may not consistently reduce local prevalence or prevent rapid reinfection [19] , [20] . Randomized field trials are now underway to examine the relative impact of increased population coverage and/or treatment frequency in community-wide or school age-targeted treatment programs for Schistosoma control [21] . However , their effectiveness relative to existing WHO protocols is not yet known . In the present study , we used a computer simulation based on a recently developed , stratified worm burden ( SWB ) model [22] of multi-village Schistosoma transmission to project programmatic success for each of seven different mass treatment control strategies in a typical Schistosoma-endemic region of sub-Saharan Africa .
In Gurarie , et al . 2010 [22] , we developed a new approach to predicting schistosomiasis transmission and community-level infectious burden using models based on a Stratified Worm Burden ( SWB ) formulation , combined with calibration using data from a completed multi-year pilot schistosomiasis control program . In this approach , a host population is subdivided into burden strata in terms of individual parasite load , with the ith stratum carrying ( i−1 ) Δw to iΔw worms , where the size of the stratum , iΔw , is prescribed e . g . , at increments of an additional 10 worm pairs per person . The transitions among strata depend on ( i ) force of infection ( determined by the transmission environment and human risk factors ) ; ( ii ) worm attrition ( natural or drug-induced ) ; and ( iii ) human demographics ( birth , aging , mortality , growth sources ) . Details are presented in Supplement S1 and [22] . The strength of this SWB model analysis is that it more accurately reflects many of the underlying complexities of Schistosoma species transmission , while yielding a more usable projection of community infection prevalence ( rather than mean worm burden ) after a campaign . The SWB includes and tracks differences in infection risk between children and adults , the highly skewed distribution of infection intensity among infected individuals , and the networked meta-population aspects of transmission among neighboring villages sharing common water sources [23] . This is an advantage where decision-making policy is made on the basis of local population or school age prevalence of Schistosoma infection [16] After the model was calibrated with available field data from pilot control programs [24] , [25] , it was used to project the likely outcomes of different targeted- or population-wide mass treatment programs now being introduced into endemic areas in sub-Saharan Africa [21] , [26] . The model can realistically account for inevitable program limitations in program resources , delivery efficiency , etc . , whereby drug treatment cannot cover every infected human in the treatment area and the frequency of the treatment cannot be too high . Non-stationary and nonlinear models of this type are difficult or impossible to solve through classical equation analysis . However , their performance and outputs can be estimated for different chemotherapy scenarios through numerical analysis [22] in the programmed simulations presented here . Taking burden strata as model variables instead of the population mean worm burden , SWB carries detailed information of the infection status of a host population and reflects its highly aggregated distribution ( typical of Schistosoma and other helminth infections [27] ) , without imposing a priori assumptions about that distribution . Statistical moments ( variance and/or aggregation ) of such a distribution can include human population dynamics and , in its most simplified format , even reflect the predictions of the classical Mean Worm Burden ( MWB ) equations of MacDonald [28] . Following the SWB approach , we were able to develop calibration procedures [22] that could readily take advantage of available prevalence and intensity data from published treatment trials [24] , [25] . The calibration methodology was , first , to estimate local force of infection in each village through fitting of SWB equilibrium equation parameters to prevalence data , then using reduced MWB and snail prevalence systems to estimate transmission parameters and the net effects of other unmeasured biotic and abiotic variables associated with transmission at local water sites [22] . These local factors include , in aggregate , water quality , rainfall , vegetation , seasonal water level persistence , human contact and sewage contamination . This model development allowed prediction of treatment outcomes across distributed village systems ( Figure 1 ) , including the significant effects of having multiple human age-strata sharing multiple water contact sites . For the present study , our base-case study system consisted of 10 neighboring villages and 5 shared water ( snail ) sites , where the village level infection exposures are linked through the shared water sites ( see Figure 1 for schematic representation of the transmission network ) . The calibration is taken from a well-studied Schistosoma haematobium-endemic region of the Msambweni District in coastal Kenya [24] , [25] , [29]–[31] . The 10 villages used to calibrate this study are located 50 km southwest of Mombasa , Kenya in an agricultural region that produces rice , coconuts , sugar cane , cassava , and maize . The area has a monsoon-type climate , with the period of January to April–May being hot and dry . The mean monthly temperature varies from 26 . 3°C to 26 . 6°C , with lows of 23 . 5°C ( July ) to 26 . 9°C ( March–April ) and highs from 26 . 7°C ( August ) to 36 . 3°C ( February ) . Annual precipitation varies from 900 mm to 1 , 500 mm , with large yearly and monthly fluctuations . There are usually two rainy seasons , the long rains starting around March and continuing until July , and the short rains beginning in October and lasting through November . In 1984 , the total population of this area was determined by household census to be 8 , 957 . It was 16 , 790 by repeat census in 2002–3 . Surveys of water use in the villages indicates that exposure to cercariae-infected waters comes through the practice of using dammed streams and seasonal ponds for washing and bathing , despite the availability of piped water ( from kiosks ) and borehole wells in most villages . No other schistosomiasis control programs were implemented in the study area during the course of these studies; routine chemotherapy dispensed by the local hospital was monitored , and there was no interval change in the level of such treatment . As is typical for schistosomiasis [32] , pre-treatment levels of Schistosoma infection school age prevalence varied significantly ( 23% to 74% ) across the landscape , even within a span of 5 km [33] . Data available for model calibration included: ( i ) demographic data , i . e . , human population numbers in different villages divided into child and adult age groups and , in addition , snail population densities at georeferenced exposure sites [34] , [35]; ( ii ) infection , in terms of individual and mean egg counts for children in each village , and also the density of susceptible , infected , and shedding adult snails in all monitored water sites [34] , [35] , [36]; ( iii ) behavioral data on water contact exposures for each village population among their adjacent water contact/snail-infected sites [24] , [31] . Full human and snail data were collected in 1983–1987 and again in 2000–2003 [24] , [31] , [33]–[35] . Human prevalence data for 2 villages were collected in 2006 and 2009 [37] . In all human surveys infection prevalence and mean egg-counts were based on standard 10 mL filtration of two urine samples [38] collected from the resident populations surveyed . The transmission coefficients from snail sites to villages and from villages to snail sites and the infection ( egg excretion ) rates for children and adults were estimated based on model fitting to these data . The calibrated parameters in [22] consisted of per capita transmission rates A ( snail-to-human ) and B ( human-to-snail ) , partitioned among different human villages i = {1 , 2 , … , 10} , snail sites j = {1 , … , 4} , and demographic groups ( “children” , adults” ) . All other inputs ( demographic , environmental ) are then entered into the model as additional factors . For instance , having human population ( “children”+“adult” ) Hi = Hic+Hia at village “i” , that carry MWB {wic , wia} , and snail number/infection prevalence {Nj , yj} at site “j” , we estimate the forces of infection between “i” and “j” asin terms of per-capita rates {Aijc/a} , {Bjic/a} . The overall force of human infection at each site λ = λc/a , along with “young/adult” demographics , will determine the SWB distribution for each group through equilibrium relations ( see Fig . 2 of [22] ) . Our basic assumption is that per-capita rates remain unchanged in time . So having calibrated model with the 1983–1987 , and 2000–2006 data , we can apply it then for any other period , changing demographics and starting infection levels ad hoc for a given time period . Because local environment remains stable , where transmission decreases it is due to the reduced number of infected humans . For the current analysis , we have taken 2009 data for demographics and human infection [37] as the starting point ( initial model inputs ) and have numerically simulated the effect of different treatment strategies over long ( multi-year ) periods , taking into account projected population changes ( growth parameters {Hic/a , Nj} - as prescribed functions of time ) , grounded on most recent human census and snail recovery trends . The effect of drug treatment on the prevalence of Schistosoma infection was implemented in the model by instantaneously shifting humans at higher burden levels ( stratum ) to lower levels according to the established efficacy of praziquantel [25] , [39] . As an example , within the model , a treatment session with an efficacy of 90% ( i . e . , killing 90% of worms ) would bring a person in the 10th stratum , who is carrying between 90 and 100 worm pairs before therapy , down to the 0th strata ( carrying 0 to 10 worms ) , as the remaining number of worm pairs is expected to be between 9 and 10 of worms after the treatment . Because egg output at this level of infection is , in practice , undetectable [40] , all persons in the 0th stratum are considered ‘uninfected’ , at least in terms of their contribution to transmission risk . In each year , community percent prevalence is estimated as [1− ( fraction in the 0th stratum ) ]×100 . This is a more realistic effect than modeled in the past , in which a treatment term has typically been added to the natural mortality of worms to indicate an extra loss . In fact , the drug-killing effect on worms occurs more quickly ( in less than one month ) than natural death ( ∼4–5 years [27] ) , and this more flexible scheme allows better prediction for the various options in timing and coverage among suggested control strategies [16] , [21] . In our model's simulations we assumed at-random treatment coverage , in the sense that a random subset of each population was treated each time [41] , [42] . In this sense , the ‘coverage’ of a control program means the fraction of all people treated in each treatment session , regardless of their past participation . Annual population growth for study villages was estimated to be 1 . 5% , based on the smoothed averages for household census surveys taken in 1984 , 1987 , 2000 , 2003 , and 2006 . Currently , the ‘standard-of-care’ for population-based drug treatment of Schistosoma haematobium and S . mansoni is based on the latest 2006 WHO guidelines [16] . Under these guidelines , populations are divided into 3 classes according to local prevalence of infection among children . These are: high-risk communities ( >50% children infected ) , moderate-risk communities ( 10–49% infected ) , and low-risk communities ( <10% prevalence ) , respectively ( Table 1 ) . For highly infected population , it is recommended that children between 5 and 15 year old , as well as adults , be treated annually; for moderately infected populations , children between 5 and 15 yr of age and high-risk adults are treated on an every 2 year schedule; for populations with low prevalence of infections , it is recommended that school children be treated twice , upon their entry into school and at school completion ( Table 2 ) . Since 2010 , new large-scale operational research trials have been initiated in seven schistosomiasis-endemic locations in sub-Saharan Africa , situated in Cote d'Ivoire , Niger ( 2 sites ) , Kenya ( 2 sites ) , Tanzania , and Mozambique [21] . These studies seek to identify more cost-effective approaches to regional and national schistosomiasis control , within the present-day framework of mass treatment campaigns for helminth control [26] , [43] . The randomized SCORE trials involve implementation in 25 villages per study arm , and will compare the relative costs and effectiveness of different drug delivery strategies for population-based control of either S . haematobium or S . mansoni infection . Two different kinds of SCORE studies will compare relative effectiveness of program implementation in communities having either low-moderate or higher prevalence of infection , defined as 10–24% prevalence among school-age children ( low ) vs . ≥25% ( high ) , respectively . For the selected SCORE villages , there was village-level randomized assignment to one of several different treatment coverage and delivery options: Some communities will receive community-wide treatment ( CWT ) , which will include treatment for all consenting adults as well as school age children; other communities will receive the more standard school-age targeted treatments ( SBT ) along the lines of current W . H . O . recommendations [16] . With each type of delivery , some communities will receive yearly therapy , some will transition from CWT to SBT mid-way through the project , while others will transition to every other year treatments or to ‘drug holidays’ . CWT will involve the most extensive treatment coverage , aiming to include as many people as possible , whereas SBT will target schools ( with non-attenders of school age also encouraged to participate ) , while drug holiday means there will be no treatment for the year . For communities with high infection prevalence ( >25% ) , there are 6 proposed strategies over the 4-year study period: For populations with lower infection prevalence ( 10 to 24% ) , no CWT is included in the SCORE trials , and treatment will be either school–based every year [S-S-S-S] , two years of SBT followed by two year holiday [S-S-H-H] , or SBT every two years [S-H-S-H] , with resulting community prevalence determined in the 5th year . An important consideration in choosing to implement a more intensive drug treatment programs will be the incremental cost-effectiveness relative to standard-of-care [44] . For this paper's analysis , we focused on two major costs: the cost for community screening ( in determining the prevalence of active Schistosoma infection in the population ) and the cost of drug treatment ( drug costs+delivery costs ) . In the SCORE program , screening and assignment is performed only at the beginning of the 4 year treatment regimen , so that screening costs are the same in each arm . However , for comparison among current WHO-recommended strategies , we have also modeled the possible impact of rescreening and reassigning treatment regimens on a yearly basis , and for the current WHO approaches ( not actually researched by SCORE ) we estimated the potential cost-savings and change in effectiveness using repeated rapid screening among school age children and reassignment of treatment strategies based on those annual results [45]–[47] . [Although it requires extra expense , screening before treatment has the potential to reduce the drug cost by adapting to a less expensive strategy when prevalence of infection becomes less intense . ] We took the cost of initial community sensitization and screening to be approximately one U . S . dollar ( USD ) per person , based on in-country costs documented by the Partnership for Child Development and other national control programs [48] , The cost of annual drug dose was estimated at 0 . 25 USD per child and 0 . 50 USD per adult . Rapid program screening is typically carried out by sampling 50–100 persons for each village [45] , so 100 USD was the maximum estimated screening cost per each village , no matter how large the population . The calculated cost for drug depended on the number of people treated , so information on total population , coverage for each treatment strategy , and number of villages treated was tailored to the intended coverage of each specific strategy . Taking the end of 2009 as the model's time baseline , we then projected the following metrics of each drug treatment strategy: i ) the number of years needed to bring down the every village prevalence to a low-risk level ( <10% ) and ii ) the number of years a village was likely to remain at this safe level following the termination of treatment .
One of the limitations of treatment delivered in population-based deworming campaigns is the risk of reinfection , sometimes referred to as ‘reworming’ [49] . The complex nature of Schistosoma parasite transmission can leverage parasite persistence in both snail and human hosts , while incomplete adherence with program-delivered treatments allows for persistent egg contamination by untreated residents [50] . Our previous analysis of a multi-year school-based drug treatment program for S . haematobium control indicated that the median time to reinfection can vary significantly depending on village of residence [20] , [24] . Over an 8 year observation period ( 1983–1991 ) , median time to reinfection could vary significantly between 2 to 8 years in adjoining villages [20] . The transmission potential within each community appeared fairly resilient to the ‘perturbation’ caused by targeted drug administration in schools , where ‘time since treatment’ and the specific study year did not significantly alter annual reinfection hazard [20] . Further , in 2000 , we observed that after an 8-year period during which control had lapsed , communities that had had the highest levels of infection in 1983 ( before any therapy had been given ) were the same ones that had the highest levels of infection after the 8-year pause ( Figure 2 ) ( Rank test rho = 0 . 927 , p<0 . 01 ) . While , in general , school-age prevalence in each community , when we revisited in 2000 , was about 32 percentage points lower than pre-control values , three of ten communities had reverted to the WHO ‘high prevalence’ category ( ≥50% , see Table 1 for definitions ) , and the remaining seven remained in the ‘moderate prevalence’ group ( 10–49% ) , while none were in the desired low prevalence category ( <10% ) associated with lowest morbidity risk after 8 years without control ( Figure 2 ) . Prior to predictive simulation of treatment program outcomes , we used the most recent 2009 infection prevalence data from the same region of Kenya to check the accuracy of the SWB model's predictions about infection prevalence following perturbation by community-based drug therapy . Treatment interventions were previously implemented in that region in 2000 , and on a limited basis in 2003 and 2006 [33] , [51]: in 2000 , 79% of those infected in the 10 village area were treated; in 2003 and 2006 , only the most heavily infected villages , villages 6 and 7 , were treated , with a coverage of 41–53% of all those infected in the area . Our new data from follow-up 2009 surveys [37] showed that the six-year post-treatment prevalence infection among children was again high: 61 . 7% for village 6 and 62% for village 7 . Figure 3 shows the comparison of observed 2009 prevalence ( black dots ) and model predicted 2009 prevalence point estimates ( gray dots ) among school age children for villages 6 and 7 . In sensitivity analysis , allowing model input parameters to vary randomly across a range of ±20% of their base-case values , observed 2009 prevalence values were well enclosed by the inter-quartile range of model outputs , indicating that its predictions were not extremely sensitive to changes in base case input parameters in this setting , and that our model had good predictive accuracy for this setting . We next used our calibrated SWB model to project the likely outcomes of implementing the current WHO treatment guidelines ( Tables 1 and 2 ) for high- , moderate- , and low-risk communities in the networked 10-village Schistosoma-endemic area ( Figure 1 ) . Using inputs based on recent community prevalence , we projected the number of years that would be required to bring all communities to a safer , low level prevalence category ( <10% ) , as a function of community uptake ( treatment adherence ) across the program ( Figure 4 ) . In doing so , we had each community continue on its original treatment strategy assignment until every community achieved the <10% prevalence level . In our simulations , coverage for high-risk adults among high-prevalence populations was assumed to be at least 60% . As our model was able to target different burden strata , we interpreted “high risk groups” as those adults in the 5 highest worm burden strata . Figure 4 shows our model's estimates of the number of years the WHO regimen would have to be implemented to bring all villages under control as a function of adherence among treated children . At high levels of uptake ( 80–90% adherence ) control was achieved in 4 years , whereas at lower levels of adherence ( 60% ) control took twice as long ( 8 years ) . We next asked whether repeated annual community screening and reassignment of treatment strategies ( again based on the WHO recommendations in Table 2 ) could offer a more cost-effective approach to infection control . Table 3 indicates the results of following a rescreen/reassignment strategy for up to 8 years in the same communities . Whereas continuation of the originally assigned treatment regimen was capable of lowering prevalence to <10% among children in each individual community , the strategy of reassigning treatment regimens based on yearly community screening was much less effective . As noted in Table 3 , the rescreen/reassign approach was unable to effectively suppress infection prevalence below 15% , even at the highest level of treatment uptake ( 90% adherence ) , likely related to the early switch to less intensive coverage as prevalence dropped . If adherence were less good ( i . e . , at a level of 60% ) , continuation of initial treatment strategies was less costly and more effective than the rescreen/reassign strategy . Notably , at that level of coverage , the final school age prevalence with the rescreening and reassignment approach was projected to remain at 35% at the end of 8 years ( i . e . , well shy of the 10% goal ) . If , however , adherence were >70% , rescreen/reassign became a less costly option but with , again , conspicuously subpar effects on suppression of infection prevalence ( Table 3 ) . In examining the success of WHO treatment regimens of differing frequencies and coverage , it was apparent that village level force of transmission ( i . e . , its reinfection potential ) played a large role in whether the program achieved or failed to achieve the goal of <10% prevalence . Local transmission factors also influenced whether a successfully treated community would stay in the ‘safe zone’ of <10% prevalence for any significant period of time after the mass treatment campaign was ended . Figure 5 demonstrates the projected prevalence outcomes for treatment of two modeled villages , one of high initial prevalence and continuing high transmission potential , and one of moderate prevalence and lower risk of reinfection . Both communities respond well to the first two rounds of drug therapy , quickly dropping to <10% prevalence . However , village A relapses to >25% prevalence after a 3 year hiatus in mass treatment , and requires ‘consolidation’ treatments to regain the <10% status . If consolidation is suspended after 4 years of annual treatments , infection prevalence rebounds to the moderate level ( >10% ) in two years . If consolidation runs for 8 years , community prevalence approaches it lowest levels , yet prevalence is expect to rebound over 10% in 3–4 years . By contrast , in the moderate prevalence village B having lower transmission potential , once local prevalence is well suppressed ( after the second round of mass CWT treatment ) then prevalence is expected to remain in the ‘safe’ zone ( <10% ) for 4–5 years . In this case , suppression of infection is used to indicate a persistent reduction of prevalence below 10% ( the WHO-recommended threshold for low-risk communities ) , but this does not imply an interruption of transmission . Figure 6 indicates the relative likelihood that a program based on WHO-recommended treatment schedules ( without reassignment ) would lead to successful long-term suppression of infection prevalence after mass treatment is suspended . The duration of continued suppression varied substantially between the modeled villages , and was much shorter in the communities with the highest levels of starting prevalence and greater links to high risk water sites ( 6–10 years suppression in lower prevalence communities vs . zero up to 3–4 years suppression in the highest risk communities ) . Results also varied the considerably according to the local adherence to treatment ( 60–90% , Figure 6 ) . Where adherence was quite high ( 90% ) suppression of infection with program intervention was much greater , suppressing local prevalences down to <5% ( Table 3 ) . Following this more aggressive level of suppression , rebound of infection prevalence ( to >10% ) took substantially longer in all villages ( Figure 6 ) . We next examined the potential of the six modified treatment regimens currently being researched by SCORE , in terms of their ability to achieve and maintain very low prevalence of Schistosoma infection . Final outcomes were quite different in the participating villages following 4 years of treatment in community-based or school-age targeted programs ( including strategies having community = >school-age crossover in coverage ( C-C-S-S ) , and those with gaps in treatments in different years ( ‘drug holidays’ ) , i . e . , C-C-H-H , S-S-H-H and S-H-S-H ) . Figure 7 shows the prevalence of each village after four years participation in each of the 6 candidate strategies . Strategies C-C-C-C , C-C-S-S , and S-S-S-S brought the prevalence of all villages well below the 10% prevalence objective after 4 years' treatment . By contrast , strategies C-C-H-H , S-S-H-H , and S-H-S-H achieved <10% in only half of the modeled villages , and these ‘successful’ villages were the villages with starting school-age prevalences of ≤36% . The projected prevalence outcomes at the end of C-C-C-C , C-C-S-S , or S-S-S-S treatment were similar , suggesting that addition of treatment for adult populations had relatively limited incremental benefits in terms of reducing local prevalence of infection . Strategies C-C-H-H , S-S-H-H , and S-H-S-H were not able to bring all villages to a safe level of infection , with post-program prevalences of 12–31% in 5 out of 10 targeted villages after the 4-year intervention period , so further treatment would be necessary in these villages ( as in discussed earlier for Figure 5 ) . Figure 8 indicates the projected number of ‘safe years’ ( local prevalence <10% ) after the 4 yearly rounds of successful area-wide C-C-C-C , C-C-S-S , or S-S-S-S intervention at an adherence level of 70% . At this level of adherence , each of these three strategies resulted in more post-program low prevalence years than projected for the six-year WHO 2006 strategy ( for the latter , when assigned according to initial high/moderate/low prevalence category and without rescreening/reassignment in subsequent years ) . This difference was noted in the six villages having the lowest pre-treatment prevalence . By contrast , among the highest prevalence ( 45–69% pre-treatment ) villages , neither the C-C-C-C , C-C-S-S , S-S-S-S , nor WHO strategies yielded a long-term , post-treatment suppression of infection prevalence . Reinfection was rapid , and there was no obvious advantage among the four . In terms of relative costs , among the three most effective strategies for infection suppression ( i . e . , C-C-C-C , C-C-S-S , or S-S-S-S ) , the S-S-S-S strategy was calculated to cost the least , based on differences in overall numbers treated ( S-S-S-S = $4185 , compared to $8455 and $6273 for C-C-C-C and C-C-S-S , respectively ) .
Results of our model's simulations indicate that in many moderate-to high prevalence areas , broad-based drug treatment programs will need to be sustained for four or more years to effect optimal suppression of local prevalence to <10% . Within the sub-district scale of our simulation , there were important differences in how the individual ( albeit networked ) communities responded to the treatment strategies that were implemented , and how long the impact of prevalence reduction could be maintained after suspension of treatment intervention . Programs with gaps in annual delivery ( ‘drug holidays’ ) were effective only in the lower prevalence villages , and did not fare as well as annual treatment programs ( C-C-C-C , C-C-S-S , or S-S-S-S ) in terms of long-term post-treatment suppression of infection prevalence . For neglected tropical diseases ‘deworming’ campaigns , there has been a recognized need for optimal community uptake , sustainability , and continued political will to achieve the objectives of the program [52] . The findings of our calibrated simulation support this thinking . In particular , our analysis suggests that schistosomiasis control programs should anticipate the need for multi-year and even multi-decade programs in some villages; initially , to achieve maximal suppression of local prevalence ( particularly in high-risk , high initial prevalence areas ) , and later , to maintain very low prevalence and maximal reduction in Schistosoma-associated disease . Intensive intervention in the initial years may result in prolonged suppression of infection in communities having low risk for reinfection , during which treatment ( but , importantly , not surveillance ) could be suspended . Program resources could then be re-purposed , and turned toward identification and retreatment of the highest risk villages . For now , these high risk villages are readily identified by their rapid re-emergence of infection prevalence ( i . e . , in 1–4 years ) . These same communities could also become the focus of non-drug interventions aiming at reduction of local transmission potential—in effect , re-forming high-risk villages into villages that have much lower risk of transmission and reinfection . Such efforts could include habitat modification at transmission sites , provision of safe alternatives for water use and recreation , and behavior change interventions focused on limiting water contamination and high-risk exposures . Based on our model simulation , we found that a policy of annual rescreening of villages ( with possible treatment reassignment according to current WHO 2006 recommendations [16] ) before each yearly treatment can appear less costly if treatment adherence is initially high ( >70% ) , but the associated results in terms of infection suppression are much less good ( Table 3 ) . This phenomenon is apparently related to a premature transition to alternate year therapy in truly high risk communities as they lower their prevalence into the ‘moderate’ category after the first round of treatment . For such high-risk communities , repeated annual treatment ( for at least 4 years ) appears to be necessary to reach fully effective suppression of infection prevalence . There are clearly limitations to our analysis . Although our projections are based on model simulations that use an advanced stratified model that appears well calibrated for a specific region of southeastern Kenya , its projections may not be fully generalizable to other endemic areas . However , we believe that the modeled mosaic of high- and moderate- prevalence villages within a sub-district area is typical of many territories found within Schistosoma-endemic areas , and that the results of our simulation will prove valid for many other locations where schistosomiasis control proves to be challenging . Wherever school age children have the most important role in disease transmission in a given area , it is likely that the predictions for other geographic locations would be qualitatively similar to ours . Among the modeling assumptions that might influence the accuracy of our predictions are the following: i ) we assume that treatment uptake is uniform across the program area and consistent from year to year . In fact , regular refusal or inability to participate , or progressive increases in non-adherence after initial years of treatment , could allow a core of untreated individuals to perpetuate transmission in a given sub-location [50] , [53] , [54]; ii ) it is possible that repeated annual treatments may provide periodic boost to anti-schistosomal immunity and relative resistance to reinfection [55] . This feature , which could gradually reduce individual risk of reinfection , was not included in our study; iii ) likewise , pond snail abundance was assumed to be stable each year over the period of the simulations . In drier landscapes where Schistosoma transmission is a rare event that is associated only with episodic flooding events , program response to therapy in high prevalence villages could be expected to be dramatically better than in our simulation , unless or until a flooding/transmission event again occurs . Of note , we have not modeled any effects of adjuvant snail control or other interventions aimed at reducing environmental transmission ( this will be the focus of a forthcoming study ) . The focus of our present analysis was only the outcomes of different possible drug-based treatment interventions . Given these caveats and limitations , we draw the following conclusions based on our modeling analysis: While these projected outcomes may not be fully realized in the seven ongoing SCORE operational trials , we feel that , for now , they offer useful , evidence-based , estimates of program outcomes where anti-schistosomal control programs are now being implemented . In terms of policy discussions and program design , the simulation results raise several new topics for consideration–There is clearly a need for ready identification of villages at high risk for reinfection . For now , annual rescreening of school age prevalence provides a basic marker of risk , but identification of other ( proxy ) features of high-risk villages could aid significantly in year-to-year planning for program deployment . In addition , beyond continuing surveillance , programs will need to decide how to manage their low-risk villages that no longer require therapy , and decide how best to bring them to the very desirable goal of complete transmission interruption .
|
Debate persists about how best to prevent disease caused by Schistosoma parasites . Current guidelines focus on suppressing morbidity by limiting average intensity of infection during childhood . However , this may not be sufficient to cure infection or prevent reinfection , leaving risk for sub-clinical morbidities such as growth stunting and anemia . More intensive programs involving broader coverage and/or more frequent treatments could potentially cure most infections and even prevent their return . Because such programs' effectiveness is not currently known , we used computer simulation ( grounded by past treatment program results ) to project short- and long-term impact in communities where Schistosoma are common . We estimated that 4 annual treatments ( delivered community-wide or only to school-age children and high-risk adults ) could effectively reduce local prevalence below 10% . Programs with gap years were less effective , particularly in high-risk communities . If a program was successful , infection stayed <10% for 7–10 years in low risk communities . However , rapid resurgence ( within 1–5 years ) was projected for high risk villages . We conclude that , given the networked transmission of Schistosoma , annual treatment programs of sufficient duration can have secondary benefits , i . e . , long-term suppression in some areas . However , high risk areas will need continuing surveillance and frequent retreatment to truly minimize their risk for disease .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"public",
"health",
"and",
"epidemiology",
"science",
"policy",
"and",
"economics",
"applied",
"mathematics",
"parasitic",
"diseases",
"preventive",
"medicine",
"mathematics",
"cost-benefit",
"analysis",
"neglected",
"tropical",
"diseases",
"urologic",
"infections",
"infectious",
"diseases",
"disease",
"ecology",
"infectious",
"disease",
"modeling",
"public",
"health",
"schistosomiasis",
"science",
"policy"
] |
2012
|
Projecting the Long-Term Impact of School- or Community-Based Mass-Treatment Interventions for Control of Schistosoma Infection
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.